Update: We reached out to LG for a statement, and it has denied claims of auto McAfee installations, stating: "LG Electronics reiterates that McAfee is not installed automatically and is never installed without the user’s explicit consent. The LG Monitor App Installer is distributed through Microsoft’s official Windows distribution process, which included McAfee as an option.

"McAfee will be installed only if a user actively chooses to proceed with the installation and provides consent. Under no circumstances is McAfee installed automatically or without user authorization. The LG Monitor App Installer does not access, collect, or transmit any customer personal data."

Original story follows below...

• LG monitors are reportedly installing bloatware on Windows 11 PCs
• The LG Monitor App Installer carries a McAfee pop-up and is automatically installed when connecting an LG monitor to a PC
• There's a fix for this, but it's a rather fiddly affair

It's no secret that Windows 11 users have voiced frustrations regarding bloat on the operating system for years, and unfortunately, LG's monitors are reportedly making the situation worse.

As reported by Windows Latest, LG monitors are installing McAfee bloatware on Windows 11 PCs without consent, and there's evidence of this in the changelog of LG's monitor app on the Microsoft Store.

Even if you don't install the app yourself, Windows Update will do so, and then treat you to a McAfee pop-up. All you have to do is plug in the LG monitor to your PC, and the automatic installation of McAfee is also shown by checking the Windows Reliability Monitor, which documents a 'successful Windows update' for the LG Monitor App Installer package.

We've reached out to LG for a comment, but haven't received a response yet. In the meantime, though, there appears to be a fix that requires users to do the following: press the Windows key and R, and type 'gpedit.msc', then hit Enter.

From that point, navigate your way to Computer Configuration > Administrative Templates > System > Device Installation > "Prevent automatic download of applications associated with device metadata", then select the 'Enabled' box and apply.

In theory, this should stop any unwanted applications from being installed, specifically with devices like LG monitors that do so via device metadata.

And if you already installed the LG monitor app manually, you should remove it, because as noted, the description on the Microsoft Store suggests McAfee is a fundamental part of the monitor application.

Bloatware can be problematic for Windows 11 users because it hogs system resources, and LG is adding to that system drain, not to mention the annoyance of continually seeing that pop-up.

Alongside LG, Microsoft hasn't provided a statement on the matter yet. I'd expect Windows 11 users will want an answer of some kind, especially because of the rather fiddly steps involved to prevent these automatic installations — and frankly, this shouldn't require extra effort from anyone.

Filmora is one of the most well-known video editing software in the industry. The company has also done well to embrace AI, adding a whole new range of advanced AI functions that have only made the platform better and more capable over time.

In this review, we've put Filmora to the test, trying out its top features hands-on to see if they live up to expectations. We also dive deeper into its other offerings, value for money, and in-use experience, as well as compare it to the best AI video editors to help you find out if it's the right choice for you.

Filmora: Plans and pricing

Much like other AI editing and content generation tools, Filmora also offers a free-forever plan, allowing you to edit videos along with access to a limited amount of creative assets such as effects, filters, transitions, music, and sound effects.

However, unlike Kapwing, Adobe Firefly, or Fliki, Filmora's free plan does not include any AI credits. This means you won't be able to generate AI videos or images for free with Filmora, which is a bit of a letdown considering the competitive landscape.

Filmora's paid plans start at $4.17 per month. The Basic plan is not much different from the free plan, except that you get 500 uses of its proprietary AI Mate, along with 1GB of cloud storage. Besides this, everything else remains pretty similar to the free plan.

You also have the option to add the Filmora Creative Assets add-on to the Basic plan for $19.99 per month. However, you do get a 14-day trial of the Creative Assets package.

There's also an option to purchase the Basic plan with a one-time perpetual license for $79.99. However, you will not receive future updates and will need to pay an additional upgrade fee to access newer versions.

You can check out Filmora by clicking here.

Next up is the Advanced plan at $5 per month, which comes with 1,000 AI credits. This is the first plan where you can create AI videos and short clips, and use AI tools like Text to Object, Text to Speech, AI Music, AI Images, AI Copywriting, and more. You also get 10GB of cloud storage, along with limited access to Creative Assets.

Compared to the likes of Veed, which costs $12 per user per month, or Kapwing, which costs $16 per user per month, Filmora's Advanced plan is one of the most affordable options available for AI content generation. Just like the Basic plan, you can also add the Filmora Creative Assets pack for $19.99 per month.

However, the Premium plan, which costs $8.33 per month, already includes the Creative Assets package and is, in our opinion, one of the best value-for-money plans you can get. You receive 2,000 AI credits per month, along with 100GB of cloud storage. The Creative Assets package also includes over 12,000 effects and filters, more than 7,000 transitions, and access to 1.5 million stock media assets.

Also, all the prices mentioned above are for the Windows version, with Mac pricing being slightly higher. For instance, the Basic plan on Mac costs $5.83 per month, with a perpetual license priced at $99.99, whereas the Advanced plan costs $6.67 per month and the Premium plan costs $10 per month.

Even though they are a tad pricier than the Windows version, they're still pretty affordable compared to other similar video editing and AI generation tools.

Filmora: Features

The best thing about Filmora is that it keeps adding new features. With the release of version 15, the platform has added a new repertoire of both AI and video editing features. For instance, there's the new AI Video Extender, which helps you expand your footage forward or backward with the help of a prompt.

This may come in handy when your footage is too short and you need a bit more footage for smooth intros and outros. Using it is also pretty simple. You just need to drag its Smart Pen onto the editing dashboard and add a prompt describing how the video should continue.

In the same vein, Filmora has also introduced Prompt Video Editing, which allows you to add, replace, or remove objects in your videos using simple text prompts instead of having to do it manually on the editing dashboard.

Another impressive new feature is Animated Charts, which lets you import CSV or Excel files and generate chart and graph animations to build dynamic line, area, and pie charts. This comes in handy for commercial teams that want to prepare engaging presentations with animated visualizations.

The platform has also recently introduced advanced pixel tracking and motion compensation algorithms for its built-in video stabilization. This helps repair shaky video footage, making it more stable.

Similarly, there's also a Lens Correction option, which helps you remove optical distortions caused by camera lenses. This includes barrel or fisheye distortion, pincushion distortion, vignetting, and perspective or keystone distortion.

This comes in handy while editing wide-angle or action camera footage, where the fisheye bulge is distracting, or architecture and interior shots, where you want walls and buildings to look straight instead of curved.

Furthermore, Filmora now has advanced planar tracking, which can follow an entire flat surface across a moving video and lets you add images, videos, text, logos, mosaics, or anything else you want to that flat surface, much like inserting an image into a video. For instance, you can attach a logo to a moving truck, billboard, or storefront sign.

Filmora also has one of the largest libraries of creative assets we have seen across similar applications. Although it's a paid add-on, the Creative Assets package unlocks a large pool of premium effects and media.

This includes stock media and original video effects, over 100,000 royalty-free audio tracks, professionally designed title templates, a wide range of basic and professional transitions, aesthetic and cinematic effects and filters, stickers, more than 10,000 ready-made templates, and over 500 AI-powered effects.

Here are some other features you get with Filmora:

• AI text-to-video, audio-to-video, and image-to-video.
• AI Storyboard Generator, allowing you to generate ready-to-use storyboards with editable characters, scenes, and styles.
• An all-new Pen Tool, which lets you draw custom vector paths and shapes directly onto the video frame.
• AI Color Palette for cinematic color editing.
• Text animation with immersive and dynamic effects.
• An automated AI-powered subtitle generator, along with more than 100 animated subtitle templates.
• AI voice cloning in 16 languages.

Filmora: Interface and in use

The Filmora dashboard can seem a bit overwhelming at first, especially if you are a beginner. The editing panel is divided into three panes: one for adding edits and elements such as audio, titles, transitions, and effects; one for viewing your image or video as you edit it; and a horizontal editing bar at the bottom, which lets you drag and drop various editing elements into your project.

At the top of your dashboard, you'll see various options such as Media, Filters, Transitions, Effects, and Templates. Simply click on any of them to open all the available options and drag and drop the ones you want onto the horizontal bar at the bottom.

Once you click on the panel where your video is being previewed, a host of other options opens on the right-hand toolbar, where you can adjust various video settings such as width, height, rotation, and flip.

As you scroll down, you'll find various AI functions such as motion tracking, planar tracking, motion blur, and stabilization. What we liked most is that you simply have to toggle the buttons to activate the features and drag elements onto your video to apply the edits. There are no complex steps involved.

In addition to video, you'll also find other advanced editing options such as speed, animation, color grading, and AI palettes on the same right-hand panel.

To be fair, it can take a bit of time for new users to get the hang of the platform. That said, Filmora is one of the most rewarding platforms once you get past that learning curve thanks to the sheer number of editing options available.

Filmora: How we tested

Since Filmora doesn't allow you to create AI content on its free plan, we couldn't test the platform's AI capabilities. However, we did put Filmora's editing functions through a thorough test.

We first downloaded the application on our Windows 11 PC, which took around a minute or two. Once set up, we imported stock media to edit. We clicked on Effects and added an RGB Stroke effect to our video of a man skateboarding. Filmora was able to identify and track the person in the video and apply the RGB Stroke effect pretty accurately.

Adding the effect was also pretty easy. We only had to select and drag the effect onto the horizontal bar. We then added a dissolving transition to the beginning of our video. Again, a simple drag-and-drop did the trick.

We then played around with various video and audio editing settings, and all of them were applied to our video almost instantaneously. There was no lag or buffering while editing even the most complex or longest videos.

Filmora: Alternatives

Filmora focuses heavily on video editing features, offering several advanced options for professional editors. But there are a few alternatives you can consider.

If you're looking for free AI video and image generation, you can consider tools like Kapwing or Fliki. Both allow you to create AI videos and images for free.

Kapwing does a good job when it comes to generating AI characters, from simple everyday characters like a baby or a cheerleader to fantasy characters like a dragon or a cyborg. There are plenty of options.

Fliki, meanwhile, gives you many pre-video generation options to choose from. For instance, you can customize your video format, template, AI avatar, voiceover, and captions before generating a video, which saves a lot of editing work afterward.

However, Kapwing's highest-tier plan can be pretty expensive, costing $50 per user per month, whereas Fliki can go up to $44 per month. Even Filmora's most advanced plan costs only $8.33 per month, making it far more affordable. Plus, both platforms lack advanced editing options such as planar tracking and video stabilization.

Filmora: Final verdict

Filmora is one of the most feature-packed AI content generation and editing tools on the market. Apart from basic features like image-to-video, script-to-video, and audio generation, you get advanced editing options such as planar tracking, storyboard generators, video stabilization, AI video and audio extenders, dual-monitor editing, animated charts, and lens correction.

Additionally, Filmora houses one of the most extensive libraries of stock assets, including licensed music, effects, transitions, stickers, and more than 10,000 templates.

That said, its interface is not among the simplest we have seen and requires a bit of a learning curve. Also, its free plan does not allow you to create any AI content, which was a bit of a disappointment. If this is a deal-breaker for you, you can consider other options like Adobe Firefly or Kapwing.

We've tested a range of video makers and editors, including the best video editing software, the best video editing apps and the best free video editing software.

Every enterprise I talk to right now has the same complaint dressed up in different language.

Their AI bills are climbing faster than anyone budgeted. Their model invoices look reasonable when viewed on their own.

But somewhere between boardroom approvals and the monthly cloud statements, money is disappearing in ways that nobody can fully explain.

Here’s the central issue that people struggle to understand: the most expensive part of AI isn't always the model itself. It's the infrastructure, orchestration, retries, idle GPUs, oversized context windows, and inefficient routing decisions that sit between a user's prompt and the final response.

This is something I call the hidden tax on AI adoption, and it's growing faster than most organizations realize.

Three numbers that should change how you think

Recently, at FinOps X in San Diego, a Goldman Sachs projection appeared on the main stage. Current enterprise token consumption globally sits at around six quadrillion tokens. The three-year projection: 120 quadrillion. That is not a rounding error… it is a 20x expansion, and it is arriving faster than the governance frameworks to manage it.

The same conference saw our launch of the Tokenomics Foundation (I’m fortunate to be a governing board member). It’s a vendor-neutral body inside the Linux Foundation dedicated specifically to the economics of AI token consumption.

The FinOps community, practitioners who have spent the better part of a decade building discipline around cloud computing spend, recognized that tokens represent a fundamentally different problem. Not a harder version of cloud cost optimization. A different one.

Here is why. When your organization runs a cloud workload, the cost is relatively legible. You provision compute, it runs, and you receive a bill. The relationship between action and expense is traceable. With AI tokens, that relationship fractures across three layers, and most organizations have visibility into only one.

Production → consumption → value: the three layers most teams ignore

The first layer is production. Before any AI model responds to any prompt, your infrastructure has to manufacture the tokens. GPU clusters, inference nodes, autoscaling policies, Kubernetes configurations: these are your token factories. Their efficiency, or lack of it, determines the base cost of everything that follows. A GPU node at 30% utilization is an expensive factory running at a third of capacity.

The second layer is consumption. This is where counterintuitive economics live, and it is what I spend most of my time thinking about. Two organizations can send identical prompts to different coding agents and arrive at radically different costs depending on how they manage context, caching, retries, routing, and the infrastructure supporting inference.

Prompt length, context window usage, caching strategy, and model routing decisions all compound. The common assumption that routing a task to a cheaper model always saves token cost, turns out to be wrong often enough to matter. A routing decision that invalidates a warm cache can make a "cheaper" model call more expensive than the frontier option it was meant to replace. These are second-order effects. They do not appear in standard dashboards but they appear in your monthly bill.

The third layer is value. This is the one that FinOps teams are comfortable with, and the one that matters least until you have the first two under control. Mapping token spend to business outcomes is a legitimate and important discipline. But you cannot govern at the value layer without instrumentation at the production and consumption layers. You are doing math with incomplete inputs.

Why 85% of your AI spend is probably misallocated

Here is a pattern I see consistently. Organizations treat frontier AI models, the most capable, most expensive models available, as their default infrastructure. Every task goes to the same model. Every prompt is constructed the same way. There is no routing logic, no tiering, no architectural distinction between work that genuinely requires the full capability of a frontier model and work that does not.

Based on my observations across organizations deploying AI at scale, roughly 15% of software development tasks actually require frontier model capabilities. The remaining 85% of routine coding, summarization, classification, and retrieval work can be handled by smaller, faster, and less expensive models, if you have the infrastructure to make those decisions intelligently and automatically.

The unlock is not picking better models manually. Manual model selection does not scale and degrades the developer experience by introducing friction at the moment when a developer needs to move fast. The unlock is building infrastructure that makes routing decisions for you: one that understands the task, routes it to the appropriate model tier, evaluates output quality, and escalates if needed. You specify the outcome you need. The system handles the economics of achieving it.

This is the direction the industry is moving, and the organizations that build this capability first will have a structural cost advantage that compounds over time.

The invoice arrives last. And you realize something has gone horribly wrong

There is a phrase I have started using with customers that captures the core problem: the invoice arrives last.

By the time you see the model provider bill, the cost decisions were made weeks ago in infrastructure configurations, autoscaling policies, and prompt architectures that nobody has reviewed since the initial deployment.

The retry logic runs silently when an upstream service slows down. The GPU nodes were reserved for peak traffic that never came. The agentic workflow, where a single user request fans out into dozens of model calls beneath it, each billed separately, none visible in the tool that generated the original request.

These costs do not live in the model invoice. They live in the infrastructure layer, in the consumption layer, and in the gap between how teams think their AI systems work and how they actually behave in production. You cannot govern what you cannot see. And right now, most teams are looking at one layer of a three-layer problem.

When AI goes from copilot to coworker, the stakes multiply

There is a shift underway that makes all of this more urgent. The AI deployments most enterprises built over the last two years were assistants, tools that accelerated individual work by handling the first draft, the next suggestion, and the boilerplate. A human remained in the loop at every consequential step. The economics were bound by how many people were using the tool and how often.

Autonomous agents change the economic profile entirely. When an AI system can receive a goal, build a plan, execute multi-step work, evaluate its own outputs, and iterate to completion without human intervention at each stage, you are no longer running an assistant. You are running something closer to a coworker, one that operates continuously, scales horizontally, and generates token consumption at rates that individual user interactions never approached.

The transition from copilot to coworker has already happened. And the governance implications are significantly more serious. A copilot with poor token economics costs you some efficiency. An autonomous agent with poor token economics runs that inefficiency at scale, continuously, without generating the natural friction that would cause a human user to pause or change approach. Infrastructure discipline and token optimization need to be in place before autonomous workloads scale rather than retrofitted afterward when the bill arrives.

The mandate for infrastructure teams

The right answer to this problem is not more dashboards. More visibility into a system you cannot control is just a more detailed invoice; it arrives with the same lag, and it changes nothing about the decisions that were already made upstream.

What infrastructure teams actually need is control that operates at the layer where costs are determined, not where they are reported. That means autonomous management of GPU and inference workloads, continuously rightsizing to match actual demand rather than peak assumptions, absorbing the bursty consumption patterns that agentic jobs produce, and moving compute capacity across providers when one environment becomes the bottleneck.

This is especially urgent now, because the transition from copilot to coworker does not give you a grace period to retrofit discipline. Autonomous agents do not pause. They do not get frustrated and choose a different approach. They run the inefficiency you built into them at scale, continuously, until something external stops them.

So the next phase of enterprise AI is defined by who can deploy models efficiently. As AI systems become more autonomous and token consumption accelerates, competitive advantage comes from understanding the full economics of AI, and not just the price of a model call.

Because by the time the invoice arrives, the decisions that shaped it have already been made.

Use the best business cloud storage to manage your data.

This article was produced as part of TechRadar Pro Perspectives, our channel to feature the best and brightest minds in the technology industry today.

The views expressed here are those of the author and are not necessarily those of TechRadarPro or Future plc. If you are interested in contributing find out more here: https://www.techradar.com/pro/perspectives-how-to-submit

Adobe is a popular brand when it comes to all things creative, including AI content generation and editing. With its newly improved Firefly tool, Adobe has taken AI content generation a step further with realistic, real-time outputs.

In addition to being highly accurate, Firefly is also one of the most affordable tools on the market today. But is it the best AI video editor for your needs? Read on to find out. In this article, we've put Firefly to the test, examining its features, pricing, ease of use, and overall value for money.

Adobe Firefly: Plans and pricing

Adobe Firefly is available online by clicking here, as well as part of the Creative Cloud suite of apps.

It offers one of the most generous free video editing software we've seen. Instead of capping your monthly or one-time usage, Adobe allows you to generate a limited number of images and videos each day. This means you can continue using Firefly for a longer period, provided you stay within the daily limits.

What we like about the free plan is that Firefly does not lock its image or video editing features behind paywalls. Even on the free plan, you get access to features such as Generative Fill, Background Removal, AI Markup, video upscaling, text-based editing, as well as audio features such as generating soundtracks, speech, text-to-avatar, and more. Unlike other platforms, Firefly's free plan can actually come in handy for limited individual use.

Its paid plans are also among the most inexpensive we've seen in the category, with the Standard plan starting at $8.49 per month when billed annually, which comes with 2,000 credits.

With this plan, you get access to all the AI models Firefly has to offer, with hard upper limits on the number of images and videos you can generate with each model. Videos, however, are capped at a maximum duration of five seconds, which might not be enough for users looking to generate long-form content. Besides this, you get access to unlimited Firefly Boards, while the remaining features stay the same as those in the free plan.

Next is the Pro plan at $16.99 per month (4,000 credits), billed annually, where you get unlimited access to several AI models such as Gemini 2.5 Flash, FLUX.1 Kontext, and others. The maximum number of images and videos you can generate is also higher than in the Standard plan.

Then there's the Pro Plus plan, priced at $29.38 per month, where you get up to 10,000 credits and higher limits on image and video generation. In this plan, a wider range of models comes with unlimited access, including Gemini 3.0, ChatGPT Image, Runway Gen-4.5, and Gemini 3 Nano Banana Pro 2K.

Finally, there's the Premium plan at $118.93 per month, offering 50,000 credits and unlimited access to almost all AI models except a few. This plan is ideal for large content creation and creative teams.

Adobe Firefly: Features

Right off the bat, what impressed us most about Firefly is its brainstorming and early-stage concepting feature called Firefly Boards. It is essentially an infinite-canvas mood board and ideation workspace that lets you upload your own images, sketches, or stock images, or generate new ones.

You can select multiple assets and combine elements while brainstorming ideas with your creative team. This is similar to Midjourney- or Discord-style iteration, giving remote creative teams a space to flesh out new ideas.

Moreover, Firefly offers access to more than 10 popular video, audio, and image models, each with its own limits on the amount of content you can generate. Some notable features are Generative Expand and Generative Fill, which let you expand the canvas of an image or video after it has been generated using Adobe's AI engine. Similarly, there's also Generative Remove, which helps you remove elements from a piece of content.

Firefly also throws in a range of audio features, such as the ability to generate speech, translate videos from one language to another, or add text-to-sound effects to your content. However, we were a bit disappointed with its video editing features, such as trimming, arranging, and refining content, all of which are still in beta. Its AI assistant is also at a very early stage and currently supports only one conversation at a time.

You also have the option to upload an image to provide Firefly with a style or structure reference and create a consistent look across a batch of images. Unlike many other AI tools, Firefly allows you to use your own images to train its AI models and create custom models. This is still an early-stage feature, and we expect it to improve over the next year or so.

Firefly has also expanded its multilingual reach and now supports prompts in more than 100 languages, along with translations in more than 20 languages. There's also an option for bulk actions such as background removal, color grading, and cropping.

Its mobile apps have also improved a lot over the past year and now offer native iOS and Android apps with an experience that's very close to the web interface. And of course, Firefly is part of Adobe's broader AI-powered content generation and editing ecosystem, integrating directly with apps like Photoshop and Illustrator.

Adobe Firefly: Interface and in use

Adobe Firefly is by far one of the most improved image and video editing software we have seen in a long time. One of the biggest criticisms of Firefly was that it did not offer any option to edit the images it generated. However, all that has changed with its revamped user interface.

Not only did it generate a pretty high-quality image for us, but it also provided several editing options. We especially liked its Tune feature, which is currently in beta and allows you to change the look of an image by selecting from a list of preconfigured options.

There is also a prompt option, which lets you keep fine-tuning your generated image with follow-up prompts.

However, if that is too much work, there are several fine-grained controls available. For instance, there is a Fill feature that lets you fill any blank spaces in the image with an element of your choice. Similarly, there's the Remove feature, which lets you remove any element you do not want in the final result.

What we liked the most is its Select feature, where you can select a particular element in an image or video and then type a prompt to edit only that selected portion. This helps avoid unwanted changes in other parts of the image or video and ensures that the edit is applied only to the section you selected.

The generation interface itself is pretty simple. You'll see a prompt box at the bottom of your screen, along with a panel on the left-hand side where you can control the generation settings.

For instance, you can select the model you want to use, the resolution, the aspect ratio (16:9 or 9:16), and the duration of the video. There is also an advanced setting where you can enter a random seed value to experiment with the AI engine settings.

What we found particularly eye-catching is the prompt enhancement feature, which lets you improve your prompts before generating an image or video. This helps you create more detailed prompts, which in turn result in more accurate outputs.

Adobe Firefly: How we tested

We tried the free version of Adobe Firefly over several days and were quite impressed with the results. We first tried to create an AI image with the following prompt:

"A tired line cook in a stained white apron, frowning with exhaustion, wiping sweat off his forehead while chopping six carrots on a wooden cutting board. Behind him, a dim, cluttered restaurant kitchen with steam rising from a pot. Photorealistic, dramatic low lighting, shot on 35mm film."

As soon as we hit Generate, Firefly got to work and produced a pretty impressive image that closely matched our prompt.

However, it's worth noting that Firefly is a tad slower compared to the likes of Fliki or Kapwing. That said, the accuracy of the results is worth the wait.

Once the image was generated, we clicked on it to open the editing panel and played around with the settings. For instance, we selected the burning steel pot behind the cook and prompted Firefly to change it to a wooden pot instead.

Rather than simply making the change, Adobe gave us an Edit Strength slider, allowing us to control how strongly the wooden effect was applied to the pot.

We then tried the Remove feature by selecting a bunch of carrots and asking Firefly to remove them. Adobe once again impressed us, as it accurately removed the carrots without distorting the chopping board around them.

Lastly, we used the Upscale feature to improve the quality of the image. This took the longest, more time than it took to generate the image in the first place. However, we were once again very satisfied with the final result.

We also generated an AI video using the following prompt:

"A woman in a red coat walks across a rain-soaked city street at night, neon signs reflecting in the puddles, camera slowly tracking alongside her."

The results were again highly accurate, the colors were vibrant, and we were impressed by how closely Firefly followed the prompt without significantly diverging from the intended output.

Adobe Firefly: Alternatives

Although Adobe Firefly is a much-improved product, it does have a few shortcomings, which is why you may want to consider some alternatives. For instance, Adobe isn't the best choice if you want to generate AI avatars. Although it offers the feature in a limited capacity, the results are not what you'd expect from a dedicated AI avatar generation tool.

In that case, you can try Kapwing, which offers one of the most extensive collections of AI avatars. You can choose from a wide range of characters, including traditional human-inspired faces and more outlandish options such as aliens, Cupid, demons, cyborgs, Dracula, and more.

Kapwing produces some of the best photorealistic AI avatars we have seen in the industry. It is also one of the few tools that can render videos in 4K quality. That said, it is a bit on the expensive side, with plans starting at $16 per month.

If you need a dedicated AI assistant to help with content generation and editing, you can try Fliki, which offers an AI Copilot that understands natural language inputs and helps you achieve your desired results more effortlessly. It also includes licensed YouTube music and advanced video generation options such as stop-motion videos.

Fliki also offers a wider range of customization options compared to Adobe Firefly, such as choosing the format and template, customizing the type of character, selecting the AI model you want to use, and adding voiceovers, captions, and AI avatars.

Adobe Firefly: Final verdict

Adobe Firefly is one of the best-known names in the content creation industry, and its recent improvements put it right at the top of our list of the best AI video editors. Firefly offers more than 10 AI models for image, video, and audio generation, helping you create photorealistic images, videos, and audio with simple prompts.

We were particularly impressed by Firefly Boards, which helps you brainstorm ideas on an endless online canvas in real time while collaborating with your content team. Firefly has also improved its editing interface and tools and now supports follow-up prompt editing, along with features like Generative Fill and Generative Remove.

That said, it has a few drawbacks, such as the lack of mature video editing tools, which are still in beta, and rather limited AI avatar generation capabilities. That said, if you already use Adobe tools like Illustrator or Photoshop, adding Firefly to your repertoire is a no-brainer.

We've tested a range of video makers and editors, including the best video editing software, the best video editing apps and the best free video editing software.

While the tech world is gearing up for the debut of Samsung’s new foldables at Unpacked next week — with the company teasing some major changes to its lineup — here in the UK we’re finally able to get our hands on the Samsung Galaxy XR headset.

It's now available outside the US and Korea for the first time since its launch in October of last year, and I’ve been taking the headset for a spin after Samsung sent me a review a few weeks ago, and seeing how it compares to my trusty Meta Quest 3.

Now, I could just rehash a lot of what our Editor at Large, Lance Ulanoff, said in his Samsung Galaxy XR hands-on — praising the OLED screen, describing the comfort of the headset over several sessions, and listing the apps I love in XR that elevate the Android experience beyond what's possible on a phone or laptop. But I won't do that here.

Instead I'm putting my body (well, head and neck comfort) on the line to bring you the return of my favorite VR headset testing format: can I spend a whole week wearing the Samsung Galaxy XR?

The short answer? Yes, but it could be comfier. Read on for more about my experience.

Into the Android XR-verse

Just to be clear, I didn't wear the Galaxy XR headset 24/7 — I haven't been walking around my local supermarket immersed in XR, or sleeping in my Samsung headset .

I spent my whole working day in the headset (taking it off only for my lunch breaks, and to use the bathroom), and also wore it for a couple of hours each evening, to unwind with a movie or a game.

I also made sure to spend at least a day relying on only the hardware in the Samsung Galaxy XR’s box — i.e., the headset.

Software-wise, I used everything I’d normally rely on my PC or phone to work and stay entertained, and also had access to the Explorer Pack — a suite of software and subscriptions worth £665 that you'll get for free with a purchase of the Galaxy XR until December 16.

A quick aside on this pack: with a year's access to premium Gemini models (and 5TB of digital Google Cloud storage), plus a year of YouTube Premium, a Calm subscription, and plenty more, it's a really solid addition to the XR experience — I'll be writing more about my experience with its services soon.

Anyway, back to the headset itself…

Things got off to a good start on day one. Thanks to the XR multitasking I was able to have a giant floating Slack window keeping me in touch with my TechRadar colleagues, and a second giant floating YouTube window featuring the latest episode of one of my many podcasts or music playlists to keep me engaged, flanking a Google doc page for whatever I was working on, which I wrote in using a floating keyboard that reacts to my finger movements.

I could open Google Chrome tabs too, and conveniently flick through and between windows using eye controls — the headset relies on eye-tracking to determine what buttons I want to press, which I indicate by simply pinching my fingers while my hand rests lazily in front of me.

I was even able to join meetings virtually, with digital avatars that were good, but not yet great at representing me or mimicking my real-world facial expressions. My Galaxy avatar looked a bit too cartoonish for work meetings, and my Google avatar looked permanently distressed (which I don’t think is my general state of being), and was also missing my glasses, which felt weird — I always wear them, so it felt like the avatar was missing some of my identity.

That aside, I quickly found that any job which relies on lots of basic operations (essentially anything that isn’t something like intense video editing, though there are Play Store apps for less sophisticated video editing needs) is a breeze on the Galaxy XR headset. However, as the novelty faded, I hit my first hurdle: speed.

Gotta type fast!

Eye-tracking controls are neat, and generally XR controls are incredibly intuitive, but the keyboard is not as responsive as even the most basic physical keyboard. This slowed my typing pace to a crawl, and with deadlines looming over me I couldn’t spend my days steadily working through sentences at a snail pace.

The ideal alternative would be a Bluetooth keyboard that pairs with your headset, and most standard models will get the job done after you’ve connected one. I gave one a quick try, and it was much more effective than the virtual keyboard for actually working productively in XR — the brief pause I had to contend with between virtual key presses was eliminated.

My predicament gave me an excuse, however, to see how the headset works when paired with a PC — giving me the massive virtual displays of XR combined with the convenience of a fully powered desktop.

The perfect setup would involve a Samsung Galaxy Book, which pairs more seamlessly with the headset, like the Vision Pro and a MacBook pair for Apple users. However I don’t have a Galaxy Book, so I instead had to turn to Virtual Desktop.

It’s an app I’ve used previously to pair Meta headsets with my PC, and it’s very effective. The connection, even in a room a few meters from where my router lives, is seamless, with next to no delay between what I do on my PC and see on my headset’s view. Plus, with the ability to drag displays around and resize them, I could create the ultimate multi-monitor setup.

The downside is that you can’t access the full multitasking capabilities of the Galaxy XR headset within the Virtual Desktop app. Other apps seem not to suffer from these issues; however the trouble is that several, such as Google’s PC Connect, weren’t available in the UK when I was conducting my tests, and it seems that every app has a downside (for example, PC Connect only lets you stream a single monitor, not multiple).

This compatibility friction with non-Samsung PCs would be the biggest change I’d like to see addressed from a productivity standpoint, as otherwise the Galaxy XR excels on the software side.

My other two big changes would be to the design: the addition of a silicone facial interface and a top strap.

I can’t feel my head when I’m with you

The UK, where I’m based, has been buffeted by heat waves in recent weeks, including while I was testing the Galaxy headset. In homes designed to retain heat, and which lack A/C, this means things can get pretty hot and sweaty, which isn’t ideal when a headset is strapped to your face — and it’s even less ideal when you can feel that sweat being soaked up by the fabric facial interface.

If I was in charge of making headsets I’d mandate that all facial interfaces are made of silicone, or at least that headsets come with a swappable silicone interface, or a silicone cover for the fabric interface, in the box. Unfortunately I can't force such a change on Samsung, but hopefully it’s a change it’ll make for the Galaxy XR 2; it’s more hygienic, and makes the headset easier to share, and I’ve found it doesn’t impact the comfort at all (if anything it makes me feel more at ease in the headset, as I know I’m not immersed in months-old sweat).

As for a top strap (one which runs over the top of my head rather than simply around it), I firmly believe it would help to make the headset comfier to wear for long stretches, as it would further spread the weight of the device across my head.

Day one wasn’t too bad, but by the midpoint of day two I was feeling a little discomfort. By day four I was ready to give up my challenge; and the end of day five couldn’t come fast enough.

Samsung gets so much else right in terms of the Galaxy XR's comfort — chiefly moving the weight of its bulky battery to an external puck, so that you’re only carrying 545g on your head — so it’s a shame that the headset is let down by something as simple as its headband.

For a few hours I found it was fine, but for my extreme testing sessions it didn’t hold up.

Yet again, then, I’ve found that I don’t really want to spend my whole day inside a VR headset, even though the Samsung Galaxy XR headset represents a serious improvement in terms of visual performance (courtesy of its superb OLED screen), the usefulness of its features, including eye-tracking (which I’ve previously found is quite gimmicky), and delivering solid multitasking performance (aided by the expansive app library of the Play Store) compared to the last time I tried a similar exercise with the Meta Quest Pro a few years ago.

Even if I won’t be spending all my time in the Galaxy XR, I plan to wear it plenty more, albeit just for more casual sessions, and I’ll definitely be taking it on a flight, as soon as I have the opportunity, to see how it fares — perhaps that’s where it’ll really shine.

• Windows 11's July update was blocked from certain Dell laptops due to nasty issues with unexpected shutdowns and poor performance
• Microsoft just issued an emergency patch to fix those problems, so the PCs in question can install the latest update
• However, that patch might end up installed on non-Dell PCs if the user has enabled the option to get the latest updates as soon as possible

Windows 11's latest (July) update was blocked from Dell PCs as it was buggy on those systems, but the good news is that it's now fixed — but there's an additional wrinkle to be aware of here for those of you who don't own a Dell machine.

Windows Latest reports that Microsoft just deployed an emergency update (KB5121767) which fixes the problems with certain Dell laptops with Intel CPUs, which include unexpected shutdowns and generally poor performance (including running sluggishly and overheating, as well as increased battery drain).

Those with an affected Dell laptop — and that apparently includes some XPS models, as well as Dell Pro and Precision (business) notebooks, and possibly others — should install the emergency patch. As Microsoft makes clear, you'll find it under optional updates, which is reached via 'Advanced options' in the Windows Update panel.

However, the wrinkle that I mentioned is that Windows Latest observes that it's possible some Windows 11 PCs that aren't Dell laptops may automatically grab and install this update — if they have the 'Get the latest updates as soon as they're available' toggle on.

Analysis: a notable side effect

This feature is one that people turn on in the hope that they'll get Windows 11 updates, and more to the point, new feature rollouts, more quickly (there can be quite a wait for certain features to arrive, such as the Start menu revamp from last year, which some folks still haven't received even now).

So, bear in mind that if you've enabled this option, you may get certain updates automatically — like this emergency patch — when you don't actually need them on your PC. It doesn't help that this particular update is confusingly named, being called '2026-07 Update (KB5121767) (26200.8894)', which could leave you wondering what on earth it is exactly (until you look up the codename KB5121767).

At any rate, if you're not happy with this situation, to turn off 'Get the latest updates as soon as they're available' you need to head to Windows Update (in Settings) and adjust the relevant slider (at the top of the 'More options' panel).

Why is this option configured in this way? It seems especially odd in this case seeing as Microsoft tells us: "This OOB [emergency] update is only recommended for devices affected by this issue." That said, seemingly it doesn't do any harm to a non-affected PC (whether a Dell model, or not) if it does end up with this emergency patch installed.

At any rate, if you do enable the option to get the latest updates, just don't forget that this can have side effects, such as installing optional updates that you don't actually want. It would be a better idea if, in these cases, Microsoft simply flagged these updates as available, rather than having them install automatically.

It's also worth noting that this isn't the first time Dell laptops have run into trouble with Windows 11 recently, as last month an update to the firm's SupportAssist app left some notebooks crashing every half an hour or so.

Kapwing is among the best AI video editors today. It allows you to create videos with simple text prompts and generate lively AI characters to build detailed storylines.

Trusted by brands like Google, Dyson, and Spotify and used by thousands of content creators across YouTube, TikTok, and digital marketing platforms, Kapwing has quickly grown to become one of the industry’s favorite video generation and editing software.

In this article, we put Kapwing to the test to see whether its features perform as advertised. Read on to learn about our real-world experience with Kapwing, its best features, and the alternatives worth considering if it doesn’t fit your needs.

Kapwing: Plans and pricing

You can get started with Kapwing through its free-forever plan, which allows up to 30 minutes of video exports per month in 720p resolution, with each video limited to a maximum length of one minute. There’s also a 2GB file upload limit, along with access to editing and AI-powered tools such as auto-subtitling, auto-translation, and text-to-speech, albeit with limited usage caps.

We also like that you get a fair share of collaboration features, such as project link sharing and commenting, allowing you to edit videos collaboratively with your team. However, the free plan is meant primarily for testing the platform rather than full-fledged use.

If you want to get the most out of Kapwing, you’ll have to upgrade to one of its paid plans. Its Pro plan starts at $16 per member per month and offers excellent value for money. The plan supports 4K video exports with unlimited monthly exports and a maximum video length of 120 minutes.

You can check it all out by clicking here.

The AI editing tools also come with higher usage limits. For example, you get up to 1,000 minutes of auto-subtitling and 500 minutes of auto-translation per month. Besides this, the plan adds several premium tools, including video and image background removal, video stabilization, and the ability to add your brand kit and custom fonts to your projects.

Next is the Business plan, designed for teams that need more advanced content creation workflows. Costing $50 per member per month, it includes up to 4,000 minutes of auto-subtitling, 2,000 minutes of auto-translation, 2,000 minutes of Clean Audio per month, and 130 minutes of Lip Sync. The remaining features are largely the same as those in the Pro plan.

Finally, there’s an Enterprise plan for large organizations that need priority support and custom billing. However, you’ll need to contact the Kapwing team directly to obtain a customized quote.

Kapwing: Features

Kapwing packs a good mix of both AI-powered and traditional video editing features. Let’s start with its AI capabilities. Right off the bat, you’ll notice that you can generate AI videos with a simple text prompt. Its video generation interface is pretty similar to that of chatbots like ChatGPT or Gemini. You simply enter your prompt and click the Send button. You can also attach media as a reference for the AI engine.

Besides video generation, Kapwing offers several other AI tools. For instance, Translate Video lets you convert the audio in a video from one language to another. There’s also a handy Smart Cut feature, which automatically removes awkward silences from your videos, giving them a more polished finish. You can also generate AI characters, which come in handy if you want to build a consistent storyline or reuse the same character across multiple AI-generated videos.

Kapwing gives you more options when generating AI characters. To get started, click the @ icon in the chat window and select the type of character you want to generate. The available options range from everyday characters like a baby or a cheerleader to more outlandish ones such as a cyborg, Dracula, or a dragon.

You also have the option to choose more than one character type. For instance, you can select both "baby" and "Dracula" if you’re trying to create something unique. Here’s the Baby Dracula character we created using Kapwing.

Apart from this, you’ll find several other options, such as Image to Video, which lets you convert a still image into a short video clip, and Text to Video, where you can enter a full-fledged script and allow Kapwing to generate an AI video for you. There’s also a handy Video Stabilizer tool, which is useful for fixing shaky footage or videos with heavy motion blur.

For longer video formats, Kapwing also supports Video Storyboards, where you can create a timestamped outline of every scene. The more traditional tools include text-to-speech, video dubbing in 40+ languages, resizing videos from one aspect ratio to another, and adding subtitles to videos.

Kapwing: Interface and in use

Almost all of Kapwing’s features are nestled into a central dashboard, with the video editor in the center, a long list of features on the left-hand panel, and an editing panel on the right. You can simply upload your media or drag and drop any video or image you want to edit. After that, you may need to spend a little time exploring the interface to find the exact feature you’re looking for.

For instance, you can access features such as Text, Audio, Subtitles, Transcript, Translate, AI Voice, and more directly from the left-hand panel. There’s also a help function that helps you quickly locate settings you can’t find right away.

The editing experience was pretty smooth. For instance, we tried adding text to an AI-generated image, and the process was seamless. We could easily change the font, color, size, and other formatting options from the panel on the right.

Note that the right-hand panel changes dynamically depending on the tool you’re using. For example, if you select Translate, you’ll see language options appear on the right.

Overall, like any video editing software, Kapwing comes with a slight learning curve. However, in our experience, it was much easier to use than many legacy editing tools. The interface is modern, sleek, and highly responsive. We didn’t experience any system lag or timeouts during our testing.

In addition to the main editing window, there is also a dedicated My Workspace page in Kapwing. Here, you can find all your recent projects at the bottom, along with a dedicated set of tools highlighted on the main page. If you’re using a Business plan, you’ll also see your team folders on the left-hand panel. Similarly, Settings, Brand Kit, and your Media Library are all accessible from the same panel.

Kapwing: How we tested

We tested the free version of Kapwing across multiple accounts to evaluate its AI capabilities more thoroughly. To start, we tried generating an AI video using the following prompt:

“A woman in a red coat walks across a rain-soaked city street at night, neon signs reflecting in the puddles, camera slowly tracking alongside her.”

Kapwing got to work as soon as we hit Send, first generating a summary of the scene we wanted to create. After that, it produced a five-second video within seconds. Although Kapwing indicated that the video would be seven seconds long, the actual output was five seconds.

We then edited the same video using Kapwing’s video editor, experimenting with features such as adding text, audio, and subtitles. Since the free plan is fairly limited, we had to use more than two accounts to fully test its capabilities.

What stood out to us was how quickly Kapwing generated both edits and the source video. Everything was processed quickly, which is a clear advantage.

Kapwing: Alternatives

While Kapwing is a capable AI video generator and editor, it is not the most affordable option out there. For instance, Adobe Firefly costs $8.49 per month when billed annually and comes with features comparable to Kapwing. You get access to popular video models such as Google Veo 3.1 and Runway Gen-4.5, allowing you to create up to 50 and 11 videos, respectively.

Adobe Firefly truly shines with its workflow features, allowing creative teams to brainstorm visual concepts on virtual boards. You can also process thousands of files at once using bulk actions such as removing backgrounds or cropping images. That said, Firefly lags a bit behind when it comes to video editing, as many of its editing features are still in beta.

It's also not ideal for long-form content generation, as most models cap video lengths at three to five seconds. On the bright side, you also get access to Adobe Express Premium and Photoshop with all paid plans.

Veed is another capable competitor, with prices starting at $12 per month, roughly 20% cheaper than Kapwing. It also offers a free plan, which appears to be more generous than Kapwing’s. For instance, you can try features like removing silences, background removal, auto-edits, social media character generation, and voice cloning on Veed’s free plan – most of which are not available on Kapwing’s free tier.

Kapwing: Final verdict

Kapwing is one of the best AI video generation and editing platforms available, with a strong mix of both traditional and AI-powered tools. Not only can you generate AI videos from simple prompts, but you can also transcribe videos, translate them into 40+ languages, create AI characters, and edit videos from a central dashboard.

Kapwing is also easier to use than many other platforms on the market, although there is still a slight learning curve involved. That said, there are plenty of video tutorials and help center articles available to help users get through this phase.

However, Kapwing can be a bit on the expensive side, especially for personal users, which is why you may want to consider alternatives like Adobe Firefly or Veed.

We've tested, reviewed, and rated a range of video makers and editors, including the best video editing software, the best video editing apps and the best free video editing software.

Apple's MacBooks seemingly have some big changes on the horizon. One of the biggest is the rumored MacBook Ultra packing that OLED screen, and this is something to be excited about, don't get me wrong. Well, worries about the price aside, and wider issues around Mac pricing overall, but there's something I'm more concerned about now.

Namely the direction in which Apple is heading with its M-series silicon, following the latest rumor dump from prolific leaker Mark Gurman. In last weekend's Power On newsletter for Bloomberg, Gurman dropped a whole lot of fresh info about Apple's CPU roadmap and the incoming M6, M7 and M8 chips.

The leaker tells us — and bear in mind, this is all just rumors, but eye-opening speculation nonetheless — that Apple is doing something very different with the next range of M silicon. The M6 will consist of just a vanilla chip, with no more powerful variants, which isn't something Apple has ever done before — there's always been a Pro and Max version (if not an Ultra, too).

This is because Apple apparently wants to usher in the M7 series more swiftly, with the M6 debuting late this year and the M7 following in the first half of 2027, just six months (or so) later. The M7 will have a full complement of Pro, Max and Ultra chips, the former two following at the end of 2027, with the Ultra in 2028. Gurman believes this is Apple's current plan and the reason for rushing to get to the M7 is AI.

Cue multiple groans, no doubt. Gurman theorizes: "The reason for this break with tradition: AI. Apple had been planning major neural-processing upgrades for the M7 family and ultimately decided those improvements were important enough to justify accelerating the next generation rather than completing the M6 lineup."

We're further told that beyond this, Apple is "already developing M8 chips with even greater AI capabilities" to land in 2028.

Gurman concludes: "The takeaway is that AI is no longer just another feature Apple's chips need to support. It is now shaping how those products are designed and when they are shipped."

A wise path to follow?

So, what am I worried about here exactly? Obviously there's a strong suggestion that AI is taking much more of a center stage role with Mac chips, and given what's currently going on with MacBooks, I've got to wonder about the wisdom of this — and how it might affect Apple's PC sales.

Currently Mac sales are strong and considerably buoyed by the launch of the MacBook Neo in March 2026, a budget laptop that has undoubtedly flown off the shelves. However, Apple's recent price hikes have poured cold water on those sales prospects to an extent, even hitting the MacBook Neo, which went up by $100 in the US (£100 in the UK, and AU$150 in Australia).

Granted, while it's odd for a device to get a price rise like this so soon after launch, given the demand around the Neo, it's not going to be sidelined with this increase (although the upper-tier model with 512GB storage now looks shakier and out of budget territory, frankly). The harder pill to swallow is the harsher MacBook Air price hikes, and as we pointed out a few weeks back, some wallet-friendly Windows 11 laptops now look a lot more tempting after Apple's price increases.

Looking on Reddit, it's not hard to find a lot of unhappiness about the MacBook price hikes. There's considerable angst from some people who were thinking of pulling the trigger on a new Apple laptop, and then watched those price increases swoop in, and are now feeling the opposite of buyer's remorse (holdout's regret?). Some of those folks are acting with their wallets, closing them and vowing to wait for a better buying opportunity now that they've missed the boat, as it were.

That could be a long wait, though. Of late, we're hearing a lot of gloomy news about the RAM crisis getting worse later this year, and in 2027 — maybe a lot worse. The memory supply situation is what caused Apple to jack up Mac prices, of course, and it's difficult to see how the MacBook maker will continue to weather this storm if it worsens next year, given that its resources in terms of built-up inventory of RAM (or SSDs) are clearly running dry (or have run dry).

Rumors have continued to persist about Apple potentially tapping Chinese memory chip makers for its RAM needs, as an alternative to the big three mainstream suppliers (Micron, Samsung, SK Hynix). That's a worrying development in itself, and it's doubtful whether even Tim Cook can pull off navigating that kind of negotiation before he leaves his Apple CEO role (there are various possible hurdles to clear and governmental complications, put it that way).

Diving into AI headfirst

I don't think any of this bodes particularly well for next year's Mac sales given the prospect of pressure around more price hikes (although yes, other PC makers face the exact same unfavorable conditions). And then we come back to this apparent renewed focus on AI, which hasn't gone down well on Reddit and other social media.

Here's one Redditor who sums all this up neatly: "I was about to buy a new MacBook to replace my M1 Pro with an ailing battery and small storage. But then Apple jacked up the prices by almost 20%. And now they are diving into AI headfirst. It might be time for me to look at Framework laptops."

There's a strong backlash against AI right now, whether that's about sprawling data centers and them chugging huge amounts of power, or Copilot in Windows 11 and indeed Apple Intelligence. So, it's no surprise that the idea of "diving into AI headfirst" has been greeted by a lot of skepticism.

What exactly is Apple Intelligence going to serve up in terms of compelling features in the next year or two? The trouble is that this feels very much up in the air, yet Apple is supposedly building its Mac strategy around this now. It seems like a bit of a leap into the unknown to say the least, and I'm not surprised that it's making Mac fans nervous.

Okay, so this AI focus is still just a rumor, although Gurman is one of the more reliable Apple sources out there, and it appears to be a well enough fleshed-out piece of speculation. Even if it's true, Apple may yet switch away from this course, but for now, I'm struggling to find the logic of shifting to a heavier AI focus so soon.

Maybe the reasons for this will become clearer as we see how Apple Intelligence develops in the near future. However, in the meantime, I think there are reasons to be concerned about Mac sales potentially flagging next year, between disgruntled MacBook holdouts, the continued RAM crisis, and cynicism around Apple's apparent new AI-led approach.

Tech companies large and small — including the likes of Meta, Samsung, and Google (and perhaps even Apple) — are betting that smart glasses become one of the leading device form factors of the future. But this week it became clear that they're facing an increasingly strong backlash against the technology.

There's one problem in particular: cameras. These tiny, embedded recording devices make it easier to record your kids' cute mishaps, your latest mountain bike ride, and a first-person view of anything without needing to get your phone out, but they also enable a certain level of surreptitious photo and video capturing that isn't possible with a phone.

And there's no doubt that bad actors are using these smart glasses for unwholesome purposes: some glasses owners have been paying modders to disable the warning light that should appear whenever recording is actively happening.

We're now at the stage where smart glasses owners are worried about wearing them in public, such is the growing negativity around them. Here's what we've learned in the last few days, and what might happen next.

An inevitable 'collapse into surveillance tech'

Smart glasses with cameras are nothing new of course, but stories around these devices being modded to operate without recording lights (something Meta has now moved to block), as well as reports that future Meta Ray-Bans could be recording all the time, seems to have prompted a new wave of anti-smart glasses sentiment.

Even the celebrities are weighing in: pop icon Lorde was less than complimentary about smart specs on stage recently, leaving her audience in no doubt about her feelings about the tech, while rapper and producer Tyler, the Creator has declared that "anyone who uses these glasses is a real weirdo".

Some of the rich and famous are still on board — you can actually buy Kylie Jenner-branded Meta Glasses, if you want — but most of the online reaction is critical of the technology. A quick peek at Reddit reveals comments like "all internet and computer tech seems to inevitably collapse into surveillance tech" and "people are choosing to be a walking Flock camera" (if you don't get the reference, read this).

Users are referencing Black Mirror and calling for the technology to be banned outright, and using terms like "creep goggles" to describe the devices. While a ban is unlikely at this point, there are some legal changes happening: Illinois looks set to be the first US state to ban smart glasses while driving, not because of any sneaky recording capabilities, but because they might prove to be a distraction to drivers.

Over in South Korea, meanwhile, a man is facing criminal charges after admitting to using smart glasses to cheat on a test. The camera captures the questions, on-board AI figures out what the answers are, and the person sitting the exam writes down the answers. There's more than one way to misuse the camera technology.

However, lawmakers don't yet seem convinced that these AI-powered glasses represent a privacy threat. Regulators in the European Union have reportedly bowed to US pressure in deciding that wearables were exempt from having to include removable batteries, like some other gadget types are when sold in the EU.

What happens next?

While there are some limited efforts to curb the use of smart glasses then, at this stage they're small scale and localized. To try and nudge things along, activists in London have been busy putting up fake Meta Glasses ads on bus stops (below), making a point about surveillance. One of the ads uses an optical illusion to turn the face of Kylie Jenner into a creepy skeleton, with the tagline "we're always watching".

Recording everything we see and do constantly? It’s giving fascism, not fashion It’s just been revealed Meta is planning to make the glasses “continuously record audio while taking photos every few seconds” without any warning light*Literally NO ONE asked for this#noncegoggles*Source: the FT

— @everyonehateselon.bsky.social (@everyonehateselon.bsky.social.bsky.social) 2026-07-17T18:40:05.934Z

Cast your mind back to 2012 when Google Glass launched, and you'll remember that we've been debating these issues for years. There are clearly benefits to wearable specs that can give you hands-free directions, translations, and information without you having to take your phone out, but it seems the concerns over covert surveillance are outweighing the positives in the public's mind.

Given the growing influence of AI tech and how well-suited these wearables are to taking in and processing information, it doesn't seem likely that the tech will be restricted any time soon. For several years now we've been seeing demos of smart glasses that can see and understand the world around them, and those capabilities are on the way whether we like it or not.

If, as rumored, Apple does make its own version of smart glasses, it's going to be interesting to see how this famously privacy-focused company deals with the issues of surveillance. A recording light that cannot be switched off is a good start, but as we've already mentioned, existing users are already cautious of wearing their smart specs around other people.

Later this year we're going to see more smart glasses from Google and Samsung, with Gemini AI ready and waiting to tell you everything you need to know about what you're looking at — which of course requires a camera. As per Google, the company is aware that people won't wear these specs if the privacy concerns are too great, and its solution is mostly to make sure the recording light is noticeable and cannot be tampered with.

"If people don't feel comfortable wearing these glasses, and privacy is part of this, they just won't wear the glasses," Google's Juston Payne told Tom's Guide this week. "It’s as simple as that." The new Google and Samsung wearables are scheduled to arrive later this year, but the companies involved still have a way to go when it comes to shifting user opinion.

• Forrester analysts warn around four in five leaders and ITDMs envision having larger budgets in 2027
• Consumption-based AI pricing is making it harder to predict outlay
• Targeted investment to improve data quality is key

Forrester is predicting software budgets could be set to rise, with more than four in five leaders expecting to increase overall budgets over the next 12 months and 82% of tech decision-makers expecting larger budgets.

While some of the extra cash could come as a result of increased confidence and readiness to spend on tech, the company's analysts warn that a shift in pricing structures could also be forcing companies to fork out more.

This comes as software vendors shift from traditional per-seat licences to token or credit pricing, which introduces so many more variables including model selection, context size, output length and agent operating time, leading to far more unpredictable outgoings.

The real reason businesses are preparing to spend more on software

"Business leaders are no longer planning for a return to stability – they’re planning for a future where volatility is a constant,” Chief Research Officer Sharyn Leaver noted.

Recent shifts from major AI providers all point toward this emerging pricing model becoming the norm, with GitHub moving its Copilot plants to usage-based billing in June and OpenAI adding pay-as-you-go Codex seats in April. Anthropic also recently removed Fable 5 from its standard subscriptions and seat-based models over difficult-to-predict demand, but set out plans to reintroduce it where capacity permits.

Acknowledging these major shifts, Forrester's report reveals two areas where companies can increase their budgets for 2027 – building machine-readable context and enterprise knowledge, and increasing brand visibility in answer engines.

The report also hints at the major role AI can play in marketing and customer-facing experiences, and the potential use that synthetic data can provide subject to testing. All in all, it's more about targeting investments rather than throwing cash at the problem, as Leaver concludes:

"The organizations that outperform in 2027 won’t be those that spend the most on AI. They’ll be the ones that invest in the foundations that make AI effective."

In May, WordPress shipped the most consequential release in its history. Version 7.0 brought AI into the core of the CMS platform for the first time, and the people who built it made a choice that's easy to miss in all the noise about the feature itself. They left it switched off.

The infrastructure is in the codebase, but nothing reaches an AI service until the site owner connects a provider and turns it on. Upgrade a site and walk away, and it behaves exactly as it did the day before.

The on-switch was handed to the person who owns the site, not flipped on their behalf.

It's worth sitting with how deliberate that was. The team had just shipped the most powerful capability the platform has ever carried, and the posture they chose for it was opt-in, plugin-based, with nothing injected into anyone's site automatically.

In an industry that loves a sensible default, that restraint was itself a statement: this decision is yours to make.

Then the more interesting thing happened.

A reasonable instinct, taken one step too far

Within days of the release, SiteGround, one of the most established managed WordPress hosting companies in the business, did close to the opposite. It pre-installed and activated its own AI product across its customer base, configured it as the default connector, and bundled in a generous allowance of free usage to get people going.

The active-install count crossed a million almost immediately. Plenty of site owners logged in to find capable new software already running on sites they had never touched to put it there.

I want to be fair to SiteGround here, because fairness is where the useful lesson lives. This is a serious operator with a long, well-earned reputation, and the product it built is genuinely good, a real piece of engineering rather than a thin upsell. The reasoning behind the rollout isn't hard to reconstruct either.

The "correct" path to native AI is fiddly, and most people would stall somewhere in the middle and never finish it. Pre-installing the whole thing, free usage attached, removes that friction in a single stroke. From an operator's chair, that's a tempting piece of customer service, and I've sat in that chair for the better part of two decades. I understand the pull of it completely.

So this isn't a story about a company behaving badly. It's a story about a reasonable instinct (reduce friction, help the customer get to the good part faster) carried one step past the line. And the reaction told us exactly where that line is.

The objection wasn't AI. It was consent.

The pushback

The pushback was quick and pointed, and the striking thing about it was its subject. Almost none of it was about whether AI belongs in WordPress, or whether the tool was any good. Many of the people objecting use AI every day. What they objected to was finding it already switched on.

That distinction matters more than it first appears, because it separates two things the industry tends to blur: the quality of a change, and the consent to it. A genuinely good feature, installed without asking, still lands as something done to you rather than for you.

The standard defense (it's optional, you can remove it whenever you like) is all true, and none of it is the same as agreement. "We switched it on and you can switch it off" quietly moves the work of noticing, understanding, and undoing onto the customer, for a change they never approved. "Here's one-click setup if you'd like it" delivers the identical convenience and leaves the decision where it belongs.

This isn't a new tension. Webhosting companies have always made changes customers never see, and most of the time they're glad we do. But AI is going to surface this question over and over, because it's the most consequential thing most of us will ever be tempted to switch on by default. Getting the principle right now, while the stakes are still mostly reputational, is a lot cheaper than getting it wrong later.

The line worth holding

The honest objection to all of this is that hosts intervene on customer sites all the time, and nobody asks permission for that. True, and the distinction is the whole point.

When a host patches a vulnerability, blocks a malicious request, or disables a plugin that's being actively exploited, it's protecting the customer's site and the wider platform from harm. Customers extend us that trust precisely because it's defensive, narrow, and in their interest. Installing a new product is a different category of act.

It isn't protecting anything; it's changing what the website is. The trouble starts when the second borrows the permission we were granted for the first, when goodwill extended for security work quietly gets spent on shipping features. That's the line. Maintain the platform freely; change the product only with a yes.

Holding it doesn't mean making customers do more work. New capabilities can arrive off by default and one click away for anyone who wants them. Multi-site managers can get a single place to see and control what's running, rather than a hunt site by site.

Anything a host pushes can be pulled back as easily as it went out. And changes can be announced in plain language before they happen, including how to say no, because the absence of a clear, opt-out-inclusive heads-up is usually what turns an ordinary product decision into a breach of trust.

Parts of the ecosystem are already moving this way. None of it is anti-AI. If anything, it's what lets hosts lean into AI confidently, because customers can trust that nothing shows up uninvited.

Whose site is it, anyway?

As AI moves from novelty to default across the web, every host will face its own version of this question. Here's a genuinely useful new capability. Do we switch it on for everyone, or do we let people choose? The convenient answer and the right answer won't always be the same one, and the gap between them is where reputations are quietly made or lost.

It helps to remember who actually lives with the answer. When a host changes something on a site, the host moves on to the next ticket. The owner is the one who stays: the one whose visitor hits a page that behaves differently than it did the day before, whose inbox fills up when something looks off, whose name is on the business the site exists to represent. We get to make the change. They have to live with it. That asymmetry, more than anything written into the terms of service, is the real reason asking first isn't a nicety. It's an acknowledgement that the consequences were never ours to carry in the first place.

The site owners who pushed back this spring weren't standing against progress. Most of them, I'd wager, will happily adopt the very tooling they objected to, the moment they get to be the ones who switch it on. They were defending something simple that's easy to lose sight of when the technology is moving this fast: it's their site. Not the site we host for them. Theirs. A host's authority runs right up to the edge of the customer's ownership and stops there, and the best operators I've worked alongside never needed reminding of it. They saw their role as stewardship rather than possession.

That trust is the real product. Not the servers, not the dashboard, not even the support, though every bit of it matters. What a customer is buying is the confidence that nothing happens to their site that they didn't choose, and that when we do step in uninvited, it's to protect what's theirs and never to quietly redraw it. Trust like that takes years to earn and an afternoon to spend. Asking first is simply how you keep from spending it.

Get the boundary right, and AI in hosting becomes exactly what it should be: useful, and genuinely welcome. Get it wrong, and even the best feature in the world arrives as something taken rather than offered. The difference was never the technology. It was only ever whether anyone thought to ask.

We tested, reviewed, and rated the best website builders.

This article was produced as part of TechRadar Pro Perspectives, our channel to feature the best and brightest minds in the technology industry today.

The views expressed here are those of the author and are not necessarily those of TechRadarPro or Future plc. If you are interested in contributing find out more here: https://www.techradar.com/pro/perspectives-how-to-submit

• A report commissioned by Mozilla accuses Microsoft of using 'harmful design' to undermine browser choice in Windows 11
• That allegedly includes trick wording, nagging, preselection, and generally dubious tactics to push Edge over other browsers
• The report does observe the situation is better in the European Economic Area, due to regulations there, and notes that "regulatory action works", urging authorities elsewhere to take a similar stand

Mozilla, the maker of Firefox, has published another report that claims Microsoft isn't maintaining a suitably level playing field for web browsers on Windows, and is unfairly pushing users towards Edge.

As Mozilla makes clear, it published the first 'Over the Edge' report – about 'How Microsoft's design tactics compromise free browser choice' – two years back, and the follow-up has just arrived (as spotted by Eteknix).

It's not actually written by Mozilla, I should note, but two independent researchers commissioned by the Firefox maker (Harry Brignull and Cennydd Bowles, the same pair who wrote the first article). It takes in user opinions from the US, UK, India and Germany regarding "key browser-choice journeys" on Windows 11 and 10.

In a nutshell, they conclude that: "Microsoft continues to deploy harmful design to undermine people's browser choice."

That specifically involves, according to the report: "Trick wording, obstruction, visual interference, preselection, nagging, and forced action — at almost every step of the user journey."

Some of the worst excesses highlighted are the pop-up banner that Edge users see when they go to the Chrome download page (trying to get them to stick with the Microsoft browser, which, it says, has the same tech as Chrome but with the "added trust of Microsoft"), and nagging within Windows around using Edge as the default browser in one way or another.

The report also notes how Edge is pre-pinned to the Windows taskbar, and how a migration from Windows 10 to Windows 11 resets Edge as the default browser, overriding any previous choice. There are a lot of accusations around trick wording, too, as well as an observation that Microsoft is using Copilot (AI) to open links in Edge rather than your default browser.

On the topic of AI, the report notes: "The researchers suggest that a sequence of seemingly minor consent requests across Windows and Edge may combine into a 'pipeline' funneling browsing data — potentially including data originating in rival browsers — into Microsoft's advertising and personalization systems."

It's notable that Germany — included as a representative country within the European Economic Area (EEA), which has different regulatory requirements Microsoft must adhere to — escapes quite a few of these excesses (the Chrome download nonsense included).

Despite that, the researchers argue that their broad conclusion that "Microsoft does not allow people to download and install an alternative browser, to set it as their default, or to continue using it as their default, without harmful interference" is, they claim, true "across every region tested".

Analysis: a call to regulate

Yes, this is a report commissioned by a browser rival, but these findings aren't at all surprising in the main, and it's long been known that Microsoft overzealously promotes Edge in Windows 11, trying various dubious shenanigans to get the browser in play as the report observes. Of course, on the flipside Google pulls similar stunts with Chrome.

Still, that's not an excuse, and no browser should be actively attempting to dissuade people from downloading another rival browser. What Mozilla is also pointing out, of course, is that Microsoft is in a particular position of power here, given that it owns the world's primary desktop OS, and it's leveraging that platform in various ways to push adoption of Edge.

Whether that's worked for Microsoft, or not — spoiler alert, it hasn't — is irrelevant, as this behavior obviously isn't ethical or fair, and while it may not have dented Chrome's reign as the top browser, it may have damaged smaller rivals like Firefox.

While the EEA may have a considerably better balance of browser equality within Windows, the report notes that this is, of course, because regulations have forced Microsoft's hand. Arguably, this is another negative in that it shows Microsoft only respects user choice in terms of browsers when it's forced to do so.

Mozilla finishes with the following paragraph that sums everything up nicely: "We again urge Microsoft to abandon these harmful patterns worldwide. And we urge regulators in the European Union, the United Kingdom, the United States, India, and beyond to consider this report as evidence that regulatory action works — there is much still to do."

So, if Microsoft really wants to talk about "added trust", how about some action on some of these sticking points? Or is this just a case of having to be more trustworthy than Google?

The software industry has always been shaped by new tools that promise speed and efficiency. From high-level languages to cloud infrastructure, each wave of innovation has reduced friction and raised expectations. The current wave is AI-generated code, but this new era of innovation is different in both scale and consequence.

For the first time, developers can produce large amounts of working software without fully understanding how it functions. This shift is creating a new challenge that engineering leaders are only beginning to articulate, a concept we’ll call comprehension debt.

Technical debt is a familiar concept. Teams knowingly trade long-term maintainability for short-term delivery and accept the cost later. Comprehension debt is a similar concept but where that debt resides is different.

Whereas with technical debt is located within a codebase, comprehension lives in the people building the systems. When developers ship software faster than they can truly understand it, the gap between output and understanding of the codebase widens. Over time, that gap can grow to be a serious organizational risk.

This tension is already visible across the industry. The most recent Developer Survey found that 84% of developers use or plan to use AI tools in their workflow, yet 75.3% say they do not fully trust AI-generated answers.

The findings highlight a growing contradiction at the center of modern software development: teams are increasingly dependent on AI, while remaining cautious about the quality and reliability of its outputs.

The disappearance of productive friction

For decades, the early years of a developer’s career were shaped by friction. Junior engineers learned by wrestling with compiler errors, reading documentation, and debugging unfamiliar systems. The process was often frustrating, but it helped developers build mental models of how software behaves and understand not just what worked, but why it worked.

Today, AI tools remove a good deal of friction. A junior engineer can generate a functional service, build an interface, and resolve common errors within minutes. The experience feels empowering and productive, but the friction was never purely an obstacle, it also acts as a teacher. When that struggle disappears, some of the learning disappears with it.

This does not mean the next generation of developers will know less. In many ways, they will be exposed to more complex systems earlier in their careers. However, developers may begin to feel productive before they have gained the deeper intuition that traditionally comes from years of problem solving and debugging. The sense of progress may outpace genuine understanding.

Faster progression, weaker foundations?

AI is also reshaping career development. Traditionally, engineers progressed from writing small pieces of code to reasoning about complex systems and eventually designing architectures.

AI compresses the earliest stages of that journey and engineers are able to contribute sooner and move more quickly towards higher levels of responsibility. In the short term, this looks like success with teams being able to deliver faster, and organizations expand their talent pipelines.

The long-term risk arises down the road, when organizations find themselves with engineers who can produce solutions quickly but struggle when those solutions fail in unexpected ways. Debugging, architectural thinking, and systems design all rely on deep mental models that are built slowly through experience. If those foundations are weaker, the consequences may only emerge months or even years later.

Enter vibe coding

Alongside this shift, the explosion of vibe coding as a practice, which entails a workflow built around prompts, rapid iteration, and intuition rather than deliberate engineering, has further exacerbated comprehension debt. There is genuine value in the vibe coding approach. Rapid experimentation has always driven innovation, and AI has the ability to unlock creativity in remarkable ways.

The danger appears when this becomes the default mode of production rather than a tool for exploration. When speed becomes the dominant priority, understanding can quietly become optional. Teams may accept generated code without fully reviewing it, and systems may be deployed that few people feel confident enough to explain in detail. The result is not immediate failure, but a gradual erosion of collective understanding.

Some skepticism about new tools is inevitable. Every generation has worried that higher levels of abstraction would weaken engineering skills. However, AI introduces an important difference. Previous tools abstracted complexity while still requiring developers to reason about behavior.

AI can now generate working solutions without the developer forming the underlying mental model at all. That changes not only how software is written, but also how expertise develops.

Building AI-native engineering cultures

The answer is not to resist AI adoption as the benefits are too significant to ignore, and organizations that fail to embrace these tools risk falling behind. The real challenge is adopting AI while protecting the growth of genuine understanding.

Some organizations are beginning to view comprehension as an explicit goal rather than an assumed outcome. Engineers are encouraged to explain generated code in their own words and document the reasoning behind AI-assisted decisions. This approach reinforces that understanding is part of the deliverable, not a nice to have.

Code review is evolving too. Historically, the focus was whether the code worked and met quality standards. Increasingly, teams are also asking whether the developer truly understands what they have produced and review conversations are shifting towards explanation, walkthroughs, and shared learning.

Hands-on learning also remains essential. Some teams are intentionally creating environments where AI plays a secondary role, whether through debugging exercises, architecture discussions, or smaller projects built from scratch. The goal is not to reduce productivity, but to preserve the experiences that build engineering intuition over time.

The most effective organizations are not choosing between AI and traditional development, but rather thoughtfully combining the two approaches. AI is increasingly used for repetitive tasks and scaffolding, while humans retain responsibility for critical thinking, system design, and architectural judgement.

This moment places a new responsibility on engineering leaders. Productivity is accelerating rapidly, yet learning must keep pace. If we succeed, AI will help create a generation of developers who build faster and think more deeply than ever before.

If we fail, we risk creating teams capable of shipping almost anything, yet uncertain about how the products they build truly work or how to fix them when something inevitably goes wrong.

We've featured the best AI website builder.

This article was produced as part of TechRadar Pro Perspectives, our channel to feature the best and brightest minds in the technology industry today.

The views expressed here are those of the author and are not necessarily those of TechRadarPro or Future plc. If you are interested in contributing find out more here: https://www.techradar.com/pro/perspectives-how-to-submit

For the longest time, I thought that, despite their annoyances, iPhones had something big over the best Android phones: PC connectivity. I'd always seen iPhone and Mac-owning friends quickly send files between the two, and assumed it was an Apple-exclusive feature.

As it turns out, this isn't true. Android users can use a Windows app that brings loads of really useful features — and it's really underutilized, which is why a tech journalist like me hadn't heard of it.

This software is called Phone Link, and we've got a guide on how to connect your Android phone to a Windows PC using Phone Link elsewhere on TechRadar. I downloaded it about a year ago, largely out of duty to test every tool, appliance, and gadget I could get my hands on. I was expecting it to last a weekend.

A year on, though, and Phone Link has become such a useful part of my workflow that I'd forgotten it's something not everyone's heard about. It's such a natural part of the way I use my phone and computer that I'd totally forgotten it was something I initially downloaded to test!

Well, no longer: TechRadar has let me wax lyrical about Phone Link, and I'm going to give the tool its time in the sun.

Sending photos from phone to PC

For the most part, whenever I want to connect my phone to my PC, it's to move pictures between the two devices.

As a smartphone journalist, you can understand that I'm sending snaps from my phone to my computer a lot. Sometimes it's camera samples from a mobile I'm testing that I want to upload for a review. Occasionally, I want to back up pictures I really like before I wipe a phone and send it back to the company.

The most frequent use case, though, is to transfer to my PC pictures I've taken on one phone, of another. I use smartphones to do review photography, and it's a great way to test out a phone I'm reviewing.

Before Phone Link, I'd have to use a USB cable to connect my phone to my computer — approving the connection on my Android's end, because they're always annoyingly suspicious — and drag and drop all the relevant pictures into a folder on my computer. It's not an especially onerous process — it's a lot quicker than when I used to use Google Drive to transfer pictures, for instance — but it takes a little while of fiddling with my phone and digging out a cable to work.

Not so with Phone Link. Its Photos tab shows a big old list of all the pictures you've taken recently, giving you options to save them to your PC, open them in an editing app, and share them with others.

These pictures appear as soon as you take them, as long as your phone is on the same Wi-Fi as your computer. I've become used to taking photos in one room of my home, and entering my office, to see them linked up on Phone Link, ready for an edit. When I've got PC notifications turned on, I can even hear the 'bong' of my computer telling me they've arrived.

Phone Link also lets you browse the contents of your mobile in the Files tab, where it appears in the Devices and drives list alongside any cloud or local drives you have. I don't use my phone to handle non-photo files, but it's a nice touch that'd be really useful if I were in another line of work.

Apps on my PC

Ostensibly, one of the main uses for Phone Link is that it lets you open your smartphone apps on your computer. Its Apps page shows you a massive list of every one you have installed on your mobile, and clicking on it opens it in a window.

This doesn't override anything you have going on on your phone; if you're using it as a second screen to watch a YouTube video or act as a music player (one of many ways you can repurpose an old Android phone), the app will open in your PC window alone.

When I first started Phone Link, I tried using it for everything, but often bounced straight off. In many circumstances, using a smartphone is simply more intuitive than using a mouse and keyboard, given that phone apps are designed for thumbs. But over time, I've found a few uses of Phone Link's app windows that are genuinely handy.

As mentioned, one of those uses is as a music player. I can let Spotify hang around as a spare window, ready to skip tracks or change playlists when I want. It's also useful as a way to check apps that don't have PC equivalents, like Too Good To Go or Mubi Go, or ones that do have web functions but are easier on the phone, like my national health service's app.

There's one way that Phone Link saved me hours, though. When I used to work for a company that didn't let you use your work Gmail account on personal computers, Phone Link was my workaround to still see emails on my PC. This saved me ages each day; time that'd otherwise be taken up checking and replying to emails on my phone, or trying to get my decrepit work laptop to turn on.

Naturally, you can manage your calls and texts via Phone Link too. Personally, I don't think I've sent an SMS or made a non-WhatsApp voice call since the 2010s, but I'm sure there are some people who'd find this handy.

I also appreciate a notification list down the left side of Phone Link. This transforms it from being software solely for controlling your mobile to a veritable hub of information from your handset. I'll often keep it up on my second monitor; it makes good use of its space.

Controlling my phone from my computer

As mentioned, you can use your computer to control your mobile's apps via Phone Link, but you can also use it to change the actual settings of the device itself, which proves itself useful in an entirely different way.

Phone Link lets you monitor your phone status; I can see how it's connected to other devices, what battery it's on, and whether it's currently paired with Phone Link or not. I can even see a little representation of what my wallpaper currently is. All useful to a small degree, but the charge is the only one of those that's seriously useful.

What's more important are the status controls. You can toggle between vibrate, sound, and silent, turn Do Not Disturb on and off, and turn on the media player.

From a cursory mention of the tools, these might not sound that important, but I've found them really useful in a pinch. I'm constantly forgetting to mute my phone before I go on a video meeting, and a quick click of the Do Not Disturb button means I'm not going to have unexpected calls.

As a lazy person, I also love the fact that you can use Phone Link to play a sound from your phone; I'm always doing this instead of spending five seconds actually looking.

My one gripe with Phone Link is that I still need my phone with me to use it. When you start using it each session, you're required to unlock your device and approve the connection. Generally, you're done with your phone after that, but sometimes I'm unexpectedly forced to re-unlock my device, which I've invariably placed somewhere stupid and forgotten about. It's a small annoyance, but if that's the price to pay for an otherwise-smooth workflow, so be it.

• Windows 11's July update has arrived
• It packs some smart features, including pausing updates at length and a new recovery option
• This patch also cures a nasty bug that slowly eats more and more drive space, but there's an issue to be aware of for Dell laptop owners

Microsoft has released its monthly patch for Windows 11 and the July update is a biggie, packing a feature that I've been awaiting for quite some time.

That would be the ability to pause updates on Windows 11 Home – beyond just a short period of time, which has been the only choice to date – and there are some other impressive additions from Microsoft here.

I'm going to pick out my top four features introduced by the July update, as well as rounding up some of the other more notable changes here. I'll also highlight an important bug fix that has been deployed for a nasty flaw that eats your drive space – along with a warning for some of those with a Dell laptop (who may not get this update for a good reason, as you'll see).

As ever, these changes were seen in the last optional update (June preview), and most of these features are being rolled out gradually, so you may not see them right away after installing this update (indeed, you may have to wait a while based on how the update winds blow regarding your exact PC configuration).

1. Pausing updates

Sometimes you want to hold off installing an update because of a bug that you're nervous about potentially affecting your system, and up until now, Windows 11 users have only been able to avoid any given update for five weeks.

Okay, so that's a fairly long time, but what if the bug is one of those persistent and niggly affairs that Microsoft takes an age to fix? You may want to delay the update beyond five weeks, and now you can. The catch is that you'll need to continually renew the delay for every 35-day period, but nonetheless, you can now keep an update at bay for as long as you want. (Well – until you're forced to update to a new version of Windows 11, that is, when support runs out for your current version).

2. Point-in-time restore

Another neat addition is point-in-time restore, a new recovery option that you can employ when something bad happens to your PC and you can no longer reach the desktop. Point-in-time restore lets you roll back a non-functioning system to a previous working state. It reverts to the "full system state" meaning that all your apps, files, and settings are recovered as captured at that previous time.

Hopefully this is a feature that you'll never need, but if you do, you'll be very glad of it. Note that the system backups obviously take up drive space, with older backups automatically deleted after 72 hours by default. Also, if you have a system drive that's smaller than 200GB, you'll have to enable this feature yourself (as it won't be turned on automatically due to potential space issues – note that you can specify a maximum amount of storage space used).

3. Fewer annoyances with widgets

The July update is making the widgets panel 'quieter' by which Microsoft means that it has cut out a lot of clutter. The main change is that by default the panel just displays your widgets and Microsoft has got rid of the promotional nonsense in terms of ads and the MSN feed. Notifications and taskbar badges are also minimized by default, and widgets no longer open when you hover over them.

There are some other streamlining moves here, and even if you don't use widgets, I think this is a noteworthy change as it shows the direction Microsoft is now heading in with Windows 11 – chilling out on the upselling as was previously promised (if not entirely banishing it).

4. Screen tint

Windows 11 has been graced with a number of accessibility improvements in the July update, first of which is a new screen tint ability. This does what it says on the tin, allowing you to tint the screen with a choice of colors to help reduce eye strain, or to make text more easily readable. You can adjust the tint intensity, and this is a very useful addition that'll save people from downloading a third-party app to get these kinds of color overlays.

Other features in the July update

There's quite a lot going on here aside from the above major moves, and that includes Microsoft boosting the speed at which File Explorer launches (helping Windows 11 to feel more performant overall).

There's a whole lot of work on the Bluetooth front, too, as Microsoft notes: "This update improves reliability and performance when connecting to and using Bluetooth devices." That includes better driver stability, improved audio routing for calls via Phone Link, and more.

File Explorer has also been honed to be more reliable, and Microsoft has implemented a performance tweak to the Background Intelligent Transfer Service (BITS) that means PCs will shut down more promptly. If you've ever sat staring at your computer while it takes 20 seconds to shut down, wondering what on earth it's doing, that kind of behavior will hopefully be a rarer occurrence going forward.

A major bug fix – and Dell laptop problems

There's some good news for those of you who have been suffering at the hands of a mysterious bug that eats storage space (due to an out-of-control database file). This is fixed with the July update (the cure was in the June preview before it), so you can grab this to resolve the issue (hopefully).

There's some bad news on the bug front, though, namely that you won't get this update on some Dell devices with Intel CPUs, as Microsoft has blocked it due to the July patch causing all sorts of weird issues. I say it's bad news, but at least Microsoft caught the problem and has prevented the update from being piped to those machines.

Microsoft informs us: "This update might not be available for a limited number of Dell devices with Intel processors due to an incompatibility reported by Dell that can potentially cause unexpected shutdowns, poor performance, increased heat, and battery drain. We are working together with Dell to prevent the affected models from experiencing the issue and plan to release a resolution for affected devices in the coming days."

So, if you were wondering why you couldn't get this patch on your Dell laptop, now you know. You'll just have to sit tight for now until Microsoft and Dell get this ironed out.

• White House, Treasury, DHS and DoW come together to launch Gold Eagle scheme
• The initiative will prevent duplicated work and prioritize vulnerability remediation
• Gold Eagle will also help to identify which systems could be at risk

The US Government has launched Gold Eagle, a new clearinghouse which looks to centralize vulnerability discovery and remediation against a backdrop of evolving AI-powered security threats.

Gold Eagle will serve as a central hub between federal agencies, AI developers, open-source software developers and critical infrastructure companies, in a bid to increase the speed of vulnerability discovery and prevent major incidents from occurring in the first place.

The scheme came about under President Trump's June 2 2026 executive order 'Promoting Advanced Artificial Intelligence Innovation and Security' and represents collaboration between the Treasury, the DHS' Cybersecurity and Infrastructure Security Agency (CISA) and the Department of War.

US Gold Eagle scheme addresses growing vulnerability exploitations

Under the scheme, vulnerabilities scanning will happen centrally to ensure multiple organizations aren't independently repeating the same work. Gold Eagle will also identify which software, networks and critical infrastructure could be at risk, before coordinating fixes. The White House described the scheme as a "force multiplier."

Although AI is largely to blame for the increase in attacks, Gold Eagle is set to fight fire with fire by employing AI to identify bugs too, using models like Anthropic's Mythos.

"Through this strategic partnership, we will expand existing security measures to safeguard software and networks in the 21st century and continue to promote advancements in artificial intelligence," DHS Secretary Markwayne Mullin wrote.

The concept of a dedicated clearinghouse centralizes vulnerability management to ensure the right bugs are being prioritized and to cut through the noise of lower-quality reports. Its assistance will most likely be felt by the open-source community, which has limited resources and financial backing to identify and fix issues as effectively as enterprise software vendors.

"Under the leadership of President Trump, we are bringing a wartime footing to the cyber domain to relentlessly patch vulnerabilities," Secretary of War Pete Hegseth added.

• Microsoft is improving Windows 11 search in multiple ways
• That includes a calmer search panel, the ability to turn off web results, a better way of prioritizing returned results, and a more stable search overall
• This is in testing for now, and the main worry is how long it might take for Microsoft to usher all this through to release

If you're fed up with the way Windows 11's search functionality works – and you wouldn't be alone – some incoming changes are set to greatly improve this experience.

Microsoft revealed the tweaks to the search box in a blog post, explaining that they're being delivered via a gradual rollout to testers in the Experimental channel for Windows 11 preview builds.

The first major change is simply to make search a calmer place, so when the panel appears, it only contains a list of your recent searches (allowing you to easily fire up one of those again if you wish). The current clutter to the right of that list, including recommendations – some of which are outright adverts – along with trending searches has all been banished.

And the second important piece of work here is that Microsoft is finally giving Windows 11 users the ability to get rid of web results in search. Currently, when searching for a file in the operating system you get not just local results (for files on your drives), but also some web results which can get in the way.

In this new scheme of things, you'll be able to switch off all web results in Settings, and also ditch Microsoft Store suggestions too.

If you keep these results on, Microsoft notes that they won't be prioritized, although this is something the company had already started to address (it used to be the case that web content could appear at the top of the returned results in a truly baffling fashion). Web results have also been stripped of any 'promotional content' so you'll only get the most relevant answers if these results are enabled.

Microsoft has also bolstered the handling of results for Windows 11 settings so that more relevant options are flagged up higher in the pecking order, and Microsoft says further fine-tuning is planned on this front.

The search box will also be able to handle typos, with an improved ability to guess what you really meant ('Chrome' rather than 'Chome' for example), and it'll start to surface possible results after typing just two characters (another enhancement we've already heard about).

While this is mostly all about streamlining, Microsoft is going the other way in one notable respect — adding a bit more detail in for the files returned in the results. Search will now provide more info about files (such as when they were last opened) and a more in-depth preview in the right-hand panel, so you can more readily tell what you'll be opening if you click on that result.

Finally, Microsoft is making search more reliable, which involves "reducing [the] likelihood of crashing and loading issues" which is obviously a welcome move.

Analysis: the rollout fly in the soothing search ointment

Windows 11 search has always been a somewhat painful affair for me, thanks mainly to the clutter of the thinly veiled (or not even veiled at all) ads, and the irrelevant results which were surfaced with a baffling level of priority, as noted. Windows 10 isn't a lot better, frankly, but at any rate, I'm very happy to see the fresh direction Microsoft has taken here.

The calmer search results are a major boon, with Microsoft following in the same vein as the streamlining it recently announced for the widgets panel, which was made a quieter place. (All this follows a broader early promise to rein in the upselling with Windows 11 and generally make the interface a more chilled place).

I was hoping that the option to turn off web results would be brought in, and this is the most important addition for me personally. I'm not alone, and the reaction to these various changes has been very positive, with the only real sticking point being impatience around when these new features will be rolled out. That's both in terms of testers wanting to try the revamped search box now, and the general computing public wondering when they will eventually get their hands on all this.

As this tester posts on Reddit: "We love this... except gradual rollouts. We don't like when things are inconsistent from PC to PC despite running identical [test] builds [of Windows 11]. It's really obnoxious and breaks muscle memory. But the changes seem great and I look forward to trying them."

As Microsoft points out in its blog post, it's worth remembering that there are now feature flags that can be enabled if you're really keen to try out something in a preview build, and it's not on your PC yet.

However, as for those asking: 'For those of us on release, when might this arrive?'

Well, that's a very different kettle of fish, as it could take quite some time for this work to progress through testing. Mainly because there are a lot of changes involved, and it's not something Microsoft will want to rush (especially given the drive to make search more reliable and stable).

And when these search improvements are eventually released outside of testing, it'll be on a controlled rollout that you won't be able to jump the queue with. However, there's a reason for that – Microsoft will need to observe the changes going live in a gradual manner to ensure no unexpected gremlins are crawling around in the works.

So, you will have to be patient, and there's some frustration around the length of time it can take for controlled rollouts to proceed these days (the Start menu overhaul from last year being a key example here – some folks still don't have it even now). But the good news is that these changes are coming, and they promise to revamp search in Windows 11 to a considerable extent. I can't wait (but I'll have to).

London Tech Week’s focus on AI - from a £12 million investment in AI for SMEs to AI bootcamps for graduates and more - has reflected the pressure to compete in an AI-era.

As this digital revolution progresses, the job economy is changing, but the mantra that AI is taking our jobs is simply not correct and potentially fueled by an undercurrent of classicism.

When the Luddites famously started to break the first machines of the industrial revolution in 1811 in England, fearing for their job as textile artisans, the “Bourgeoisie” would describe them as “ignorant workers”, with no understanding of basic economics.

More than two hundred years later, with the rise of GenAI, it is no longer the blue-collar workers who fear for their job, but the white-collar workers. This time it is the “bourgeois” who live in the anxiety of an uncertain world.

Since ChatGPT introduced AI into the everyday lexicon, it has been clear that we would experience an unprecedented revolution. The rhetoric that immediately began to dominate social discourse has been that AI tools would render most jobs insignificant.

Furthermore, whilst other technological revolutions ended up being creative destruction, ‘this time it was different’.

But is that really the case? Or are we more fearful, more concerned about destroying the status quo, because this time it’s a different ‘class’ of people being impacted? This time it’s the desk workers, not the physical laborers, who risk losing jobs, and suddenly there is alarm.

Artificial Intelligence relies on humans - and more humans than ever

AI is a human creation and still relies on humans to evolve. First, we have those who build the infrastructure, like data centers, which accounted for almost all of the United States’ GDP growth in the first half of 2025 (according to Harvard economist Jason Furman).

Then, we have those who train the models, which still need to be constantly retrained. Even if models are able to train themselves eventually, there is no consensus that human intervention in training will become obsolete, because human behavior and the entropy of organizations are in a constant state of flux and evolution.

And even when trained, AI constantly needs to also understand the “context” in which it is prompted to perform efficiently. AI then needs to be deployed. Managing security, defining guardrails for agents, understanding how to use AI and tracking agentic AI’s actions, all comes with inherent challenges.

The CIOs of the largest global corporations are already investing hundreds of millions of pounds to understand this. Startups based in San Francisco - a city I recently visited where 95% of out-of-home ads were about AI agents - are focused entirely on resolving these problems for large enterprises.

The fact that both Anthropic and OpenAI have launched their own consulting companies is proof that managing AI complexities in the coming years will be the biggest source of growth for all consulting and outsourcing companies of the world.

Sourcing the right human talent in the AI era is the biggest challenge

Software engineering is a job category where GenAI - perfectly trained on open-source code and GitHub repositories - can now code better than even the most experienced developers.

Additionally, developers in AI labs - with privileged access to “tokens” on Claude Code or OpenAI Codex - now develop 100% of the time without writing a single line of code. Nonetheless, when asked about their biggest challenges, all AI startups would point to recruitment.

A report by the UK's National Foundation for Education Research showed a 50% increase in tech job adverts between 2019/20 and 2024/25, with entry-level roles particularly affected. However, we’re now seeing a surge in demand driven by Gen Z, according to Employment Hero’s March Jobs Report.

This demand for AI expertise is reflected in a new Malt Tech Trends Report, which analyzed 1.2 million searches of tech freelancers in 2025. It reveals that AI is now the second most-in-demand skill, irrespective of company size, industry, or project type. More specifically, demand for freelancers with agentic AI expertise exploded by 5,800% in just twelve months.

Observer of the AI revolution, Andrew Ng, explains that if, for example, a team of 3 developers builds 10 times faster, then they need more designers or product managers to fuel the creative process. Doing more faster, with fewer people creates more work to fuel and execute the output.

More people are echoing the same rhetoric as Ng, calling out the phenomenon of ‘AI washing’, whereby companies have justified mass redundancies with AI disruption. In reality, in many cases, they were either adapting to geopolitical and economic uncertainties or had simply employed too many in the crazy post-COVID bull market.

The AI job apocalypse is not yet here… Still, the fear is real and needs to be understood

Software engineering is a perfect example of a job category that has constantly evolved. Since the inception of computer science, programming has become progressively more about “natural language”. Whilst there were 50,000 developers worldwide in the 1960s, today there are almost 50 million. Undoubtedly, the eradication of barriers to entry to build software increases that number tenfold.

History, data, and observation shows us that the AI job apocalypse is not yet here. Still, the fear is real and needs to be understood. The reason every science fiction novel paints an inhospitable world and unattractive paradigm is because the human mind always fears change. We assume the worst.

AI transformation, like all transformations, will be a cultural change first. And it’s companies, not professionals, who are most at risk if they fail to adapt. If one thing will be different in this digital revolution, compared to the last (arguably comparable is the advent of the internet), it’s the rate of change.

CEOs will have to be imaginative, change org charts and processes, admit they are not omniscient, take risks, and invest in training. Schools and universities also face the challenge of teaching soft skills: how to adapt to live and work in a more uncertain world. Because we can only harness top-tier AI talent if we understand how to truly adapt to change.

Independent professionals - those who create their own roles - from freelance developer to fractional manager and strategic consultants - have already redefined work.

On average, freelancers spend 4 hours a week on upskilling and keeping up with the job market and already have the habit of switching from one client project to another. They were the first to adapt to AI and realize that a job is more than just a bundle of tasks.

As Jensen Huang, CEO of Nvidia, recently said, if someone were to observe him at work, we would conclude that his day consists of tasks like making hundreds of calls and sending emails. AI will replace, augment, and improve these tasks. But it will not take Jensen’s job.

We've featured the best AI chatbot for business.

This article was produced as part of TechRadar Pro Perspectives, our channel to feature the best and brightest minds in the technology industry today.

The views expressed here are those of the author and are not necessarily those of TechRadarPro or Future plc. If you are interested in contributing find out more here: https://www.techradar.com/pro/perspectives-how-to-submit

• Microsoft has blocked Secure Boot updates on some devices
• There have been issues with some devices updating from the 2011 certificate to the latest 2023 certificate
• Some older devices or those not supported by their OEM may be restricted in downloading the latest Secure Boot certificates

Microsoft has blocked some Windows 11 PCs from installing Secure Boot updates due to known issues with certificate updates.

The company is currently rolling out an update for Secure Boot on devices using certificates issued in 2011, which are now expired. The new 2023 certificate is being applied through Windows Update, but issues on devices with faulty firmware have forced Microsoft to halt the rollout.

“Devices in this group are affected by a known issue. To reduce risk, Secure Boot certificate updates are temporarily paused while Microsoft and partners work toward a supported resolution,” Microsoft said.

Secure Boot issues

Secure Boot has long been a device-saving feature when it comes to removing malicious files, as it allows the device to verify and load only authentic software before booting to Windows. However if the device cannot receive certificate updates it can fall victim to threats at the boot-level before Windows is loaded.

Microsoft is currently working with manufacturers to issue a patch that will allow affected devices to install the new Secure Boot 2023 certificate, with HP issuing a BIOS update to allow the installation of the latest certificate.

What this means in practice is that some older devices, or devices that no longer receive updates via their Original Equipment Manufacturer (OEM), will not be able to apply Secure Boot and Boot Manager protections. Microsoft clarified that, “this results in a gradual reduction in long-term security—not an immediate risk or system failure. Continue to follow standard security practices, including staying current with Windows updates.”

So even if your device is blocked from installing the latest Secure Boot certificate, it will continue to work properly, other Windows updates will continue to work, and your Secure Boot version will continue to protect against known vulnerabilities. It’s just future vulnerabilities that users affected by this issue will need to be aware of.

Many users may not be aware of issues until they need to use Secure Boot, so the silver lining in Microsoft’s warning is that now is the perfect time to check if your Secure Boot is working properly.

How to check Secure Boot is up to date

To check if you are using the latest Secure Boot certificate, take the following steps:

• Open the Windows Security app using the search bar
• Navigate to the Device Security dashboard using the menu on the right hand side
• Look at the Secure Boot section, and check for the following messages:

1. "Secure Boot is on"

If you see this message, Secure Boot is likely working properly. However, this does not display your certificates’ current state. Microsoft has been rolling out an update to show if your Secure Boot is running on the latest certificate, so make sure your don't have any pending Windows updates.

2. “Devices in this group are affected by a known issue.”

Devices with this message will likely be able to install the latest certificates once a firmware update has been issued by your OEM. Check your OEM update channel for availability.

3. “Secure Boot is on, but your device does not support the automated Secure Boot certificate update due to hardware or firmware limitations.”

Devices with this message may no longer be supported by your OEM, or the OEM might no longer be able to provide the firmware updates needed. Microsoft recommends checking your OEM’s Secure Boot support page to confirm whether your device is out of support.

Via WindowsLatest

• Windows 11's Copilot app has a new feature in testing
• 'PC insights' provides an easy way to receive clear answers to hardware-based questions about your device and its specs
• While there are some fears over privacy (and bloat), Microsoft has made it clear that Copilot needs to be granted permission to access your system and files

Copilot is getting a new ability to answer questions about your PC's hardware, allowing the AI to tap into the relevant hardware details to do so – and while Microsoft is treading carefully with privacy here, that's unlikely to stop some level of paranoia.

Windows Latest flagged the introduction of 'PC insights' for the Copilot app on Windows 11, which as Microsoft explains, "enables customers to conversationally ask Copilot questions about their Windows PC and receive clear responses based on their device's state without having to dig through system settings."

This is currently an experimental feature, so still in testing, and an optional ability that you must turn on for it to be in play. Windows Latest notes that it's gradually rolling out, but only in the US for now.

You can ask Copilot how much RAM you have, or storage space left, or what your GPU is, and the current level of usage for your processor, and a whole bunch of similar component-related queries. You can ask about elements as diverse as whether you have an antivirus running, or what your laptop's battery health is, diving into mild troubleshooting territory should you wish.

To get its answers, the Copilot app hooks up to Windows APIs to analyze your system, and the AI asks for permission to do this. You can allow it access to your PC's hardware details on a one-time basis for that session only, or you can elect to 'always allow' if you're happy to give Copilot this access on a more permanent basis.

Analysis: fears over hallucinations and bloat

As ever, this is AI, and as Microsoft notes, Copilot "may not always provide complete or accurate information", especially during this testing phase. So, if you do get a chance to try out PC insights, maintain a healthy sense of skepticism with the responses you get.

As Windows Latest makes clear, there's also a certain irony about a Windows 11 user checking up on resource usage, perhaps due to system sluggishness, employing the Copilot AI to run diagnostics when the app itself uses the best part of 1GB of RAM when running in the background and doing nothing.

That doesn't stop this new PC insights feature from being situationally useful, of course. Some of the reaction has come from a place of disdain, though, as you might guess, with comments such as the one from this Redditor: "Oh hey it's like Task Manager except instead of lightweight and authoritative, it's bloated and might be lying to me."

Of course, this is a feature aimed at less well-informed PC owners, not those who can easily understand what's happening in Task Manager at a glance. Criticism around the bloat of the Copilot app is fair enough, mind, and this is because in its most recent incarnation, Microsoft changed things so the app is essentially a standalone spin-off of the Edge browser.

Another worry is that of privacy, and having Copilot 'snoop around' on your machine, but as noted, there are clear requests for permissions, and the new feature is strictly opt-in. You don't ever have to go near PC insights if you don't want to. It's also worth noting that giving the Copilot app access permissions doesn't mean it can read the actual contents of files, but only their sizes (for weighing up questions about storage and the like).

At the moment, this is a purely informative or troubleshooting feature, and in the case of attempted diagnostics, it may point to issues with your PC, but won't resolve them for you. However, it's not difficult to envision where Microsoft might head with this, in terms of getting Copilot to implement fixes for certain issues that the AI flags up. I'm talking simple Settings changes rather than anything in-depth, and this has always been the idea of Copilot (even though it hasn't yet been realized to much of an extent).

When we get AI agents in Windows 11 – and they are coming, make no mistake – this kind of functionality may turn into a full-on troubleshooting agent. The trouble (pun not intended) with that being that the mistakes and hallucinations that AI can make could be considerably more aggravating in this kind of scenario.

There used to be a time when you’d buy an Apple app by paying for it once and never again. Then along came the idea of app subscriptions, and slowly but surely, Apple has started selling more and more of its software products through recurring payments. From iCloud+ to Apple TV, there are many ways to spend your cash on Apple’s wares on an ongoing basis.

Today, Apple flogs bundles of its own apps to its users, with Apple One in particular offering a cluster of apps for a single monthly price. And Apple hasn’t stopped there, with the Creator Studio package being introduced a few months ago for users of its artistic apps.

But while these app groupings ostensibly give you a way to get the software you need without paying multiple subscription fees, I’m starting to worry about how they restrict your choices and force you down avenues you might not want to travel.

Not only can they be expensive, but they’re also inflexible, giving you very little ability to customize them to your needs. If subscriptions are the future of app monetization, then Apple needs to do a whole lot better.

Enter the Creator Studio

Apple likes to portray itself as the company of creatives — think about its “Here’s to the crazy ones” or “I’m a Mac” commercials of days gone by. And the firm already sells apps made specifically for this demographic, including Logic Pro, Final Cut Pro, Pixelmator Pro, and more. So it makes sense that Apple would want to offer a software suite that grants interested users access for a single fee, instead of requiring several payments, whether one-off or ongoing.

That’s what you get with the Creator Studio, which is priced at $12.99 / £12.99 / AU$19.99 a month or $129 / £129 / AU$199 a year. In return for your cash, you get a pass for the following apps:

• Final Cut Pro
• Logic Pro
• Pixelmator Pro
• Keynote
• Pages
• Numbers
• Motion
• Compressor
• MainStage

And yes, some of these apps — like Keynote, Pages, and Numbers — are already free, but pay the Creator Studio entry fee and you get “a library of high-quality, royalty-free photos and graphics, and powerful intelligence features” in addition to what comes with the toll-free editions, Apple says.

Considering Logic Pro alone costs $199.99 if you want to buy it outright (rather than pay an ongoing subscription), this package might seem like a rather good deal. But it’s not the price that I’m interested in — it’s the way app bundles like Creator Studio work.

Right now, there are two main app collections available from Apple: Creator Studio and Apple One. While Creator Studio contains the aforementioned creative apps, Apple One starts at $19.95 / £18.95 / AU$24.95 a month and comes with 50GB of iCloud+ storage, plus memberships to Apple TV, Apple Music, and Apple Arcade.

There are two additional tiers, with the most expensive costing $37.95 / £36.95 / AU$49.95 a month and throwing in 2TB of iCloud+ storage, Apple Fitness+, and Apple News+, in addition to what’s already in the entry-level option.

But what if you want to mix and match your apps and services from Apple? What if you want, say, Logic Pro and Final Cut Pro from Creator Studio, plus Apple Music, Apple Fitness+, and Apple News+? Well, you’re going to have to either take out the top Apple One bundle and separately buy those creative apps outright, or you’ve got to pay for memberships of two distinct ongoing subscriptions.

In this instance, combining Creator Studio with the most expensive edition of Apple One means paying $50.94 a month when all you want are five apps. The rest of the content you pay for is superfluous. Throw in AppleCare+, and you’re forking out even more.

Apple gives practically no flexibility or customization with its app collections. You can’t mix and match apps and services, and if there are some you want from both subscriptions, you’ve got to pay for both. That makes it an incredibly expensive way to go.

But with Apple seemingly getting evermore enthusiastic about subscriptions — and the revenue they bring in — don’t be surprised if this state of affairs continues.

Is there a better way?

You might feel that, since Creator Studio is relatively new, Apple is still working out the logistics of how to sell the apps and services it contains. But Apple is no stranger to subscriptions. It’s been offering the likes of Apple Music and Apple Arcade for years now, so it knows a thing or two about how to structure subscription offerings.

Is the lack of flexibility here and the absence of any way to pick and choose apps a deliberate policy on Apple’s behalf? We can’t know for sure, but it’s not a good look.

While the Creator Studio might seem like a decent bargain on its face, it becomes considerably less so if you’re not interested in everything it offers and want to pair its apps with those from the Apple One bundle.

Considering how keen Apple seems to be on subscriptions, I’m surprised it doesn’t offer individual memberships for each component app within these packages. For instance, why is there no individual subscription for Pixelmator Pro? Instead, you’ve got to choose between a one-off fee or the full Creator Studio.

If Apple were to go down this route — and it's what I'd do if I were CEO-to-be John Ternus — it would instantly solve the problem of pairing different apps from different collections. You could simply add whichever apps you wanted to your own “build a bundle” package and get exactly what you need for a more acceptable price.

But that kind of process doesn’t seem to be anywhere on the horizon. There are no rumors saying it’s coming, no telltale clues nestled in Apple’s code. Don’t hold your breath for its imminent arrival.

Until we do get something like that, Apple customers are going to continue to feel ripped off by a rigid, inflexible system that is happy to take their cash in return for apps they don’t want or need. It’s a situation that’s in desperate need of a rethink.

It wouldn't be a difficult argument to make that the USA is the television capital of the world, or at least the birthplace of modern TV culture.

From Walter Cronkite to Fox & Friends, I Love Lucy to The Simpsons, The Twilight Zone to The Sopranos; whether it's news, current affairs, comedy or drama, US television has produced some of the most iconic, enduring shows committed to the format.

And that's before you even get to sports. What would the sporting calendar look like without Super Bowl Sunday, the Fall Classic, Indy 500 and WrestleMania? Or, most recently, the 2026 World Cup that the USA is co-hosting.

America's shows and sports are collectively the most popular in the world, which means you can continue to watch them on local channels when you're overseas. But it isn't always the case – college basketball, for example, has limited appeal outside the States. And if you go to watch your usual streaming websites, apps and platforms in a different region, they're unlikely to work due to rights restrictions.

Thankfully, there's a very simple way to get around these blackouts. By using one of the best VPNs, you can watch as much US TV as you wish when traveling outside of its borders. Keep reading to discover how.

• Use NordVPN to watch US TV from overseas now (Save 75% with our exclusive deal)

Why can't I watch US TV overseas?

As we say, the international proliferation of US television and sports means that there's a good chance you'll be able to see the content you want in many other parts of the world.

However, due to licensing and rights agreements, you won't necessarily be able to watch them on the platform to which you've become accustomed (and, more commonly than not, pay for).

Try to watch Love Island USA on Peacock, Welcome to Wrexham on Hulu, or a numbered UFC event on Paramount Plus, for example, and you'll be met with an error message telling you that you can't stream them. The same goes for OTT cable alternatives like YouTube TV, Hulu + Live TV, Fubo and Sling TV – all are region-restricted to the US.

It can even happen with certain content on the likes of Netflix, Disney+ and Amazon Prime Video. They have slightly different libraries depending on which territory you're in. So you can forget watching The Great British Baking Show on Netflix if you're not Stateside.

And even if you find the show or sporting event on a local station, you may end up having to subscribe to stream. That hardly seems fair if you've already paid a premium back home to watch that exact same thing.

How can I watch US TV from overseas?

This is where having a good Netflix VPN or Peacock VPN can pay dividends.

The clever engineers at these VPN companies work constantly to make their tools compatible with the world's biggest streaming platforms, thereby letting you watch them regardless of where in the world you are.

The technology effectively spoofs your IP address so that your laptop, smartphone, tablet or streaming device looks like it's in an altogether different location.

Specifically, in this case, a location back within the USA.

Then, when you load up your app for Disney+, Apple TV, DirecTV or anything else, it will think you're back at home and so won't hit you with that annoying error message.

That's in theory, at least. Only the best streaming VPNs are able to unblock a wide range of these platforms. That's why it's worth making sure you get one that will come to your rescue when you're overseas on business or vacation...

Which is the best VPN for watching US TV when overseas?

Step-by-step guide on using a VPN to watch US TV overseas

Using a VPN to watch US TV when overseas is a doozy...

• Sign up and install a VPN on your device (NordVPN is our top pick)
• Connect to one of its United States servers
• Head to the website or app of the US streaming service you wish to use
• Hit play and enjoy!

We've learnt that it's best to connect to your US VPN server first before loading up the streaming app or website that you're attempting to unblock. We've found that if you do it the other way around (i.e. heading to the streaming platform first) you can, in effect, 'tip off' the streamer, which makes it less likely that the VPN will work its magic properly.

And another tip we have is to use an Incognito window you're streaming within a web browser. Again, it just helps to make sure that none of your cookies or usage history triggers the streaming platform to know that you're currently overseas.

What sports can I watch on US TV from overseas?

As a sports-mad nation, US TV hosts just about every event you can imagine from both home and abroad.

The NFL, NBA, NHL and MLB all go out on an intricate variation of networks and their associated streaming platforms. Anything that would normally be available on cable can also be watched on an OTT cord-cutting service.

Paramount Plus and Netflix are the respective homes of UFC and WWE, while Fox is the existing Nascar provider and has exclusive action from the 2026 FIFA World Cup. Apple TV has upped its game in recent years to win the rights to show MLS soccer and every race of the Formula 1 season.

Extensive coverage of the most recent Summer and Winter Olympic Games was carried by NBC's Peacock platform, as is Tour de France cycling.

In terms of overseas soccer, Peacock shows English Premier League matches, ESPN has La Liga from Spain and Bundesliga from Germany, and CBS carries the UEFA Champions League.

And what about sports that are traditionally less popular in the States? Dedicated platforms like RugbyPass and Willow TV respectively feature global rugby union and cricket action.

What news, dramas and comedy can I watch on US TV from overseas?

While some 24/7 rolling news casts will work wherever you are (e.g. Fox News, MS Now), others will not (e.g. NewsNation) without the use of a VPN.

Both current and prestige dramas can be streamed on an array of streaming platforms. George R. R. Martin's stories are an apposite example of this – Game of Thrones and House of the Dragon can be watched with a subscription to HBO Max. Peacock Originals include Ponies and The Five-Star Weekend, while Disney+ has plenty of Doctor Who.

Thankfully, most of the proprietary shows to land on Netflix and Apple TV can be watched from region to region. So the likes of Squid Game, Stranger Things and Severance can be streamed from pretty much anywhere.

It's a very similar situation in terms of comedies. Friends makes its home on HBO Max, while Seinfeld sits on Netflix and all seasons of Cheers are on Hulu. For something more contemporary, you'd need Paramount Plus for Colin from Accounts, Prime Video for Bait and Apple TV for Ted Lasso.

Just remember, if you're struggling to watch any of these when away from the country, you'll need to invest in a good VPN. They're more affordable than you might imagine.

• Meta reportedly has a new AI glasses prototype
• It is helping the company finalize a super sensing AI model
• Meta execs apparently aren't yet certain what rules the all seeing AI should follow

Meta is reportedly trialling a new super-sensing AI glasses prototype that aim to offer an ultimate level of assistance, for the small price of capturing your every moment.

The basic idea is that, because personal AI assistance gets better the more the AI knows you, by having you smart specs watch every detail of your life, the assistant can then offer more insightful help.

It could, for example, know if your fridge has milk in because it has seen inside it, or remind you not to forget your keys as you go to leave home. It could remember that gift your friend was desperately hoping to get for their birthday, or remind you the name of someone you’ve met before, if you're forgotten and are too embarrassed to ask.

This kind of tool would undeniably be pretty handy, but the extreme cost would be that your glasses would have to be always on — otherwise the AI might not see or hear the crucial information you’ll need later.

However, this incredible level of insight could be very easily abused — so you’d either have to really trust the company capturing it, or just not care if it knows everything about you.

This latest super-sensing leak comes via the Financial Times (behind a paywall) which reports that Meta is currently testing a pair of specs that achieve super-sensing by capturing continuous audio, and snapping photos every few seconds.

The more staccato image feed is likely an effort to conserve battery — according to previous leaks, existing Meta glasses have all the hardware they need to pull off super sensing, but their batteries wouldn’t be able to last long enough for it to be widely useful.

According to the report, Meta is also still trying to work on some other aspects, like whether the recording light LED should be displayed or not. When you’re recording it normally would be — super sense would be capturing people around you who might wish to know if they’re on camera — though sources have told the FT that there are plans for it to not to be displayed.

Considering that Meta just made a big post about privacy, and even updating the specs to disable tricks modders have been using to allow the specs to record without the light coming on, I’d be surprised if it took the no-light approach, but we’ll have to wait and see.

Man vs machine

You might also remember a story we shared in March about Meta contractors who claimed to be able to see images and videos taken with the glasses.

Essentially, in order to use Meta’s AI you agree to allow the company to see information about your AI communications, including videos and photos. This seemingly includes photos and videos you take using hands-free voice controls — which does technically involve Meta AI, as it has to action your request.

If an always-on AI is seeing every detail of your life there’s potentially a lot more scope for you to capture sensitive data that you'd rather not let Meta's contractors (or anyone else frankly) see. Encouragingly, according to people familiar with Meta’s super-sensing technology, there are plans to never store the raw footage and audio — meaning Meta nor the user could access it.

Instead, the system would extract the metadata from the capture, and only that would be uploaded — metadata is data about data, so imagine if instead of showing someone a photo you just described what’s in it and where it is.

There’s still some room for personal data to leak through this system, but with the correct safeguards it would have far fewer privacy implications. The big obvious gap would be the privacy of those around you — people who, unlike you the glasses wearer, might not have consented for Meta to store any kind of data about them.

The FT report adds there are also debates over whether Meta’s AI should be allowed to use this metadata for training purposes, in order for it to keep up with the capabilities of OpenAI, Google, and Anthropic’s models.

While leaks should be taken with a pinch of salt, Meta CEO Mark Zuckerberg has made it clear that he wants to offer these kinds of features on his company’s glasses. Most recently he said in an investor call that he wants smart glasses “to be a personal agent that’s with you all day long, helping you remember things and achieve your goals.”

So the advent of technology like super-sense is most likely a matter of when not if, though with battery hardware constraints persisting (especially if you want slim and light glasses), and with privacy being such a hot-button issue, I expect it might be a while before we see super sense in action.

Whenever it arrives, we'll just have to hope that it’s implemented in the right way, and with appropriate safeguards. Such a tool could potentially offers some incredible accessibility benefits, but if the privacy cost is too great the I don’t see it taking off in the way I’m sure Meta would want it to.

Do you remember the metaverse? If you don’t, don’t worry. In 2026, four years into the AI revolution that’s changing the world forever, you could easily be forgiven for thinking it was a strange fever dream you had back in 2021. You might even have odd memories of seeing a blocky version of Mark Zuckerberg floating about in a Minecraft-inspired hellscape, conducting meetings with people who could walk with no legs, while the real Mark Zuckerberg was looking at the whole thing through VR goggles. At least, that’s my memory of it.

I could be suffering from the Mandela effect, but I distinctly remember something off about the legs. They fixed that in a later version, but that’s my overriding memory — no legs. Oh, and Mark Zuckerberg assuring us that this was the future. He’d spent tens of billions of dollars on it, even changing the company name from Facebook to Meta, just to let us know he was really serious about the metaverse. Even if the legs didn’t work.

The problem was, it looked laughable. While everything in the technology world to do with games and special effects was moving in the direction of hyper-realism, the metaverse was moving in the opposite direction, towards the sort of blocky graphics that small children enjoy. But even that didn’t really answer the most basic question about the metaverse. Why? What possible advantage was there for us all to meet in a VR space where clunky avatars of ourselves could interact… badly?

Then AI happened and Meta abruptly forgot about the metaverse and pivoted towards the mission of putting personal superintelligence in all our hands instead, which sounds as terrifying and dangerous as it actually is, but we are where we are. At least when Zuckerberg was obsessed with the metaverse, we could ignore him. It existed somewhere “over there”, in Meta-land, where we could let him get on with it. Now he’s right up in our business again.

The rise of a bad idea

I’ve always been fascinated by how Zuckerberg got into the metaverse and why he became so obsessed with it. The origins of the metaverse go way back. In March 2014, Facebook bought Oculus, the VR company, for about $2 billion. This was where his passion for VR started. Think of it as the seed, not the full obsession.

By July 2021, Zuckerberg gave a long interview to Casey Newton at The Verge about Facebook becoming a “metaverse company” and described it as an “embodied internet”. Then, on October 28, 2021, his obsession became the company identity. Zuckerberg announced that Facebook the company was becoming Meta at Connect 2021, saying the new company brand would focus on bringing the metaverse to life.

I can see the logic. As a business strategy, that made sense. As a product ordinary people were expected to use, it was much harder to understand. Zuckerberg did not want Meta/Facebook to be trapped inside someone else’s platform again. Facebook had won on social, but on mobile it remained dependent on Apple and Google for distribution, privacy rules, app-store policies and hardware. The metaverse looked like a chance to own the next operating system of social life: hardware, avatars, identity, payments, meetings, gaming, work, commerce — the whole stack. In his 2021 founder’s letter, he framed the metaverse as the “next chapter of the internet” and said Meta would become “metaverse-first, not Facebook-first.”

Facebook had obviously had its problems — it was scandal-ridden. It had let advertisers target vulnerable teenagers, helped fake news spread, and enabled the spread of hate speech linked to atrocities in Myanmar. Perhaps Zuckerberg was looking for a way out of Facebook, and the metaverse offered that.

What I still didn’t understand was why he didn’t see what the rest of us saw — that it looked terrible and offered no real benefit to users. Then I read Sarah Wynn-Williams’ tell-all book about Facebook, Careless People, and it all started to make sense.

To say the book made my jaw hit the floor on several occasions would be an understatement. It’s an absolute page-turner, and your reactions grow from mild amusement to shock, then disbelief, then absolute outrage the further through the book you get. I’m aware of the criticisms of Wynn-Williams: that it is a book written by a disgruntled employee, and that she dodges a lot of personal responsibility for her part in the various misdeeds of the company. However, in another perfect example of the Streisand effect, the fact that Meta obtained a legal order in the United States to prevent her from saying anything negative about the company — at all — made me want to pick it up, and I’m glad I did.

Because now I get it — Zuckerberg seems to have spent years in an environment where too few people were willing to tell him when his ideas weren’t good. According to Wynn-Williams, he was surrounded by sycophants. When he had bad ideas, like the ill-fated Internet.org, he wouldn’t let them go and persisted with them, even when they were obviously going to fail. The people around him enabled him because he was simply too powerful. They even let him win at the board games he liked to play with them at his house or on his jet, and — crucially — he didn’t notice that they were letting him win. I can imagine that in that environment, nobody inside Meta would want to tell Zuckerberg that his metaverse was the equivalent of the emperor’s new clothes, especially if they wouldn’t even risk beating him at Settlers of Catan.

Wynn-Williams only mentions the metaverse in her epilogue. It happened after she was brutally fired from Facebook. Perhaps selfishly, I wish she’d been there for the metaverse period, because I would love to read firsthand accounts of how and why Zuckerberg persisted with such an obviously bad idea.

The fall of the metaverse

Maybe I’m being too harsh on Zuckerberg. The metaverse graphics did get better over time and Apple ventured slightly into the same territory with its Apple Vision Pro, even after the metaverse had turned into a smoldering wasteland. The fact is, people don’t enjoy wearing VR goggles for extended periods of time, and for normal people, VR lacks that one killer app. There doesn’t seem to be anything you can do in a VR space that you can’t do elsewhere much more easily.

The metaverse didn’t really die with a bang, but with a whimper. It faded through layoffs, spending cuts and the AI pivot. If I had to put a date on it, I’d say early 2023 was when Meta’s narrative moved on. In February and March 2023, Zuckerberg started talking about Meta’s “year of efficiency” and announced huge layoffs and cost-cutting. OpenAI had launched ChatGPT in November 2022, and by early 2023, generative AI had swallowed the oxygen that ideas like the metaverse need to survive. Every tech company was talking about AI now, not virtual offices and avatar legs.

The metaverse was over. We all forgot about it and moved on.

I’m glad I read Wynn-Williams’ book, because now I can understand how Facebook let the metaverse happen. And if there’s one thing I learned from reading it, it’s that money and power can bring you a lot of things, but common sense requires neither.

A study has revealed that approximately 40% of UK employees receive ‘work slop’ caused by low-quality, AI-generated content. As a result, it is estimated that each instance of work slop takes up to 3.5 hours each month to correct, resulting in millions of pounds in lost productivity.

It has been reported that as of mid-2026, approximately 78% of global businesses are using AI in at least one business function.

Despite this, studies have revealed that 62% to over 80% of workers lack confidence or training in AI, with many reporting they don’t have the skills to use the tool in their daily tasks, which can lead to errors and work slop.

What is AI work slop?

Many workers are using AI to make daily tasks such as generating reports, emails, or code more efficient. However, if used incorrectly, workers can be faced with low-quality, AI-generated content that may look like a high standard at first glance, but after closer inspection, is inaccurate and requires hours of manual checking and corrections.

To minimize AI work slop, there are several steps businesses can take:

Make training a key objective: It’s inevitable that without appropriate training and guidance using AI software, errors will occur. Business owners should ensure that employees are trained in more efficient prompting and understanding AI limitations to reduce work slop. If training is implemented during the employee onboarding stage or as early in the project as possible, the chances of work slop will be significantly reduced.

Start by brainstorming: AI must be used to enhance work, not to create it on its own. Before completing tasks that incorporate AI, it’s important to brainstorm how it can be used and the objectives you want to achieve. Where possible, create the bulk of content from scratch and use AI to support and enhance the end goal.

Introduce a review process: AI-generated work should never be presented to senior stakeholders or clients unless checked by an experienced team member. Business owners should ensure that all AI-generated tasks are reviewed and fact-checked before being presented as the final version.

An effective method is to create a mindset that AI-generated work has been created by an intern who has little knowledge of the industry. Reviewing work using this method will help to spot errors faster.

Encourage transparency: To reduce AI work slop, business owners should encourage workers to use AI to support and enhance tasks, not to complete them altogether. It’s also important for workers to be clear about how AI has supported the tasks they’re working on to make the reviewing process more efficient.

Encourage team feedback: One of the main ways to prevent continued work slop is to make colleagues aware that it has been spotted. More often than not, people will correct mistakes themselves, rather than be transparent and provide constructive feedback to their peers. Raising work slop in the early stages of the project will save time and prevent the issue from recurring.

Create manageable workloads: Workers who are unable to manage their workloads will often complete tasks quickly using AI, which can result in work slop. To prevent this, promote quality over quantity throughout the business and ensure that unmanageable workloads don’t hinder this ethos.

The key to AI success

For businesses using AI, the software should be used to elevate work and make processes more efficient, not cause delays and errors. If your business is looking to adopt AI, it’s important to be transparent with employees about when to use it and when to avoid it.

Employees should have a strong mindset towards AI and have quality and value in mind at all times when understanding the outcomes they’d like to achieve. When planning tasks, we should ask ourselves: Will AI add value and additional insights to the project, and how will it move your goals forward?

Once these points have been considered, it’s important to understand what AI can’t assist with, such as human perspective and strategic creativity. If you decide to use AI for the task you’re looking to complete, which is weighted towards human-powered thought and creativity, this is when work slop can be encountered, resulting in wasted time and resources.

To summarize, AI should be used to generate valuable content and results. Asking yourself the questions mentioned above before you turn to the software will help to generate the best possible outcomes.

An effective mental model that helps to prevent work slop

Before employees can use AI responsibly, they need to understand what it actually does. In "AI for Startup Leaders," we present a framework called A-R-C that captures the three core capabilities of AI:

• Agency: AI can work with tools, run code, and complete tasks on an employee’s behalf. For example, it can query data from your CRM, draft documents, or update a spreadsheet.
• Reasoning: AI can plan ahead and think through problems step-by-step. The latest models can break down complex challenges, consider multiple approaches, and work through logic chains.
• Context: AI can base its answers on the information you provide in each conversation. It can understand natural language, identify objects in photos, and parse data in spreadsheets.

The A-R-C model is crucial, as many workers use AI as an encyclopedia and to state facts, which isn’t what the software was created for. When people rely on AI for facts, work slop can occur. To avoid this, workers should be educated on the main purpose of AI, reasoning over their business context and taking action.

When employees understand A-R-C, they approach AI outputs differently. They recognize that AI is best used to reason from their business context rather than look up answers. Instead, they use AI to speed up how they collaborate with other teams and improve efficiency.

We've featured the best AI website builder.

This article was produced as part of TechRadar Pro Perspectives, our channel to feature the best and brightest minds in the technology industry today.

The views expressed here are those of the author and are not necessarily those of TechRadarPro or Future plc. If you are interested in contributing find out more here: https://www.techradar.com/pro/perspectives-how-to-submit

For the past year, one idea has dominated conversations about enterprise technology: AI agents will replace Software as a Service (SaaS). It’s a compelling case on paper. If AI can reason across tools, write code, execute workflows, and interact with systems through APIs, then traditional SaaS applications start to look like unnecessary middlemen.

Seats become less relevant. User interfaces matter less. Software becomes cheaper to build. Custom internal tools become easier to create.

In that reality, much of SaaS gets pushed down into infrastructure, while value shifts to models, agents, and orchestration layers. This reaction is being driven by several parallel shifts.

Software is becoming cheaper and faster to build, AI agents are becoming better at navigating tools and executing work across systems, and the economics of agentic execution are bringing new attention to cost and latency, or time to serve.

The market is not wrong to think this way. AI agents will change software economics.

They will push platforms to become API-first, deeply connected, and capable of supporting autonomous activity at scale. They will challenge seat-based pricing and accelerate the move towards consumption and outcome-based models. They will expose weak products and reward systems that are trusted, extensible, and embedded in real operations.

This is the next phase of SaaS, rather than the end of it.

From replacement to expansion

The assumption behind the “end of SaaS” narrative is that software exists primarily as an interface. If agents can bypass that interface, the application becomes redundant. That logic holds for some categories of software, but it does not apply uniformly. Enterprise software has never been just a presentation layer.

Enterprise platforms derive their value from managing structured data, enforcing permissions, executing workflows, and maintaining the audit trails organizations rely on. Agentic AI does not remove that requirement; in many cases, it sharpens it. What AI changes is not whether these capabilities are needed, but which types of software are most exposed as agents become more capable.

Products whose value is largely defined by navigation, basic interaction, or shallow workflows are more vulnerable when agents can reason directly over APIs and complete tasks without human mediation. More configurable, deeply integrated platforms behave differently.

Solutions that act as systems of record, coordinate workflows across services, apply policy, manage state, and provide security, auditability, and evidencing are not easily displaced. Their role is not simply to present functionality, but to ensure work is executed consistently, safely, and at scale.

In that context, AI agents are not replacing platforms so much as being embedded within them, extending how work is initiated, coordinated, and completed. That direction is already visible in the data.

Gartner forecasts that by 2030, 85% of enterprise agentic AI investments will be bundled into existing SaaS and cloud renewals rather than delivered through net-new contracts, up from 55% in 2025.

SaaS as the execution and control layer

As agents become more capable, they shift where value sits and how platforms compete.

For years, software differentiation has been driven by features and user experience. In an agent-driven world, those differences begin to matter less. Agents can move across systems, access functions directly, and orchestrate tasks programmatically.

What matters instead is which platforms can actually complete work. This means coordinating processes across systems, applying the right type of automation at each step, and ensuring outcomes are reliable, traceable, and compliant.

In this model, SaaS does not disappear. It becomes the execution and control layer for enterprise AI. As this shift plays out, value and margin move away from individual features and towards the platforms that control execution and data access. Agents themselves are unlikely to be a sustainable point of differentiation. As capabilities converge, the focus moves towards control.

That’s because enterprise processes still need to be predictable, auditable, and, in many cases, reversible. Regulatory frameworks are reinforcing this, with increasing expectations around explainability and oversight. This places new importance on the platforms that sit at the center of operations.

Organizations that can orchestrate work across systems, apply automation selectively, and produce a clear evidence trail will capture more value over time. Those that cannot risk becoming passive data stores, rather than active systems of execution.

A multiplier effect on demand

There is another misconception shaping the conversation that AI will reduce the amount of work organizations need to do and, with it, the reliance on software.

In practice, the opposite is happening. As the cost and effort required to execute processes falls, more work becomes viable. More cases are identified, more customer needs are addressed, and more processes are triggered automatically.

We are already seeing this in service environments, where AI-driven detection and automation increase the volume of actionable work entering the system.

In the context of public services, for example, within a case management environment, a missed bin, pothole, or housing repair issue can be identified automatically, raised as a case instantly, and routed into the right workflow without waiting for a citizen or staff member to log it manually. In that model, the value sits not in who created the case, but in how effectively the platform absorbs, processes and resolves it.

This creates a multiplier effect. AI does not just reduce effort; it expands what organizations can do. AI expands demand, and SaaS platforms are where that demand is fulfilled.

The question is no longer how many users a platform supports. It is how much work it can handle, complete, and evidence effectively.

What this means for SaaS platforms

This shift will not benefit every platform equally.

As software moves from presenting information to completing work, value becomes tied to what a system can actually deliver, not just how it is accessed. This will also shape commercial models, with more platforms likely to combine traditional subscriptions with pricing based on consumption, throughput, or completed outcomes.

Organizations still need oversight, approvals, and ways to manage exceptions, but the focus is shifting towards how effectively platforms can absorb, process, and complete work at scale.

That change is already exposing the gap between strong and weak software, placing greater scrutiny on systems that offer limited workflow depth or rely heavily on manual effort. In contrast, platforms that are deeply embedded in operations, with strong data, logic, and execution capabilities, will become more valuable.

As agents become consumers of APIs, software needs to be connected and capable of supporting autonomous activity at scale. Those that are not will struggle to keep up.

The next phase of SaaS

Ultimately, what we are seeing is not the end of SaaS, but an evolution. SaaS is being reshaped, not just as software for human interaction, but as software designed for both humans and machines operating across systems.

A shift from systems that present information to systems that complete work. From user-driven interaction to orchestrated execution, and feature competition to control over how processes run.

This is not theoretical. It is already happening. Low-code platforms and productivity tools are being augmented by AI and agentic workflows, changing how applications are built, adapted, and used day to day.

But this shift also raises the bar for how organizations adopt AI. Those that simply layer agents onto existing processes risk creating more complexity rather than less. Introducing automation without understanding how work really flows can make outcomes harder to predict and harder to evidence, particularly in regulated environments.

The organizations that succeed will start with a real operational problem and introduce AI selectively. Where existing workflows provide sufficient structure and control, value can be delivered quickly without redesign. Where AI exposes friction or inefficiency, that insight can then inform targeted process improvement.

AI agents amplify the platforms beneath them; they do not replace them.

The result is not less software, but different software, running on platforms built to execute work, absorb demand, and stay in control. More automation, more cases handled, more throughput, more outcomes achieved through systems that reliably complete more work at scale.

We've featured the best vibe coding.

This article was produced as part of TechRadar Pro Perspectives, our channel to feature the best and brightest minds in the technology industry today.

The views expressed here are those of the author and are not necessarily those of TechRadarPro or Future plc. If you are interested in contributing find out more here: https://www.techradar.com/pro/perspectives-how-to-submit

• A new update is coming to Meta's smart glasses
• It should stop modders from disabling the light that lets people know you're using the camera
• Meta is also targeting modders on and off its platform

Last month we shared details of reports that Meta glasses were being modded to bypass privacy protections and turn them into secret spy glasses. Now Meta has revealed it will update the glasses’ software to detect whether its light has been tampered with (or destroyed) to prevent recording.

Whenever you take a photo or video with Meta glasses, a white light appears on the front of the glasses to signal to people around you that you’re filming.

For creeps looking to be more secretive with their recordings, this light is a hindrance, but Meta has imposed more basic tamper-proof features since launch. That is, any attempt to use the camera while the light is blocked — such as by a piece of tape — wouldn’t be allowed. The trouble is, modders have found ways to open up the glasses and disable or damage the light and its mechanisms that prevent it from showing, without getting flagged by the system — meaning you can use the camera as you normally would, but without anyone else knowing.

This is (to put it bluntly) not good, and when the reports came out, a Meta spokesperson told me that the company was looking into ways to disable this workaround.

They also explained that Meta is working hard to stamp out advertisements for these kinds of services — some of which appeared on its own Facebook Marketplace platform — with its latest announcement explaining this means banning accounts, taking down listings, and taking legal action against people or businesses that tamper with its tech.

Beyond the detail of updating software to prevent tampering, the whole article from Meta is focused on privacy, and crucially how Meta keeps you and others safe.

Thanks to people misusing its tech, the wider notoriety these gadgets are getting again, and reports of contractors seeing recorded images and videos that Meta glasses users might not have fully realized they could see, Meta and smart glasses makers have been facing major privacy questions.

With this article, Meta seems to answer many of them, though we’ll have to wait and see if it can convince users — or if they might be tempted by the rise of cameraless smart glasses.

The only way is cameraless?

Even Realities, another smart glasses firm, recently announced it has raised $150 million in investment at a $1 billion valuation — not too shabby for a company that only launched its first XR gadget in 2024.

Instead of Meta’s camera-first approach, Even Realities went for a display-first approach. Information is shown visually as green text and basic diagrams to provide features such as AI advice, navigational directions, or a speech appearing on a virtual prompter. Its specs also lack speakers, though that’s not true for all glasses of their kind.

Losing out on the camera is, of course, a major privacy win for some, as there’s no possible way for the glasses to see something they shouldn’t or be used to spy.

The thing is, I think these kinds of glasses are pretty terrible. Having tested a few at home, the ones without a camera just aren’t worth wearing.

Smart glasses, despite being increasingly popular, are in their infancy, which in the tech world means you can expect to pay a fair sum for relatively limited features — that’s the price of being an early adopter. That’s especially true, I’ve found, for these XR glasses specifically, because while they can offer several tools like navigation, on-screen translations, a prompter, and notification pop-ups, their usefulness is pretty limited.

How often do you need a prompter? Or one-way translation tools? In the case of the latter, because these kinds of glasses often have you rely on your phone to set up the translation feature or access other features, you might as well just turn to something like Google Translate — which has conversation modes so that two people can talk and see translations through a single device.

The software I’ve found for these types of specs can also be pretty terrible with sluggishness, inaccuracies, and crashes — and if I, as someone who tests smart glasses professionally, have trouble, I can’t imagine what less techy people must think.

Additionally, the green text can be hard to make out on a bright day if you’re outside, making on-screen directions difficult to see.

While admittedly more limited feature-wise, the Meta glasses and smart specs like them feel like way better value for money. From Meta specifically, the software is generally very reliable, and more broadly, the ability to snap a photo whenever — either to capture a moment or to provide context for an AI’s response — comes in handy so often.

Even if it isn’t as good as my phone camera, the ability to record a memory, hands-free and without being taken out of the moment, is so utterly delightful.

Yes, there are privacy challenges which need to be hashed out more formally, as even without the ability to record privately there is still plenty of room for creeps to harass people with this kind of gadget, but if you want a pair of smart glasses right now there simply isn’t a better option.

“Tokenmaxxing” – the idea that AI coding success comes down to using as many tokens as possible – is an appealing metric.

Tokens are the fundamental unit that AI coding tools use to read, write, and reason. So on the surface, more tokens should mean more output, more productivity, and more impact.

But when we analyzed 12,000 developers across 200 companies, the data revealed that while more tokens do correlate with more output, they come at a significantly higher price per unit.

Some organizations are pushing software engineers to use as many tokens as possible, using leaderboards to promote the biggest AI users. But that’s not a sustainable strategy. CFOs are starting to push back on uncontrolled AI spending and asking coders to show receipts.

Leaders may be willing to spend money to move fast, but they can’t do it without proving their engineering teams are having an impact.

The best approach to “toxenmaxxing” isn’t to blindly push for AI adoption. Instead, the best path forward for companies is to push AI coding adoption more broadly, moving more engineers into the middle of the curve while avoiding both underuse and expensive overconsumption.

Why ‘tokenmaxxing’ doesn’t scale

We found that the top 10% of Claude Code users consumed about 10 times as many AI tokens as the median developer but produced only about twice the output. In other words, increasing token consumption does increase output, but not proportionally.

The research also shows a small but growing group of power users dominating total token consumption. At the 90th percentile, users are burning around 225M tokens per week, about 3x what they were using six months ago, and about 7x the median.

Many engineering leaders are now looking at their highest adopters and trying to figure out how to get the rest of the organization to the same level. That approach is misguided. With the cost per merged PR increasing from $0.28 at the lowest adoption tier to $89.32 at the highest, scaling extreme token usage simply cannot drive value.

Instead, engineering leaders should focus on smoothing the curve. Broad, moderate token consumption is far more cost effective than having a small group of power users at one end of the spectrum and everyone else lagging behind. When most of the organization is operating in the middle of the curve, AI becomes a durable advantage: enough to drive real productivity gains but not so much that engineering teams burn money chasing marginal output.

Maximize impact, not token consumption

The organizations that burn through the most tokens aren't necessarily getting the furthest with AI. When token consumption is high, most are spent on automating manual tasks with tools like Claude, Copilot or Cursor. Developers essentially have a better tool to do the same kind of work as they did before.

To really drive impact with AI, engineering organizations need to move towards new, truly agentic modes of working. However, agentic systems require major investments in IT infrastructure, including context engineering, orchestration, and sandboxed environments. Until organizations address these issues, the productivity gains will remain blocked by an "agentic barrier" that no amount of tokens can overcome.

How established enterprises can follow the AI-native lead

Conversations around AI and software development focus on coding, but writing code is just one part of an engineer's role. Taking a product to market also involves roadmap work, deployment, go-to-market enablement, and more. If engineers are spending tons of tokens on writing code as fast as possible, everything else needs to catch up.

Changing roadmap cadence and accelerating sales enablement requires major cultural shifts that many organizations aren't prepared for. As a result, extra cadences are often poured into the backlog or other things that may deliver value down the line but won't move the revenue needle in the short term. Teams can consume millions of tokens every week but have little to show for it by the end of the quarter.

AI-native companies are more likely to see an immediate return on their AI investments. While established enterprises may not be able to start from scratch, adopting AI-native principles can help remove bottlenecks and turn token spend into measurable business returns faster. By designing workflows with automation in mind, they can continue to accelerate coding without creating technical debt.

"Tokenmaxxing" is having a moment, but engineering leaders need to move beyond token count and start finding ways to prove value. By measuring how AI impacts delivery, quality, and productivity across the software delivery life cycle, leaders can demonstrate ROI and make sure every token counts.

We've featured the best no-code platforms.

This article was produced as part of TechRadar Pro Perspectives, our channel to feature the best and brightest minds in the technology industry today.

The views expressed here are those of the author and are not necessarily those of TechRadarPro or Future plc. If you are interested in contributing find out more here: https://www.techradar.com/pro/perspectives-how-to-submit

• Microsoft's new Windows 11 recovery method is available for Windows Insiders
• Cloud Rebuild allows users to reinstall Windows 11 and necessary drivers via the cloud, without a USB drive
• The feature should begin rolling out to users on stable Windows 11 builds

Microsoft is continually making adjustments to its Windows 11 operating system through patches that address user pain points, and, fortunately, its latest move is certainly welcome.

As reported by Windows Central, Microsoft has implemented a new recovery method for Windows 11, known as Cloud Rebuild, which is available to Windows Insider users. Cloud Rebuild allows users to reinstall the operating system and drivers from the cloud without the need for a USB drive.

It's a major step in the right direction for users who need to reset their PCs, whether due to data corruption, malware, or simply wanting to start anew, especially since not all users have immediate access to another device to download a Windows image or a USB drive to install it on.

Unlike the Windows Recovery Environment's 'Reset this PC' option, Cloud Rebuild doesn't allow you to keep personal files, but Microsoft says Cloud Rebuild reinstalls Windows with the appropriate drivers and "without depending on the integrity of the installed operating system".

Notably, Cloud Rebuild can still work when users can't boot into Windows 11, making life easier in a dire situation where the operating system is dysfunctional.

Perhaps the biggest advantage of Cloud Rebuild is for the acquisition of drivers, saving users plenty of time from manual downloads. This comes in handy for Windows 11 handheld gaming PCs, where reinstalling drivers can be slightly complicated, especially without the necessary peripherals being on hand for quick and easy navigation.

Cloud Rebuild isn't available to all Windows users yet, but the gradual rollout phase shouldn't be too far off. Fortunately, it's not a feature that most users urgently require (at least, I hope), so the wait for its arrival in stable Windows 11 updates shouldn't be frustrating.

How many unique passwords do you use to secure every single one of your hundreds of online accounts? If your answer isn't, "I have a strong unique password for every account," then you're among the millions of people at risk of having multiple accounts stolen from just a single password leak. All it takes is the right combination.

But that doesn't have to be the case. Using a password manager to generate and store your credentials is a secure and convenient way to prevent your accounts from being hacked. The best part is, Keeper has cut prices on its personal, family, and business starter plans by up to 50%.

Keeper's password manager offers secured password storage vaults, a password generator that easily complies with password strength requirements, and includes convenient features for privately sharing regularly used passwords - like the one for the Wi-Fi.

Why we recommend Keeper

Our Keeper review found the password manager to be an excellent addition to anyone wanting to improve their personal or household security.

The service uses zero knowledge architecture and device level encryption to keep the contents of your personal vault hidden from private eyes.

Rather than having to enter your master password each time you need to fill in your credentials, Keeper uses biometric security to access your vault. This uses the built in facial scan or fingerprint tech on your device. Keeper will also recognize when you are trying to log in to a website or app, and allow you to autofill the exact credentials you need.

Four households, the family plan includes five private vaults, allowing you to quickly share Wi-Fi or streaming passwords using Keeper's shared vaults.

These discounts apply to the first year only, so the value is highest for new customers or anyone switching from a monthly plan. After that, pricing goes back to standard rates.

MacPaw has spent years building a reputation as one of the most design-conscious developers in the Mac ecosystem. Its flagship product, CleanMyMac, has long included a malware removal module powered by Moonlock's engine. In October 2025, the Kyiv-based company spun that security technology into a standalone product: Moonlock, a full-featured antivirus app that goes well beyond a simple scanner.

Rather than leading with threat counts and detection percentages, Moonlock frames itself as security software that treats users like adults, explaining what malware is, why it matters, and what to do next, instead of firing off opaque alerts. The marketing centers on a 'care, not scare' approach, essentially promising to educate you rather than just bombarding you with red-text alerts.

While many live in the mythical belief that Macs are immune to viruses, MacPaw's own research reports that 66 percent of Mac users encountered at least one cyber threat last year, with a 67% increase in registered macOS backdoor variants in 2025. The research shows a key message: macOS is not immune, are users are being targeted more frequently than ever.

Plans and pricing

Moonlock starts at $54 per year for a single Mac, with licenses available for 2, 5, or more than 10 devices per subscription. Monthly billing and one-time lifetime license options are also available for those who prefer not to commit to an annual cycle.

Discounts of up to 67 percent are advertised on multi-year plans, which is worth exploring if you intend to stick with the product long term.

New users get a seven-day free trial, though a credit card is required to start. That is a common enough practice, but it does mean you will need to remember to cancel if the product does not suit you. To soften the blow of that annual fee, Moonlock offers a 30-day money-back guarantee, which is a considerably more generous safety net than the case-by-case refund process offered by some competitors.

Current Setapp subscribers get access to Moonlock at no additional charge, which may be the most compelling value proposition for those already in MacPaw's subscription ecosystem. At $54 per year for a single device, standalone pricing lands considerably higher than ClamXAV's three-Mac Home plan at $29.95, a gap worth weighing if budget is a primary concern.

Features

Moonlock is organized into six sections: Home, Malware Scanner, VPN, Network Inspector, System Protection, and Security Advisor. That framework reflects a deliberate decision to bundle a security suite rather than deliver a focused antivirus, giving the product a notably broader footprint than Mac-only rivals like ClamXAV.

Real-time protection runs continuously in the background, monitoring file activity, app behavior, and Mail attachments even when the main application window is closed. The Malware Scanner supports on-demand and scheduled scans, with built-in quarantine and removal tools. Detected threats are accompanied by plain-language explanations rather than raw file paths and specialized terms.

The bundled VPN is a simplified version of MacPaw's ClearVPN, covering around 60 server locations across more than 45 countries. Independent testing found no DNS or WebRTC leaks, and MacPaw maintains a no-logging policy. Speed retention is strong, holding around 82 percent of baseline download speeds on transatlantic connections and up to 96 percent on closer servers.

Network Inspector adds a country-level connection blocker, permitting users to block outbound traffic to specific regions. System Protection audits macOS's own built-in security settings and walks you through any gaps. Finally, Security Advisor provides a checklist for basic digital hygiene, including practical guidance on habits such as two-factor authentication and app permissions. AI assists with malware classification on the backend, helping the team update threat databases before new strains reach your device.

Privacy and Security

From a top level perspective, Moonlock was tested by the third-party laboratory AV-Test in September 2025 and earned it's AV-Test certification. It scored a 5.5/6 in Protection, 4.5/6 in Performance, and a full 6/6 for Usability (which I'll dive into in the next section).

As for the credibility of the underlying research arm, Moonlock Lab has made several notable contributions to the antivirus landscape, being the first to identify PyStealer on VirusTotal, and the lab has also been cited by the SANS Institute for discovering new variants of the Atomic macOS infostealer.

Regarding privacy, the VPN operates under a strict zero-logs policy, and all data is processed locally. MacPaw publishes a Trust Center at security.macpaw.com describing its data-handling practices, certifications, and security standards, which is a nice change in transparency from many other antivirus providers.

The one caveat worth noting is that Moonlock is a recent standalone launch. While the underlying engine has been in use in CleanMyMac for some time, the app itself has a limited history as an independently tested product. But it is worth noting that in the time since the last time AV-Test handled Moonlock, MacPaw have likely taken steps to improve protection and performance.

Interface and in use

The interface is highly polished, modern, and immediately legible, with a two-panel home dashboard that separates tasks on the left from status information on the right. Everything is where you would expect it to be, and the visual hierarchy makes it easy to tell at a glance whether your Mac is protected.

Instead of a generic 'Threat Resolved' notification, Moonlock tells you what was found, why it poses a risk, and what your options are. I found I was the one to make the final call on whether to remove a flagged item, which sidesteps the infuriating experience of automated deletion that occasionally catches legitimate software.

The system requirements make it suitable for older devices too, requiring macOS 13 or later and 515MB of disk space. The app runs quickly and, in day-to-day use, does not noticeably drag on performance. Installation requires a MacPaw account, which adds a step that competitors like ClamXAV skip entirely for home users, but the tradeoff is a unified login for managing licenses and accessing support.

Ultimately, Moonlock is a great option for those looking for an accessible and easily navigable Mac antivirus that doesn't bombard you with any overly-technical language, and performs as though you are the one in control.

Support

Moonlock support runs on MacPaw's established infrastructure, with a dedicated knowledge base that covers installation, configuration, and troubleshooting, and those with questions can submit immediate inquiries through the support portal. In-app feedback is also available via the Help menu.

As with many Mac-focused security products, live chat or phone support does not appear to be offered as a standard option. For most home users, the knowledge base and email channel will be sufficient. Teams with more complicated environments should verify support response times before committing, particularly given that Moonlock is a relatively new standalone product and the support documentation is still maturing.

The competition

ClamXAV is the most direct rival in the Mac-exclusive antivirus space. At $29.95 per year for three devices, it is considerably cheaper than Moonlock's $54 single-device starting price, and it also holds a perfect AV-Test score compared to Moonlock's test results. It does not include a VPN, network inspection, or the polished onboarding experience Moonlock offers, but for those who want focused antivirus protection at a lower cost, it is a strong option.

Intego Mac Internet Security X9 sits at a comparable price point and includes a network monitor, with a longer track record in independent third-party testing. Bitdefender Total Security and Norton AntiVirus Plus both offer wider platform coverage and larger feature sets, making them better fits for households with mixed Windows and Mac devices.

Those who are already subscribed to CleanMyMac should also note that its built-in malware-scanning module, powered by the same Moonlock engine, continues to function independently. Therefore the question is whether the full Moonlock standalone app adds enough to justify an additional subscription or an upgrade in spending.

Final verdict

Moonlock is one of the most carefully designed security apps I've encountered in the Mac ecosystem. Its interface is excellent, its feature set is broader than most Mac-specific alternatives, and the research team behind it is doing genuinely credible original work. The 30-day money-back guarantee is also a nice addition, despite the need to enter your payment details first.

At $54 per year for a single Mac, it costs nearly twice as much as ClamXAV's three-device plan. The added value of the bundled VPN and Network Inspector goes some way toward justifying that gap, but those who already have a VPN solution elsewhere may not find the extras compelling enough. Setapp subscribers, on the other hand, get all of this for free as part of a subscription they likely already value.

For long-standing CleanMyMac users who already benefit from the embedded Moonlock engine, the standalone app offers greater depth, visibility, and control, but it's not a replacement for anything missing. It is a fuller version of the protection they have already been relying on, now with a VPN, richer reporting, and a proper home for the security features that were previously contained within a Mac cleaning utility.

For Mac users who want a single subscription that covers antivirus, VPN, network monitoring, and system security guidance, Moonlock makes a strong argument. Just go in aware of what you are paying for relative to the alternatives.

You might also be interested in our report on the best Mac apps of the year.

Walk deep into Walt Disney Imagineering’s sprawling, surprisingly unassuming Glendale campus, and you’ll eventually find Studio C.

It’s where Imagineering mixes the audio for its attractions, but when I recently stepped inside, I quickly realized it was much more than a recording studio. The room is lined with dozens upon dozens of speakers designed to recreate the sound field of Disney attractions, letting Imagineers hear a ride long before guests ever do.

During my visit, I watched a live mix session for Zootopia: Hot Pursuit, Shanghai Disneyland’s trackless dark ride, and the experience was almost uncanny. Sound moved seamlessly around the room, tracking the ride vehicle as though I were actually inside the attraction.

That same room became the proving ground for one of Imagineering’s most unusual engineering challenges yet: figuring out how to mix the audio for Soarin’ Across America.

The reimagined version of Disney’s iconic flight simulator debuted this summer at both EPCOT at Walt Disney World in Orlando, Florida, and Disney California Adventure in Anaheim, California, timed for America’s Semiquincentennial celebration. Disney’s Unscripted behind-the-scenes video confirmed that Imagineers used the Apple Vision Pro during production of Soarin’ Across America, but it left one obvious question unanswered: why? That’s what I wanted to find out.

So I spoke with Greg Lhotka, Sr. Manager, Audio Media Design at Walt Disney Imagineering, who walked me through the surprisingly complicated process behind mixing one of Disney’s most technically demanding attractions.

“For Soarin’ Across America, we utilized Studio C not only as a production facility, but as a proving ground for an entirely new field-mixing approach developed specifically for this film. On the sound design side, we partnered with Skywalker Sound, and together we used Studio C to prototype and validate a custom speaker configuration that closely mirrors the unique audio architecture of the Soarin’ theaters.”

If you’ve ridden Soarin’, you already know why that matters. Riders are lifted into the air in three stacked rows facing a massive curved projection screen, while sound moves not only around the theater but above and below you as well. As Lhotka explained, the attraction uses a four-tiered speaker system — upper, mid-level, lower, and a dedicated floor array — creating an audio environment that’s far more complex than a traditional movie theater.

Having now ridden Soarin’ Across America several times, it’s easy to appreciate the result. The updated film sweeps from New York Harbor and the Grand Canyon to California’s rolling hills and countless landscapes in between through soaring aerial cinematography, practical effects, and a soundtrack that glides almost effortlessly alongside you. The audio shifts just as fluidly as the visuals, making each transition feel natural as the ride carries you across some of America’s most recognizable landmarks and scenic vistas.

Studio C could simulate much of that during production, but not all of it.

“For Soarin’ films, mixing historically required the construction of scaffolding inside the theater to reach the appropriate listening height for final mix decisions,” Lhotka explained. “Even then, the process was far from ideal, as the ride canopy significantly influences the acoustics. For this project, one of the key requirements was that the theater remain fully operational and open to guests throughout production — eliminating the possibility of scaffolding altogether.”

That constraint forced Imagineering to rethink its entire workflow.

“The challenge was clear: how do you mix a show with over 1,000 audio tracks and an extremely complex panning matrix while seated in a ride vehicle? The solution came through an innovative use of new AR technologies. By leveraging advanced screen-sharing capabilities, we were able to remotely control our production system — directly connected to the theater — while seated in the ride vehicle itself. This allowed us to place virtual screens in our field of view, effectively bringing the mixing console into the theater environment.”

Before anyone tested it inside the attraction, Studio C became the test lab.

“Studio C once again played a critical role as our testbed. Over the course of a week, we evaluated multiple configurations to confirm that things could be controlled precisely in this way, and that the AR headset’s external cameras would allow us to view the film clearly in the theater’s low-light conditions while simultaneously monitoring the virtual screens. The final configuration exceeded expectations.”

Once validated, the workflow moved from experiment to production.

“This workflow was ultimately used as the solution — and it became the method by which Soarin’ Across America was mixed. The result is a soundtrack that was shaped directly from the guest’s perspective, using tools and techniques that reflect our continued commitment to innovation, immersion, and storytelling.”

Having experienced Studio C for myself, that revelation immediately clicked. The room is already designed to recreate the feeling of sitting inside a Disney attraction with remarkable accuracy. But even a purpose-built facility couldn’t perfectly replicate Soarin’s unique acoustics. The only place left to finish the mix was inside the attraction itself.

It’s an elegant solution that solved two problems at once. Guests never lost access to Soarin’ during production, and Imagineering’s audio team was able to shape the soundtrack from the exact seat every future rider would occupy. Rather than approximating the experience from scaffolding or even Studio C, the final mix was crafted where it mattered most: inside the attraction itself.

For the past year or so, predictions of SaaS's demise have dominated the tech conversation.

Headlines warned that AI tools would replace traditional software, engineers would become obsolete and the SaaS industry itself was headed for extinction.

The narrative became so widespread that Wall Street gave it a name: the ‘SaaSpocalypse’.

Fortunately, reality has proved far less dramatic.

While the SaaS market - once the darling of investors - has undoubtedly taken a substantial hit, it is a long way from dead. In fact, I would argue that software doesn't die; it evolves.

If history has taught us anything, then it’s that major technology shifts rarely eliminate software altogether. Instead, they reshape it, with the biggest winners often being those agile enough to adapt.

From the rise of the internet to the shift to cloud computing, time and time again we’ve seen established players demonstrate an ability to evolve alongside technological change.

The AI era is unlikely to be any different.

SaaS subsectors best positioned to unlock AI-enabled growth

And I’m not alone in this view. Over recent months a number of high-profile economists, technology historians and analysts have arrived at that same conclusion. JPMorgan continues to champion the SaaS subsectors best positioned to unlock AI-enabled growth. Meanwhile, Goldman Sachs projects the global app software market could still reach $780bn by 2030, with agentic AI ultimately expanding the market rather than shrinking it.

Put bluntly, AI isn’t killing SaaS but rather separating the bad from the good. For years, SaaS startups have benefited from intense market hype and lofty valuations, often with minimum regard given to operational efficiencies, product differentiation or true value creation. Today that playing field looks a lot different. In effect, AI is now acting as a filter, revealing which SaaS models are superior and offer a genuine point of differentiation and, in turn, those which have for far too long favored style over substance.

For SaaS founders then, it isn’t time to panic. In our experience, investors-in-the-know will be more than willing to overlook the misguided SaaS narrative and recognize a genuine high potential opportunity should it arise.

Take, for example, the mid-market. Although much focus tends to surround the startup scene, I would argue that established software vendors are actually best positioned for AI transformation. After all, they already possess powerful structural advantages, including proprietary customer data, deeply embedded workflows and long-term contracts.

Moreover, they often sit on years of customer data. In this way, unlike startups where the focus is typically product, these firms actually have something meaningful to transform.

Well positioned to use AI to strengthen their market position

Far from becoming obsolete overnight, many are well positioned to use AI to strengthen their market position. Take, for example, Keyword.com. Having early recognized that search itself is changing, it launched a product to track brand visibility across AI platforms like ChatGPT, Perplexity, and Gemini. That repositioning opened up an entirely new growth vector.

In terms of how founders can best navigate this new terrain, it’s important to look beyond product and focus on structural resilience. For a long time, SaaS benefited from assumptions such as high multiples on ARR, heavy adjustments for stock-based compensation, and a willingness to prioritize growth over almost everything else.

That is now being replaced by a much more grounded framework. Investors are increasingly looking at real profitability, on a GAAP basis, and asking harder questions about cost structure, for example around sales and marketing efficiency, and now AI-related compute costs as well.

At the same time, as founders will be aware AI is introducing new variables. Revenue may be less predictable if it’s usage-based, and margins can be more dynamic because of inference costs, meaning investors won’t simply rely on simple rules of thumb anymore.

Valuation is becoming less about applying a multiple and more about understanding the underlying business, with a focus on true earnings quality, defensibility and efficiency.

Another significant but largely underappreciated upshot of the current environment is the emergence of roll-up opportunities.

The SaaS landscape is becoming increasingly fragmented

As valuations normalize and weaker players struggle to differentiate, the SaaS landscape is becoming increasingly fragmented. This creates fertile ground for consolidation strategies, particularly for investors with deep sector expertise.

Importantly too, AI enhances this model by enabling efficiencies across portfolios, from automated customer support to streamlined marketing and product development.

For investors, this is opening up a different kind of opportunity in terms of not just backing individual high-growth companies but participating in broader aggregation strategies that can unlock value through scale and operational improvement.

This means that purposefully positioning your business for ease of integration and standardization could pay dividends.

The short of it? Software isn’t going anywhere. Just like the major tech trends before it, it is simply evolving from a market that is less about hype and surface-level metrics to a more practical, mature model driven by genuine business outcomes.

For founders, this may mean navigating a more complex operating environment. But it also creates an exciting opportunity to differentiate.

Not all SaaS firms are primed for this new investment landscape, so those who pay due diligence by prioritizing intelligence over features, operational prowess and AI interoperability to deliver clear, measurable value will be best positioned to succeed.

We've reviewed and ranked the best business cloud storage.

This article was produced as part of TechRadar Pro Perspectives, our channel to feature the best and brightest minds in the technology industry today.

The views expressed here are those of the author and are not necessarily those of TechRadarPro or Future plc. If you are interested in contributing find out more here: https://www.techradar.com/pro/perspectives-how-to-submit

• Microsoft posted an ad promoting Copilot on social media
• It said that Copilot is the "button you can press to fix everything"
• This has elicited quite the outpouring of rage from some folks, and it's not a wise promo given the prevailing climate around AI and Windows 11

Microsoft is catching some flak once again over the topic of AI in Windows 11, following what might be regarded as an overzealous piece of marketing.

Windows Latest noticed that Microsoft's marketing department posted an image across its social media (Facebook, Instagram, and X) which promoted the dedicated Copilot button and the AI assistant it summons.

The text featured in the pic was: "Them: There's no button you can press to fix everything," followed by a reply as if in a conversation that read: "Me: Wanna bet?"

Predictably this has stirred up some controversy, with responses to Microsoft's little ad that range from accusing Copilot of being "AI slop" (or "Microslop") through to comments along the lines of "no one wants this" and similar negativity.

Undermining the fix Windows 11 effort

This is a particularly poorly timed piece of PR, given that, as Windows Latest observes, Microsoft is finally bringing in the ability to change the Copilot key back to function as Right Control. This is the key that the Copilot button replaced on 'AI PCs' (laptops), but as some people have said, Right Control is crucial to their workflow (and has accessibility ramifications for one-handed keyboard use).

With that change coming to Windows 11 later this year — an effective softening of Microsoft's initial stance that you should have the Copilot key and like it — this marketing snippet feels even more badly judged.

What's worse here for me is the exaggeration of the usefulness of the Copilot key. The AI can "fix everything" can it? It's a frankly ridiculous statement to make, and yes, I know that's not exactly uncommon in advertising, but you've got to bear in mind your target audience here and the context.

AI in Windows 11 has triggered a lot of bad feeling in the user base since last year, and indeed Microsoft's blinkered focus on pushing AI was one of the reasons that people were up in arms about fixing the OS rather than adding Copilot trimmings. Or in other words, the big fix Windows 11 campaign — which is the focus of this entire year for Microsoft — was partly the result of all the anti-AI feeling.

So, erm, let's market around the Copilot key being this amazing solve-all feature, shall we? Sounds like a great plan, everyone. Full steam ahead! Let's talk about how great it would be to add more AI into Notepad while we're at it, yeah? (Apologies, I remapped my Copilot key to be a 'Sarcasm' key and it appears to be jammed down right now).

Microsoft also says the Copilot key is a button with "main character energy" in this advert, the problem being that if AI in Windows 11 was a movie, it's been a box office turkey for the software giant so far. Microsoft's hope is that AI agents in Windows 11 will turn things around, but it remains to be seen how that'll pan out.

Meantime, while Microsoft is trying to generate positivity and good vibes around Windows 11 with all its (very commendable) work to fix the OS, it'd be better for the software engineers and designers in the thick of it not to have those efforts undermined by the marketing crew wheeling out unintentional ragebait like this Copilot button nonsense.

Now, I'm not saying that Microsoft's PR team can never mention AI, of course, but this particular snippet on social media is a perfect example of how not to do it in the current Windows 11 climate.

• Capture One has added native Hasselblad .3FR RAW file support
• Hasselblad's X2D II 100C, X2D 100C and CFV 100C are included
• Tethered capture is set to follow later this year

Hasselblad photographers, I have the news you've been waiting years for: Hasselblad cameras finally have native Capture One support.

The support covers Hasselblad's 16-bit .3FR RAW files and includes three of its 100MP medium-format cameras: the X2D II 100C, X2D 100C and CFV 100C digital back (which is the sensor module for the 907X). You'll need the latest version of Capture One 16.8.3 or Mobile Version 3.3.4.

It's truly a match made in heaven; Hasselblad's 16-bit RAW files and 'Natural Color Solution' render possibly the most true-to-life colors I've seen in my 15 years of testing cameras, while Capture One is industry-favorite editing software for color correction and tethered shooting.

This partnership between two Scandinavian companies had to happen some time, and now it has, with tethered shooting support (a studio setup where you can control the camera through the software) also promised for later this year.

It's a big deal for both parties — Hasselblad shooters can ditch Adobe, which is held in lower regard among pro colorists, in favor of Capture One and its precise photo-editing toolset.

The move might also attract new professional and commercial photographers, who have previously been put off because of the lack of Capture One support, to Hasselblad's medium-format digital cameras.

As for Capture One, it can potentially attract a new wave of existing Hasselblad photographers, with a seamless workflow from capture to the finished image.

Letting Hasselblad cameras sing

Until now, workarounds were needed to bring Hasselblad RAW files into Capture One, at the cost of color accuracy and editing latitude. Meanwhile, Adobe's photo-editing platforms, such as Lightroom, are held in less regard than Capture One for outright color fidelity.

Now Hasselblad's 100MP medium-format RAW files are supported, with profiles for Hasselblad's entire range of XCD lenses and the ability to organize files as well as edit them and, later this year, tethered capture.

I reached out to Hasselblad for comment, and a spokesperson told me: "One of the most consistent requests has been native support for Hasselblad RAW files in Capture One. This partnership is a direct response to that feedback, and we've been working with the Capture One team for some time to make it happen."

Capture One echoed Hasselblad's sentiment about the new partnership, saying "Giving Hasselblad photographers the native Capture One workflow they've asked us for is genuinely rewarding. Both of us are firmly rooted in the commercial and professional space, and that's a big part of why this partnership makes sense."

And when I asked about the challenges in bringing native support to a new format, Capture One told me, "The real challenge was less about simply reading the files. Getting a new format right means dedicated color profiles for each of the three cameras, so Hasselblad's files render with the true-to-life color our users know Capture One for, plus lens profiles for the XCD lenses that correct distortion, chromatic aberration, and light falloff.

"Throughout, the priority was making sure the unique character of these cameras comes through with the image quality we pride ourselves on."

You can sign up for a free Capture One 7-day trial if you've yet to try it, with subscriptions starting at $18 / £16.67 AU$27 per month, or a one-time purchase costing $349 / £336 / AU$523.

• Some Windows 11 users have found their drive space has been mysteriously dwindling
• In some reported cases, this is down to a database file for a Windows component that grows massively due to an apparent bug
• The optional update for June fixes this, albeit that patch is still in testing

If you use Windows 11 and have been wondering why your drive space seems to have been mysteriously dwindling of late, the culprit could be a background process called Capability Access Manager.

Capability Access Manager (CAM) is a rather obscure Windows component which, as the name suggests, manages the apps on your PC in terms of permissions to access the camera, mic and so forth.

CAM keeps a database that logs app requests and this is stored as a file called CapabilityAccessManager.db-wal – and it's that file which is growing hugely on the drives of some unlucky Windows 11 users.

Neowin reports that Microsoft just implemented the fix for this aggravating issue in the latest update for Windows 11. This is patch KB5095093 which is the optional update for June (so it's still in preview, or testing, at this point).

In the changelog for this release, Microsoft tells us: "This update improves disk space usage for the CapabilityAccessManager.db-wal file."

So, with any luck, this will be an end to the gremlin, and those who install the June preview will find this database file has been changed so it doesn't spiral out of control and eat their storage like some kind of log-based black hole.

Analysis: a baffling case of bloat

This is a nasty little problem, especially for those who've got a modest main drive for Windows 11, such as a 256GB SSD (or maybe even smaller). With reports of the database file expanding to a size of 60GB to 100GB, or even larger (200GB according to this Redditor), you can see how problematic this could be when you're already short on space with your SSD.

Microsoft doesn't explicitly say that this is a bug, but surely it must have been? Albeit a situational one that only occurred on certain systems by all accounts (Dell PCs are mentioned in a few reports, I should note). It's clear, however, that there are a lot of reports of CAM woes on Reddit and Microsoft's own help site, with some folks left scratching their heads as to what on earth was going on – and why their drive space was disappearing into some apparent void.

Unfortunately, these problems have been around for some time, with those reports going back to March 2026, and some dating from almost a year ago. As one Redditor explained in a post some 10 months ago: "Turns out it's the 'CapabilityAccessManager.db-wal' file which ate up 25GB. Looked through some forums and seems like it's a notorious culprit for system bloat. That file is supposed to slowly build up then be emptied each month, but mine bugged and never emptied itself."

There are workarounds to delete the CAM database file and effectively empty it out, but they're somewhat fiddly, and the log will rebuild itself if buggy conditions still abound on your PC. As one Redditor observed: "I spent hours trying to delete that file. Like in a horror movie, it just wouldn't die, even when I thought I had killed it. Your procedure [workaround] worked perfectly. Thanks!"

So, the whole thing has been a bit of a headache really, and while it's good to see an apparent cure now, it's taken Microsoft long enough. And frankly, I'm not impressed with the communication on this issue from the software giant.

To be clear, the fix right now is to download and install the June optional update for Windows 11, but bear in mind that as noted, it's a preview, so there's no guarantee it'll work as it should. The resolution will be provided with the full July update, mind, in tested form, and that's just around the corner – so you might just want to wait for that. Assuming your drive isn't chock-full and causing you a whole load of grief, anyway.

• Conversation focus is being limited to three hours a month for some Meta glasses users
• For a longer 15 hours a month limit you'll need to pay for Meta One Premium
• The feature is handled on device, so it's unclear why it has been paywalled

The Meta Ray-Bans and other AI glasses can perform a slew of useful tasks thanks to their wearable hardware and digital assistant — but Meta has just announced that one of those features will be getting serious limitations as it introduces a paid subscription plan that will unlock more usage time.

Conversation focus was showcased at Meta Connect last year, and is essentially an audio mixer for your real-life conversations. When you switch the feature on — using the voice command "Hey Meta, start conversation focus" — the specs’ microphones will pick up the voice of the person you’re looking at, the AI will separate their voice from the background noise, and then your glasses’ speakers will play what they’re saying to you to amplify their speech and make it stand out.

It’s an impressive feature, and very useful if you struggle to hear people in crowded spaces — or are starting to become a little hard of hearing and want a voice boost much of the time — but this is the feature Meta is now starting to limit. If you only have the free account that you create when you set up your glasses in the Meta AI app, you’ll just have three hours of conversation focus a month (as spotted by The Verge).

If you want more hours you’ll need to sign up to a paid Meta One Premium subscription, which costs $19.99 per month. Notably this subscription doesn’t unlock unlimited conversation focus access; instead your limit is bumped up to 15 hours a month.

In addition to this, Meta One Premium will also give you access to more advanced Meta AI reasoning models through the app, and expanded access to image and video generation models.

Was this inevitable?

We know that unlimited free AI access isn’t sustainable. Every AI query has a data center and energy cost to AI companies like Meta, and eventually they'll need to recuperate those costs or risk serious financial struggles.

However, no matter how inevitable a Meta AI glasses subscription seemed, this isn’t how I imagined it would roll out — frankly it’s quite a bad look for Meta, and I hope it reverses course.

My frustrations with this change as a Meta AI glasses user are three-fold. First, having something that was previously free taken away is never fun, especially as features like conversation focus felt like they’d already been paid for via the purchase of Meta’s glasses.

If Meta One had instead locked a new more powerful AI model behind a subscription, or some powerful new tool, that would feel very different. Your glasses would still do everything you were told they could do at launch, but to get some serious upgrades you’d need to pay up — sure this wouldn’t feel all that amazing either, but at least Meta wouldn’t be seen to be taking a feature away and holding it hostage.

My next frustration is that it’s conversation focus specifically that is being taken away. As I've mentioned, this is something of an accessibility tool, and while I’m sure most users only activate it periodically if there’re in a very noisy bar or somewhere similarly packed, for some I can see conversation focus being a tool they rely on frequently if they have more general hearing troubles but aren’t quite ready for a more sophisticated aid.

The flip side, unfortunately, is that if conversation focus isn’t a tool everyone uses all the time it’s the one least likely to actually inconvenience most Meta glasses users, allowing the company to experiment with paid access to tools without upsetting a large section of their consumer base.

Finally, according to Meta itself, conversation focus is handled entirely on-device. Your Meta AI glasses can offer the function without an internet connection, meaning it shouldn’t be costing its data centers anything to process.

In a way, taking unlimited conversation focus usage away is like Meta limiting how many photos your glasses can take a month. All the hardware and software tech is on your device; it’s just being locked away in what seems like a classic case of corporate greed.

Now, the silver lining at the moment is that if you haven’t noticed any rate limits yet, or heard about Meta One through your specs, it’s likely that your glasses are still working the same as they always have. As explained in the help article, “Meta One is currently in limited testing and isn't available everywhere yet.”

Additionally, I’m hoping the fact that Meta One is still in testing means Meta won’t be afraid to reverse course, especially as this decision feels so nonsensical for all the reasons I’ve outlined. Or perhaps it’ll be forced to rethink things if its AI glasses rivals (including the incoming Android XR specs) boast similar features at a similar price with no such paywall or usage limit.

Otherwise, this feels like yet another unforced error for Meta in the smart glasses space. How can we now trust tha tMeta won’t paywall any AI glasses feature we’ve previously had access to?

Following the announcement of its more budget-friendly AI glasses — including a pair styled by Kylie Jenner — I hoped it might be getting back on the right track. This latest announcement has knocked my confidence, but the ball is still in Meta’s court — perhaps Meta Connect 2026 will see it properly turn things around.

• Windows Subsystem for Linux gets new container feature after being teased last month
• CLI and API form the basis of the latest update, Intune management also added
• Other changes being made available to alternative container CLI tools

Microsoft has released Windows Subsystem for Linux (WSL) containers in public preview following its earlier announcement at Build 2026, as a way to create, run and manage Linux containers directly on Windows.

When the feature update was first revealed, Microsoft described it as "a built-in way to create, run and interact with Linux containers using familiar CLI & API."

The new container feature is now available in the latest pre-release version, which can be downloaded from GitHub, and includes two new core features.

WSL containers now available in public preview

As part of the update, Microsoft has added a built-in container CLI and an API to let Windows applications run Linux containers.

The company noted that the API is particularly useful for reusing existing Linux specific code, like running cloud applications locally, while the CLI tool uses a familiar format so users can "use [their] existing muscle memory when running Linux containers."

Microsoft also added new management settings to let admins control whether people can use WSL distros and containers via Intune.

Besides the headlining container update, Principal Product Manager Craig Loewen also highlighted a series of other changes the company has made to WSL, including a new 'virtiofs' file system that makes Windows file access 2x faster and a new 'consomme' default networking mode to improve compatibility.

Loewen noted that the lower level platform changes are also being made available across "other fantastic Linux container CLI tools" like Docker Desktop, Podman Desktop and Rancher Desktop.

Microsoft has also published a support page covering how to install the API, which includes support for C, C++ and C#.

The company hasn't yet provided a timeline for general availability.

• An ex-Microsoft software engineer has created a lightweight alternative to Notepad
• TinyRetroPad turns back the clock to avoid the bloat Microsoft has introduced to the text editor
• The engineer observes that the app has no bloat or telemetry, and that it's "just pure old school Windows done right"

Ever long for the days when Notepad was a lean, mean, text editing machine? If so, a software engineer who used to work at Microsoft has just released something you may well be interested in.

The Register noticed that Dave Plummer — who was likely one of admittedly many catalysts that sparked Microsoft's fix Windows 11 campaign — has created TinyRetroPad. (It's a fork of Dave's Tiny Editor or DTE by Matt Power, which, in turn, was built on the foundation of Plummer's HelloAssembly — the world's "smallest possible complete Windows application" no less).

TinyRetroPad is a fully functional text editor in the style of the original Notepad, completely streamlined and with all the bloat removed, so it's, well, tiny as the name suggests, weighing in at 2.5KB.

Plummer explains that he isn't keen on Notepad as it is, and so he "rebuilt it from scratch", with: "No bloat. No telemetry. No nonsense. Just pure old school Windows done right."

Analysis: note to Microsoft – debloat Notepad

In case you weren't aware, Notepad has been accused of being a bloated application for quite some time now, as Microsoft has expanded its features to cover all kinds of bases beyond what you'd expect from what's supposed to be a basic text editor.

Of course, the problem is that WordPad — which used to be the app that covered the middle-ground between Notepad and Microsoft's fully-fledged Word — was ditched back in 2024. Since then, Microsoft has drafted more and more features into Notepad in what's essentially covering for the removal of WordPad.

The trouble is that this is very much at odds with Notepad's core philosophy of being a lightweight text editor, and Windows 11 users now fear it's being bloated and will eventually end up less and less responsive, and therefore less useful as a quick-and-easy editor that puts a premium on convenience.

What all this means is that some people have abandoned Notepad and searched out third-party alternatives for Windows 11. Of course, TinyRetroPad represents another of these offerings, albeit about as pure and compact an alternative as you'll find.

How is this app so very small? Essentially, the program can be extremely compact because it taps components already installed in Windows.

As Plummer explains: "TinyRetroPad is basically a wrapper around the RICHEDIT50W control from the WinAPI."

So, Plummer notes, if you want Notepad to be "exactly like you might remember" from the Windows XP era, this is what TinyRetroPad does. I think the ex-Microsoft engineer may have a few takers on his hands.

• AI is forcing Apple to deliver security updates more often
• iOS 26.5.2 is part of Apple's new update strategy
• More updates mean better protection against AI-powered cyberattacks

It seems that the number of ways AI is changing the world is increasing. The requirement for more RAM to run new AI features in Apple products is already being blamed for the recent price increase in Apple products, as well as for the current high price of RAM itself.

Now, according to a recent Reuters report, AI is also to blame for the number of iOS and macOS updates we'll need to install. The bad news is that it's going up, all because of the threat posed by the latest AI models and their potential to aid cyberattacks.

Instead of waiting for the next scheduled operating system update for the latest round of security fixes to arrive, Apple is now delivering individual security updates ahead of the next iOS and macOS 26.6 update.

iOS 26.5.2 is here now

If you look in Settings > General > Software Update on your iPhone, you'll see that iOS 26.5.2 is waiting for you now, unless your iPhone already installed it overnight.

The description for the update reads, rather vaguely: "This update provides security fixes for your iPhone," but it's the dangers posed by AI that are driving this update, so make sure you install it promptly.

Details of the security updates for all Apple operating systems are available on the Apple Security Updates page.

Malicious hacking tools

The Reuters article states that the urgency of the update is due to the risk that AI now poses to Apple devices.

"The company told Reuters on Monday it was adapting to the reality that, given the ability of artificial intelligence to speed the development of malicious hacking tools, it needed to reduce the time between when updates were first made public and when they were put into customers' hands."

It looks increasingly likely that this will become the new normal for security updates, and companies like Apple won't be able to rely on bundling the latest security fixes into the next scheduled operating system update. Stand-alone security patches delivered whenever they're needed are something we're all going to have to get used to.

While it might be annoying to keep updating our tech gadgets more often, it's a small price to pay for better security in the AI era. We'll just have to get used to software updates being less about new features and more about staying ahead of increasingly sophisticated cyberattacks.

OpenAI and Anthropic are publicly disagreeing about whether their new AI cyber tools should be shared with European regulators.

OpenAI has offered Brussels access to its model. Anthropic is holding back, with Commission talks described as being at a different stage. Both have framed their position as the responsible one, and both arguments have merit.

But whether openness or restriction is the right call is ultimately a policy question and one that will take time to resolve. For organizations managing cyber risk today, the more immediate question is whether the teams are equipped to handle what these tools can already do.

AI systems can now autonomously carry out multi-step cyberattack tasks in controlled environments. Anthropic's Mythos completed a 32-step simulated corporate attack in testing.

Before it existed, no AI had ever done that in this type of full-chain simulation. Regulatory access to that kind of model matters for policy development. But the organizations that will be on the receiving end of attacks it enables are not waiting for that process to conclude.

The question of who decides when a powerful cyber tool is safe to deploy is important. But responsible deployment cannot just mean responsible release. It also means ensuring the organizations expected to defend against these capabilities actually have the people and skills to do so.

Most organizations are underprepared

Recent UK survey data found that only 27% of UK organizations are fully prepared for AI-powered attacks. Seven in ten are operating with partial or no AI-specific readiness, even though the vast majority of senior leaders already recognize that AI is increasing their risk. The awareness is there. The preparation is not keeping pace with it.

Part of the issue is that cyber security has long been treated as a technical problem with a technical solution. Buy the right tools, run the right software and you are covered. AI fundamentally changes that assumption.

When attack tools can learn, adapt and probe defenses continuously, finding weaknesses, failing and trying again without getting tired, the humans on the other side need to be able to keep up. That requires expertise, not just familiarity with a dashboard.

The skills gap is an operational risk

AI identifying a vulnerability is only the first part of the problem. Someone still needs to understand what they are looking at, assess how serious it is, prioritize it against everything else on their plate and act quickly. That judgement does not come from a tool.

It comes from trained, experienced people who have built that capability over time. This is especially important as AI-generated attacks become harder to distinguish from legitimate activity. Threat recognition at speed requires pattern-matching built through experience and training, not simply access to the right software.

The data supports this. Among organizations that have invested in ongoing certification training, 86% report a measurable reduction in cyber risk, with an average reduction of nearly 48%. Certified teams also recover faster when something goes wrong.

Nearly half of UK organizations surveyed experienced at least one attack in the past 12 months, with the financial cost most commonly landing between £100,000 and £199,999 once recovery, downtime, regulatory fines and reputational damage are factored in.

Regulation is moving, but slowly

This is also where the governance question gets more practical. Giving regulators access to frontier AI models is useful for understanding what they are dealing with. But that access is only meaningful if the organizations it is meant to protect have the capability to act on what those models can do. A policy framework built around tools most security teams are not yet equipped to respond to does not close the gap.

AI security standards are still being written. Most security teams have limited awareness of what frameworks even exist, let alone what is coming. The EU AI Act, NIS2 (Network and Information Security Directive 2) and emerging sector-specific guidance are all moving targets. Organizations that build continuous training into how they operate will be better placed to keep up as those requirements take shape.

The fix is known

For most organizations, the question of whether they have the skills in their people to respond when it matters is the gap between awareness of risk and readiness to manage it.

The investment in trained and certified security professionals has a measurable impact on an organization's ability to deal with attacks. It also builds the kind of internal capability that makes it easier to maintain regulatory compliance as requirements evolve. This isn’t a glamorous answer but the evidence for it is consistent.

Organizations that view training as a core part of managing cyber risk, rather than something to be revisited after a breach, are generally in a much better place. The tools and the threats will evolve all the time. It is the difference between resilience and vulnerability.

Connect securely online with the best VPN service.

This article was produced as part of TechRadar Pro Perspectives, our channel to feature the best and brightest minds in the technology industry today.

The views expressed here are those of the author and are not necessarily those of TechRadarPro or Future plc. If you are interested in contributing find out more here: https://www.techradar.com/pro/perspectives-how-to-submit

• Windows 11's File Explorer is receiving major improvements via the latest preview update
• File Explorer should perform much faster, including mounting disk images
• The update is currently in the rollout process

Microsoft is still hard at work fixing pain points within Windows 11 for its users. While this case isn't focused on gaming improvements, it addresses an issue that should improve one of the operating system's fundamental aspects.

As reported by Windows Latest, Microsoft is rolling out File Explorer improvements with its latest preview update, KB5095093. The update includes much better speed and performance of the File Explorer and increased responsiveness when mounting disk images.

This is a long-awaited update for Windows 11 users, as File Explorer hasn't always been the most responsive or fastest, especially compared to macOS or SteamOS, both of which feature fairly responsive file managers.

It's also worth noting that PC hardware, specifically SSD speeds and storage space, can have an impact on loading times. Still, the onus has been on Microsoft to fix its Windows 11 File Explorer, which has been noticeably slow even with a high-speed SSD — so this update is very much necessary.

Perhaps the most important improvement being rolled out is better performance for mounting disk images, which was arguably the worst aspect of Windows 11's File Explorer. This would often result in a five-minute wait for the disk image to fully open, and that's frustrating, especially for users with high-speed hardware.

Windows Latest also highlights that the 'Properties' tab on a file will be seeing a visual overhaul, with a more modern look using WinUI 3, as opposed to the Windows 95-era UI.

It's great to see Microsoft address these issues, seemingly as a wider attempt to improve Windows 11's functionality and appearance due to consistent user complaints. For sure, there's still more to be desired in the game performance aspect, but Microsoft should get praise where it's due.

My only concern is a potential update that could undo all of the hard work, since Windows 11 updates are notorious for breaking functionality, whether that's via File Explorer or as far as GPU stability — which have previously impacted Nvidia GPU users.

For the past two years, the generative AI conversation has been dominated by one piece of hardware: the GPU.

GPUs supplied the parallel compute needed to train large language models, and their scarcity quickly became a proxy for AI readiness.

But that shorthand is now incomplete.

The next phase of enterprise AI will not be defined by accelerators alone.

It will be shaped by CPUs, memory bandwidth, cloud capacity, networking, and the workflow systems that allow AI to move from casual experimentation into daily business operations.

AI’s true economic impact will not come from model access; it will come from whether businesses can turn AI into reliable, cost-efficient operational capacity.

AI is Becoming an Infrastructure Problem

The first wave of generative AI adoption was largely experimental. Employees used standalone tools to draft emails, summarize documents, or write code. These ad-hoc use cases were useful, but they did not require companies to redesign how work actually gets done.

The next wave is different. As AI moves deeper into enterprise workflows, IT infrastructure requirements become exponentially more complex.

A customer service tool that drafts a response is simple. An AI system that reads account history, checks policy, updates a CRM, logs the interaction, and triggers a follow-up task is an entirely different beast. This system does not just need a powerful model; it requires compute orchestration, secure data access, software integrations, permissions, audit trails, and fallback logic.

This is where the GPU-centric view fails. While GPUs remain critical for heavy inference, CPUs coordinate how these workloads interact with databases, APIs, security layers, and operating systems. As a result, memory bandwidth, latency, and power availability are becoming the true strategic constraints.

The High Cost of Unstructured AI Usage

The early enterprise playbook was simple: give employees access to powerful tools and see what happens. While this accelerated learning, it also exposed a massive financial vulnerability. Individual, unstructured prompting is expensive, difficult to measure, and hard to tie to tangible business outcomes.

We are seeing a major corrective shift play out among tech giants. Microsoft recently began pulling back internal licenses for Anthropic's Claude Code—which was costing between $500 and $2,000 per engineer monthly due to high token consumption—and is forcing its Experiences and Devices division to transition to GitHub Copilot CLI ahead of its June 30 fiscal year-end.

Similarly, Uber completely exhausted its entire AI coding tools budget in just four months. The ride-hailing giant deployed Claude Code to roughly 5,000 engineers and aggressively stoked adoption using internal leaderboards. The experiment was incredibly effective—assisted systems generated nearly 70% of committed code—but token usage scaled faster than anyone anticipated, forcing Uber's leadership to publicly question the net ROI.

Consequently, the future of enterprise AI will move away from fragmented prompting toward a central intelligence model. Rather than thousands of disconnected interactions, companies will rely on shared intelligence layers—centralized systems that understand corporate data, apply consistent business rules, route tasks across applications, and track performance.

This model is inherently more efficient because the same intelligence is reused across workflows rather than recreated from scratch by individual users.

From Answers to Workflows

The most critical shift in enterprise tech is the transition from tools that answer questions to systems that perform work.

Traditional software is deterministic: a user clicks a button, and a system performs a known action. AI workflows are more dynamic. An agentic workflow can retrieve real-time data, reason through a multi-step process, interact with third-party software, and loop in a human for approval.

This puts immense pressure on the full technology stack. To unlock actual productivity gains, businesses need clean data infrastructure, disciplined governance, and robust integrations. Advanced models are useless if layered on top of fragmented, disconnected corporate systems.

Unprecedented Change Management and the "AI-Native" Workforce

As these agentic systems mature, the impact on global employment will trigger a corporate change management crisis on a scale never before seen. AI will fundamentally alter hiring patterns and role requirements long before it eliminates headcount at scale.

Historically, headcount was the default lever to scale capacity; more customers required more support staff. AI breaks that linear relationship. Instead of asking how many people are needed to handle an influx of volume, leaders will increasingly ask how much of a process can be handled by automated systems.

This environment will aggressively reward adaptability. Professionals who stay ahead of the technology curve, learn to design AI-enabled workflows, and manage systemic exceptions will disproportionately benefit.

Conversely, the risk of displacement is starkest for those relying purely on legacy industry experience. Traditional technical and managerial paradigms are being disrupted by a new cohort of AI-native developers, product managers, and team members. These professionals do not just use AI as an assistant; they build, manage, and think in terms of automated, model-driven systems.

Those who fail to transition from traditional operators to AI-native orchestrators risk being replaced by those who do.

AI Infrastructure is Economic Infrastructure

The broader economic impact of AI will be determined by how deeply it can be embedded into the core systems that run global businesses.

GPUs, CPUs, networking, and data centers form the physical foundation. Agent orchestration, security, and observability form the operational foundation. Together, they dictate whether AI remains a novelty or becomes a scalable business capability.

The GPU race was merely the opening chapter of the AI boom. The next chapter will be defined by the holistic compute, data, and workflow systems that allow AI to do real work at scale. That is the moment AI stops being a tool and truly becomes infrastructure.

Check out our list of the best IT automation software.

This article was produced as part of TechRadar Pro Perspectives, our channel to feature the best and brightest minds in the technology industry today.

The views expressed here are those of the author and are not necessarily those of TechRadarPro or Future plc. If you are interested in contributing find out more here: https://www.techradar.com/pro/perspectives-how-to-submit

When Apple unveiled its upcoming software updates at WWDC 2026 on June 8, we didn't really get much on macOS 27 — aside from an animated skit about how the Golden Gate name was chosen. But having given the operating system a trial run, I can report that there is in fact a lot to look forward to.

It's worth emphasizing that installing the developer beta is risky: it's no exaggeration to say it can potentially brick your Mac, or at least some of its apps. Unless you're sure about this, you should wait for the public beta in July or the full release of the software sometime in the fall (for the northern hemisphere).

Just because I haven't come across any problems in my testing doesn't mean it'll be the same for you, but I can tell you that I'm impressed by what I've experienced so far. This is still very much a work in progress from Apple of course, so don't treat this as a review — features may come and go before it's pushed out to everyone.

As for compatibility, macOS 27 Golden Gate leaves Intel Macs behind. You'll only be able to install and run this if you're on an Apple Silicon machine. Here are my three favorite things about it so far.

1. It's smooth and fast

Apple has promised performance upgrades with macOS 27 — though it's been a little vague on the details — and while I haven't run any benchmarks, I'd say my MacBook has felt noticeably faster and snappier. Perhaps Apple is taking advantage of not having to take Intel chips into consideration any more.

Other users have noticed the speed increases too, and it seems that high performance tasks are particularly benefitting from this. Even though I don't do much beyond writing, web browsing, and photo editing, I've seen less in the way of lag and sluggishness than I did before, which bodes well.

These performance gains should hopefully translate into battery life improvements as well, though I haven't noticed any real change in terms of time between charges. Bear in mind that there are still months of development to go on macOS 27 Golden Gate, so it's likely to get better over time (this is still only the developer beta, after all).

2. Siri AI is a genuine upgrade

I have to mention Siri AI, which is now up and running on my MacBook. With a little bit of help from Google and Gemini, it feels like Siri is now genuinely useful on the desktop: answers are accurate and informed, relevant, and personalized to you.

One of the most helpful upgrades is the way that Visual Intelligence now works on macOS 27. You can highlight anything on screen (Shift+Cmd+Space is the shortcut you want), and then ask Siri something about it — and the assistant then uses clues about what's on screen and image recognition to serve up an answer.

It's the sort of feature that should've been in Apple Intelligence from the beginning, but at least it's here now. In addition, the dedicated Siri app and the integration with Spotlight works really well too, making the AI more accessible and more versatile.

3. The interface tweaks

There are numerous interface tweaks here that aren't major on their own, but which all add up to make a significant difference. Even something as simple as having an overflow button for menu bar icons is really effective — it means if you've got a lot of them, they won't start disappearing behind the notch.

The Liquid Glass slider has been given a lot of attention, and it works as advertised. You can find it in the Appearance section of System Settings, and I've moved it all the way to the right — it's as little transparency as possible for me, please. I'm actually hoping Apple gives us more control over this in the final release.

As we've written about before, the icons that were plastered all over app menus are gone as well, leaving behind an interface that suddenly seems more elegant and clean. This isn't a user interface revamp by any means, but it feels as though Apple's engineers have thought long and hard about what changes to make.

• Windows 10 is getting extended support through to October 2027
• That's an extra year on the original deadline, although Microsoft didn't formally announce the extension
• It gives those who were worrying about what to do next with their Windows 10 PC (especially if it wasn't compatible with Windows 11) some welcome room to breathe

Microsoft has given consumers another year of extended support for Windows 10 without any formal announcement of this move.

Windows Latest spotted that Microsoft changed its info on Windows 10's Extended Security Updates (ESU) scheme to indicate that support now runs through to next October.

The section about the deadline now reads: "Windows 10 support has ended. You can enroll in ESU any time until the program ends on October 12, 2027. If you're already enrolled, your coverage will automatically continue through that date — no action needed."

So, as noted, this means that if you're signed up for the ESU, you will continue to get updates not for another four months — as the original deadline for support ending was October 2026 — but for 16 months. It's still possible to sign up for the ESU program if you haven't yet done so, too.

Microsoft confirmed to Windows Latest that this isn't a mistake in the text, and the ESU does indeed now run for another year.

Analysis: a commendable move from Microsoft – but a jaded reaction

I still run Windows 10 on my main PC (Windows 11 is on my secondary PC), and I was about to fire up the upgrade most likely next month, or perhaps August — but soon, anyway. This gives me, and no doubt plenty of others, some extra breathing room. A lot of it, actually, which is very welcome.

I may still upgrade to Windows 11 this year, but given my procrastinatory ways, I might well wait until next year now. Especially seeing as Microsoft is busy fixing Windows 11 in all sorts of ways currently, so it kind of makes sense to wait until all that work is done anyway.

Clearly enough, this is a positive move for consumers, and I'm pleased to see it. I've argued before that a year of extended support wasn't enough — considering the hardware requirements rule out many PCs from upgrading to Windows 11 — and that Microsoft should give consumers a second year. (Especially given all the e-waste issues raised around Windows 10's end of the line in the past).

With that second year now confirmed, I'm rather surprised that Microsoft doesn't appear to have announced this anywhere, and just made the change to its website info quietly in the background.

Perhaps the idea is to keep it on the down-low, so as not to put off any imminent upgraders? At any rate, much of the reaction to this decision is pretty muted. One of the most upvoted comments on this Reddit thread is: "So are they finally admitting that Windows 11 isn't good enough almost 5 years later?"

You don't have to look far to find other jaded commentary like: "The moment the support expires I'll go with Linux."

Also, some folks have taken the line that it's not a surprise to see Microsoft extend support, and indeed that it always does this. And yes, that's true for businesses, where the company may support an outgoing Windows version for multiple years. But this isn't the case for consumers; remember Windows 10 is the first time Microsoft has ever offered extended support outside of the business world.

As a reminder, the support is free (as long as you're willing to sync your PC settings to OneDrive, which isn't that big a deal, at least not in my opinion), or you can pay $30. If you've already signed up, there's no extra charge for the second year, as you might hope.

As observed by a Redditor above, it's almost five years since Windows 11 was launched — and the OS was formally announced half a decade ago this week, in fact. I just wrote about that at length, and how I'm more optimistic about the future of Windows 11 at this point than I've ever been. Still, as mentioned, I won't be rushing to upgrade just yet on my primary computer, but I'll put the trigger on that update eventually.

• Only 31% of companies say they have accurate AI software visibility – 36% for overall IT visibility
• AI is complex to track because it's a layer that sits across existing categories
• IT leaders are also under pressure to prove AI ROI

New data revealed in Flexera's State of ITAM report has revealed that only 31% of companies have accurate visibility into their AI software – an area of IT spend that continues to grow as new use cases emerge and models develop.

This comes as nearly three in five (59%) reveal wasted AI spend has increased over the past year, with increasingly complex software stacks being blamed for an overall drop in visibility.

Flexera found that complete visibility across IT assets has fallen to 36% – only slightly ahead of AI visibility – thanks to complex and intertwined SaaS, public cloud and hybrid environments.

IT stack visibility is getting worse, not better

According to the report, ITAM professionals now spend more of their time optimizing software (32%), but audit response (22%) also takes up a considerable amount of their time.

As for AI, Flexera believes it's difficult to track because it doesn't occupy its own category. Instead, models, agents and platforms sit across multiple systems as an additional layer.

"What we’re seeing is a familiar pattern of rapid adoption followed by a scramble for visibility and control, as spend surges," Chief Product Officer Becky Trevino wrote.

But ironically, Flexera also found that AI promises to automate some ITAM work, including processing software contracts, purchase orders, licences and renewals.

With IT professionals under more pressure to prove AI ROI, the next steps move beyond identifying which AI tools are being used to consider how they're being used, how much they cost and what data they access.

Can you believe that it's now half a decade since Windows 11 was revealed? The operating system was first announced by Microsoft at a press event on June 24 back in 2021 (although the OS was leaked just before that, in typical fashion).

Five years ago today we were told that Windows 11 was inbound as a free upgrade for all Windows 10 users — but there was no stampede to adopt it, that's for sure (and the hardware requirements certainly didn't help the cause).

It wasn't until July 2025 that Windows 11 overtook Windows 10 as the dominant version of Microsoft's desktop OS according to Statcounter's figures, but it now holds a comfortable majority of over 70% of that market. As it should do, considering Windows 10 ran out of support last October (and only has a few months of extended support left).

However, putting aside the sluggish pace of adoption and the various problems that have plagued Windows 11 through the years (all the bugs and some notably missing features in the main), I think there's now cause for optimism for the future of Microsoft's operating system.

So, let me share my thoughts and reflect on what has been a half-decade of Windows 11, and tell you why I'm way more positive about the OS than I was last year — and why I think that Microsoft is finally on track (with, of course, some inevitable caveats).

The great fix-athon

Most of my hopefulness about where Windows 11 is heading comes, of course, from the big campaign Microsoft kicked off in March 2026 to fix Windows 11. Since that announcement — which I would say is the biggest statement to have been made since the OS was first announced in 2021 — Microsoft has very much proven that it intends to tackle all sorts of shortcomings and pain points with the OS.

Taskbar repositioning? We've got it. A much greater level of Start menu customization? Check. Spam removed from Windows search? Yep. More control over Windows updates? Certainly, and in fact way more control than I'd have ever believed might happen, including the ability to put off an update indefinitely, should you wish, on Windows 11 Home.

In fact, Microsoft has hit many wish-list features that I never expected would come to Windows 11, and the extent of the crowd-pleasing measures so far is heartening. These are features that are actually being delivered already, too, they're not just promises.

Furthermore, Microsoft appears to be listening to feedback and requests more closely, and generally engaging more with the community online. There's even a new initiative consisting of a research panel where Microsoft will consult testers directly on how to change aspects of Windows 11.

This genuinely feels like a fresh direction for Microsoft, and a serious commitment to change Windows 11 for the better based on what the users themselves actually want.

A more thoughtful, cautious approach

The other key driver for optimism with Windows 11 is the way in which Microsoft is taking more care over how the operating system is developed and coded.

Not so long ago, matters were less organized and more chaotic. Cast your mind back to the introduction of Qualcomm's Snapdragon X (Arm-based) chips in Copilot+ PCs back in 2024, alongside which Microsoft brought in a new underlying platform for Windows 11 (complete with the tinkering required to support that Arm silicon). While nothing was ever officially admitted, this is a move that I believe could have at least contributed to the mess that was the 24H2 update, which was laden with a ton of (sometimes very annoying) bugs.

Whatever the case in terms of how those glitches came to be, things have changed a lot since then. Microsoft is now being a lot more cautious with its Arm and x86 strategy — Windows 11 is split into two development paths, with the 26H1 update for Arm devices, and the 26H2 update for traditional x86 PCs — and the company has switched to use a fresh approach for these annual updates.

Instead of big annual updates — the last of which was the problematic 24H2 — Microsoft is now deploying small 'enablement packages', essentially very minor bumps to a new version of Windows 11. The actual features, the meat of Windows 11 changes, are pushed out in monthly updates as and when they're ready — in sometimes quite tightly controlled, carefully paced rollouts. This more gradual drip-feed of features is a more reliable method of deployment compared to dropping a lot of stuff all at once.

In short, Microsoft has learned its lesson from the nightmarish 24H2 release, which suffered from far too many bugs, to take on a fresh new way of operating. True, there will still need to be big updates at times, when the underlying codebase of Windows 11 has to be changed (quite possibly with 27H2). But it looks like Microsoft wants to mainly stick to compact, easily applied annual updates in the main, with features pushed out elsewhere in general.

Optimism abounds — with a notable catch

Between the ardent push to fix Windows 11 and the better thought-out deployment of features and updates, Microsoft has come a long way, but as I mentioned before, there are caveats here.

It's worth mentioning that while the new system of continual feature deployments, rather than weighty annual feature drops, is commendable (in my opinion), the controlled rollouts of these various features have come in for some criticism. Mainly because they are so cautious in some cases that something like the Start menu revamp (the one from last year, I should clarify, not the current work) took ages for some Windows 11 users to get, and those folks found that rather frustrating.

Part of that caution is likely down to Microsoft's paranoia around bugs, and sadly, the truth is that there are still too many bugs in Windows 11, and some of them are disappointingly weird. And by disappointing, I mean odd things that just shouldn't be happening.

I only need glance back to last weekend for one such example where there was a glitch with the Recycle Bin whereby the delete confirmation dialog (when junking a file from the bin) showed the internal file name instead of the proper name. While this only applied to the dialog box — so it was hardly an important or dangerous glitch — it was confusing some people, and more to the point, this sort of thing shouldn't be happening in the release version of an operating system.

This kind of bizarre slip-up also leads to folks blaming AI for being involved in Windows 11 coding. And while there is absolutely no evidence for that, it's the very nature of the strange bug that means people will easily jump to these conclusions when Microsoft has previously admitted AI is used to quite an extent in programming its software.

Whatever's at fault, Microsoft still needs to have better processes in place to catch these kinds of glitches, and other more critical bugs which have turned up in Windows 11 this year.

One thing I've called for in the past is a commitment from Microsoft to confirm that it's addressing its quality assurance processes, and improving bug squashing, and this is a notably missing part of the fix Windows 11 campaign. A vital part, in fact, I'd argue, for better stability going forward which is one of the big overarching goals (alongside better performance, and those crowd-pleasing feature additions).

Still, all in all, I've got to underline that right now, I'm as optimistic about Windows 11 as I've ever been. If Microsoft can tackle the bug blot on the OS landscape, and keep on listening to users — and its new research panel of testers from the community — Windows 11 could be in great shape come next year.

• Meta unveils $299 / £269 / AU$599 smart glasses (Meta Glasses) designed with EssilorLuxottica
• They match Ray-Ban Meta AI glasses on virtually all features
• They add adjustable nose pieces, and a lot of design and color choices

"It’s pretty easy to make glasses that don’t look good, it turns out," chuckled Meta CTO and Head of Reality Labs, Andrew 'Boz' Bosworth, shortly after unveiling a bumper crop of new Meta eyeware, simply called Meta Glasses, all fashioned in collaboration with EssilorLuxottica.

Bosworth says the team argues over "every gram, every quarter of a millimeter" in an effort to bring ever lighter, ever more comfortable, and ever more fashionable AI eyewear to the market. With this foray into making frames with a slightly less well-known brand than Ray-Ban, Meta is bringing the new frames — Adventurer, Fury, and Starfire Kylie [Jenner] Edition — to market at a somewhat startling price: $299 /£269 / AU$599 (not counting prescriptions).

No one size or style fits all when it comes to eyewear, Ankit Brahmbhatt, Meta's Senior Director of Product for AI glasses, told me as he walked me through some of the new frames. There are, he added, eight colorways and 26 different style options.

Ogling the style choices

As I looked around the room in the Manhattan event space, I spotted black, something like ivory, tortoiseshell, deep black, green, and a dark maroon. The frames range from larger and almost boxy-looking to thinner and lighter frames. In fact, many of the EssilorLuxottica frames are so relatively thin and light that you might miss the cameras hidden in the front, and the slightly thicker stems to accommodate components and batteries, and mistake them for normal frames.

Kylie Jenner's cat-eye-style frames are particularly fetching, and surprisingly looked halfway decent on me.

Brahmbhatt told me Meta worked closely with Jenner to develop the design, adding signature touches like a tiny gem in the frame, a mirror in the case, and even Kylie Jenner's voice in the Meta AI.

There are structural changes new to the Essilor Luxottica frames. The nose piece adjusts with a push to three different positions, the stems are bendable at the ends (Bosworth noted that the wires are coated with a kind of cellulose plastic), and the stems actually flex outward. I tried on almost every style I could find, and they were all quite comfortable.

One of the biggest changes, though, is the addition of a small button behind the traditional Meta AI glasses button that you might use to capture a photo or start a video: it's a tiny Meta AI summoning button, and I used it interchangeably with saying, "Hey, Meta."

More and better AI

This is also the first set of Meta AI glasses to feature a Meta AI system backed by the company's more robust Muse Spark models, which provide a more conversational voice, better context awareness, and the ability to tap into the zeitgeist by checking out social media (at one point I asked Meta AI if there was chatter online about fake food, and it confirmed that many were talking about it on social media).

I tried the new Meta AI in a few scenarios, and it ably identified whatever I was looking at (I could hear it snap a picture before the analysis), launched a music playlist based on my surroundings, and translated Arabic print for me.

That all of this comes in for under $299 (Ray-Ban Meta frames start at $379), and without compromising on the 3K video-shooting quality, 12MP photos, microphones, array, or speakers, is remarkable; but these are still relatively early days in the wearable AI space.

Getting it right and making it safe

Design is "really important if you want people to wear them as daily driver glasses," said Meta Head of Industrial Design Peter Bristol, who joined Bosworth on stage and took some questions from reporters.

In perhaps a nod to how thick and oversized AI glasses can look, the pair talked about how they made subtle changes to the designs, slimming down the frames, or simply making them look thinner by, for instance, adding a chamfer along the top edge of the frames, near the brow.

A good design means less friction, which Bristol believes can help with AI adoption.

For Meta, the goal is to "reach every corner of the market," said Bosworth, but that approach does come with risks.

When asked about growing concerns about the privacy of these glasses (there have been reports of people wearing them to illegally photograph women and even tampering with the glasses to turn off the LED "I'm filming you" light), Bosworth acknowledged these issues, but reminded us that Meta had actually "pioneered putting LED on the glasses," and talked about the anti-tampering technology they put in Gen 2 Ray-Ban Meta AI glasses. But, he added, it's "a cat and mouse game."

As for what the future holds, I pointed out to Bosworth and Bristol that, while they now have a lot of styles, not everyone wears, or wants to wear glasses. What about smart contacts?

"Absolutely, that one’s top of mind for the design team," said Bristol, adding, "We are thinking and trying the other potential paths, but it’s a complicated space, so glasses is front and center for us.”

Bosworth agreed with the premise of my question, admitting that he’s not a glasses wearer, but is happy to wear Meta AI glasses “because they brought a lot of value — but I’m aware that I’m doing it.

"The design team is absolutely captivated by this question. What are the other ways that we can deliver this capability to people who don’t have glasses on?"

"You can take the MacBook back. I don’t have the patience to learn a new thing," said my wife as she slid the MacBook Neo back across the kitchen counter.

It was the unceremonious end to a two-week-long experiment in which I encouraged my wife, a decades-long Windows user, to give Apple a try, more specifically, the flavor of Apple found in a lovely, citrus MacBook Neo.

Look, I am no pusher. Ever since I started testing the $599 laptop, my wife had been eyeing it. She was becoming Mac curious. This had a lot to do with her creaky Microsoft Surface Pro 4, a decade-old system that would soon face the dreaded end of Windows 10 security updates (no TPM 2.0 on that old Surface Pro).

As a realtor, she kind of loved the old girl. It ably ran all of her Web-based business software, handled emails, basic art needs for whipping up new listing sheets, and social media materials. She also knew, however, that it was almost time for an upgrade. We both wondered if the affordable and mid-range-powered MacBook Neo could ably step in for the Surface Pro.

My wife even told me that, if she went with the MacBook Neo, Cirtus would not be her choice; the blush looked rather nice. Of course, Citrus is what I had on hand, so that's what she got.

I told her that, while different, macOS would not be completely foreign. Sure, it moved some things around, but my wife was no Luddite; she picked up new tech pretty well.

One thing about her, though, should have been a warning: she hates change.

Taking the macOS plunge

Simply buying her a MacBook Neo — even if there are some nifty Prime Day Deals right now — without knowing if she'd take to it made no sense, and since I'd been testing and using this on-and-off for months, we decided to set her up with it as if she'd bought the system brand new.

I reset the MacBook Neo. It's a $699 model with Touch ID and 512GB of storage, and I told her that this would likely be the one we would buy anyway since I wouldn't want her to run into storage issues.

Next, we sat side-by-side as I walked her through the setup. She took to this part quickly, though I realized that the placement of Touch ID on the power/sleep button was non-obvious. When I told her to register her finger to unlock the laptop, she stared at the system for a few seconds, clearly looking for something with a fingerprint symbol.

You're on your own

For the next couple of weeks, I would watch her slip the laptop into her work bag or use it at the dining room table. I did catch her occasionally trying to tap the screen, which was unsurprising. After all, she'd spent a decade with a touch-screen convertible. I'd ask her how it was going, and she'd give me a slightly less-than-enthusiastic "OK".

At work, she said she struggled to connect to the office printer and finally had a coworker step in and help her.

When she was at home, I showed her how to add her OneDrive account to access work files, a move that seemed to both confound and confuse her. The Windows system automatically integrates the drive. For the Mac, there's an app and then a few steps.

Even the benefits I found in adding a Mac to my Apple ecosystem were lost on her:
"Why do my iPhone notifications keep popping up on the MacBook? That's annoying."

I kept waiting for that moment, the epiphany that triggered, "Why didn't I make this switch years ago?"

It never came.

She found the need to use two fingers to enact right-click functions confusing and didn't seem all that thrilled with the trackpad.

It was clear she appreciated the MacBook Neo design, and yet, she told me, “It’s nice and sleek and all of that, but I could get a new PC that’s nice and sleek and all of that."

As we walked through a Best Buy looking at both MacBook Neo and a bunch of comparable Windows machines from Lenovo, HP, Dell, and Acer, it was clear that the MacBook Neo had not won her over.

After she returned the system, my wife told me that she just couldn't get used to all the differences. Why was the control center at the top? Why did the dock look like that? How could they not have a Start button?

For the average Mac user, like myself, there were all good reasons for these design elements and placements, but, having been a switcher myself, I understood the confusion. Working on a MacBook after 30 years on Windows means you are living in a constant state of "Who moved my cheese?" And when you, as my wife does, have important work to get done, you can't have a system's quirks getting in the way of your productivity.

Getting real

None of this is a commentary on the still excellent MacBook Neo, but it does say something about a certain class of users who, while Apple might try to attract them with the affordable MacBook Neo, are unlikely to make the switch.

The reality is that the best market for the MacBook Neo is still the back-to-school market, where it will likely have an easier time of swaying kids who've used Chromebooks or have only been on iPads.

They'll become Apple's newest customers. As for aging Windows users, getting them to make the switch might be more than Apple can or should hope for — even with the appealing MacBook Neo.

• Microsoft has confirmed that 26H2 will be a minor update
• It'll be a "small enablement package" like Windows 11 25H2
• This means less likelihood of bugs, with major features being rolled out separately in monthly updates – a sensible approach, although that said, 27H2 could be different

Microsoft has confirmed that Windows 11's next annual update will be another minor affair, all in the cause of avoiding a repeat of the chaos around the 24H2 update.

Windows 11 24H2 was the last version of the OS to land bristling with major changes, as 25H2 was a small incremental update – and now it's confirmed that this will be the case for 26H2 as well.

Windows Latest reports that Microsoft confirmed that Windows 11 26H2 is a "small enablement package" just like 25H2 before it.

In reality, this means that the upgrade is essentially a lightweight download with no meaningful feature additions – so isn't that disappointing?

In a word, no, because it's just indicative of how Microsoft has switched to a more sensible way of working in terms of yearly updates for Windows 11.

Analysis: a more stable way of working

Instead of unloading a whole raft of new features in the annual update, Microsoft is now deploying those features in stages throughout the year. And that strategy of drip-feeding important new features in different monthly updates – with the usual controlled rollouts, the pace of which varies depending on how cautious Microsoft is feeling about any given piece of functionality – is a better way of working overall.

Not only does it mean that we don't have to wait until the latter half of every year to receive major new changes for Windows 11, but we don't have a massive download and complicated installation to deal with when the H2 update arrives, too. On top of that, there's more chance of things going awry when a big annual update lands, as we saw with the 24H2 update, which suffered a whole load of annoying bugs – and I think Microsoft learned its lesson from this.

So, what's the point of the annual update now? Well, 26H2 provides another marker for support, extending your ability to get updates. Those still on Windows 11 24H2 should note that support runs out for this version in October 2026 – which is when 26H2 should roll out (or just before, perhaps in September) – so those on 24H2 will need to upgrade to 26H2.

That said, the other possibility with installing a newer version of Windows 11 that's an enablement package is that while it doesn't pack any new features as such, stepping up to the latest release such as 26H2 may trigger a feature rollout for your PC that you wouldn't have got otherwise. There's no guarantee you'll get anything, mind, but it could work out that way.

All in all, I'd rather Microsoft worked this way for the noted reasons around the increased stability of piecemeal feature deployment, rather than them arriving in one big lump with the inherent danger of a bunch of gremlins therein.

Microsoft can follow this path because the foundations of the underlying codebase remain the same – which has been the case since 24H2 – although at some point, it'll need to take Windows 11 forward onto a new codebase, which is when we'll see a bigger update.

And we've already had signs that this will happen with next year's update, and that 27H2 could represent a big change for Windows 11, one that unifies the Arm and AMD/Intel (x86) architectures (with the former currently split off into its own branch of the OS, on the 26H1 update instead of 26H2).

Hopefully going forward, Microsoft will still stick broadly to the new philosophy it appears to have embraced, in terms of fewer big updates and multiple enablement packages in-between.