Dreamforce 2024: All the news and updates as it happened

Good morning and welcome to our live coverage of Dreamforce 2024!

TechRadar Pro is here in San Francisco after flying in last night, and we're looking forward to a busy week of news and announcements.

Dreamforce 2024 begins tomorrow (Tuesday 17 September) - so we'll be in the opening keynote to report everything going on!

Ahead of the opening, we've just been to pick up our badge, and it seems the Dreamforce campus is taking shape nicely.

As per previous years, there's a campgrounds-esque theme to Dreamforce, and we'll be sure to upload more photos tomorrow!

Morning from day one of Dreamforce 2024! We're awake (just about) and heading to get some sustenance before the opening keynote starts in a few hours time - stay tuned for all the details!

We're here! It's a packed entrance to the Dreamforce campgrounds, with everyone excited to get in for the keynote.

And we've made it to the press lounge! Coffee and pastries abound, before we head to the keynote theatre shortly...

Get ready to hear a *lot* about AI today - particularly when it comes to Agentforce, Salesforce's major new platform release.

Described by the company as, "what AI should be", Agentforce will let any business quickly get up to speed with the technology.

We're in and seated for the keynote - as expected, it's absolutely heaving. We've been told this is the biggest Dreamforce since 2019, with the pandemic hitting attendances - but there is apparently 45,000 people here this week.

As also expected, we're greeted by a soothing collection of surf rock-flavored pop and rock covers - if you're new to Dreamforce (or Salesforce as a whole), the company is big on Ohana, which (if you never watched Lilo and Stitch) - means family.

All the Salesforce mascots are here - Einstein, Astro, Rhodey - the gang is all present, sporting an AI-flavoured update in the form of space-age suits and accessories.

It's not an understatement to say the queue for photos with Einstein in particular is massive...

As usual, attendees are greeted to Dreamforce with the classic Hawaiian welcome - the Akaka'ohana.

Always a special moment, ahead of the tech and AI overload.

We're then given a typically bombastic video introduction, outlining the importance of AI for businesses everywhere (via Agentforce), narrated by Salesforce MVP Matthew McConaghey.

And with that, the lights go down, and it's time for the keynote - Salesforce CEO Marc Benioff takes to the stage.

"This Dreamforce is like no other...this Dreamforce is about humans with agents dirivng customer success together," Benioff declares.

"We're here to inspire you, to motivate you, to energize you," he adds.

Benioff also gives a shout-out to the state of California, as the company has essentially taken over downtown San Francisco for the week. Governor Gavin Newsom and San Francisco mayor London Reed also get a shout out.

Benioff takes to the floor for his usual walkaround - so it's time for an outfit check.

As befits a billionaire tech CEO, Benioff is snappily dressed, sporting some incredibly stylish Dior sneakers.

The first shout-out for AI - Benioff looks to highlight how the technology has always been a big focus for Salesforce.

"This was a moment...of change," he adds, noting the company saw the first generation of AI happen, and how it was set to evolve. He notes he uses Einstein regularly to sum up meeting notes, with some copilots not always performing how the should.

Salesforce's AI platforms are already generating two trillion AI results per week, he notes.

"That's why this show is our most important Dreamforce ever...you're going to see technology like you have never seen before."

"This is the third wave of AI - agents," Benioff notes.

"This is going to be something we're not going to forget...we're doing something no one has ever done before," he chuckles.

"We helped our customers get into AI first."

"This technology is going to be so important, it has the best AI in the world," Benioff notes.

"We've been proud for 25 years," he adds, noting how Salesforce has helped customers move from software to cloud, to AI.

Now, it's on to Agentforce - the next step along the journey.

Workforces are overwhelmed and customers expect more, Benioff notes - workforces are different now since the pandemic.

So can AI help solve these issues? Benioff says Salesforce's dream for these agents is to do just that, ticking off low value and repetitive tasks for businesses in many industries.

"We can build a different kind of technology platform...the next capabilitiy that's going to make our companies more productive, and deliver better business results."

"Agentforce has to be the biggest breakthrough we've ever had on technology, and I think it's the biggest breakthrough I've seen in a long time in artifical intelligence," Benioff declares.

Now we're on to some early customer successes for Agentforce - including Wiley, an education provider, and restaurant booking firm Opentable (turns out Benioff is a creamed spinach fan).

Benioff also hints at the second version of Agentforce, named Atlas - and notes he's off to Disneyland next week after Dreamforce, to highlight the company's work with Disney, which is using Salesforce to help create amazing guest experiences.

Agentforce is what AI was meant to be, Benioff notes - expanding on the company's foundations of Data Cloud and Customer 360, providing a huge scale of services and tools for customers.

This has all changed within the last 12 months, Benioff notes, with agents transforming the way we all work, offering security, customization, and..."it just works".

We've all been told DIY is the way to go with AI, Benioff says - but DIY means you're actually putting it all together on your own - whereas what you need is something smart and deployable.

"I think we're also going to inspire at this show a whole new generation of apps build on our platform," Benioff claims.

Agentforce can deliver the "first Agent ecosystem", Benioff says - and welcomes Clara Shih, CEO of Salesforce AI, to the stage.

"In the very near future, agents will be as commonplace as apps and web pages," Shih declares, giving us an insight into how to build agents for the enterprise.

Only the Salesforce Platform can offer everything you need to get AI agents up and running as fast as possible, Shih notes, with quick integrations with Customer 360, Data Cloud, actions such as prompts, channels such as Slack, and built-in trust and security.

"Everything that you've ever done has led you to Agentforce," she declares.

Now, another look at Atlas - the "brain" of Agentforce, which helps the platform learn and reason from Data Cloud, then takes action across Customer 360 to create outcomes.

Agentforce is now going to be in every Salesforce app, Shih reveals, and the company is also releasing a host of Trailhead tools to help workers learn more and experience the platform, including Agent Builder.

We're now getting a look at what Agentforce is doing for customers already, with clothing giant Saks showing how it can help shoppers get customized recommendations, giving them a deeper interaction with the company.

The platform isn't replacing human workers, but augmenting them, the company insists.

We've just seen a lengthy demo of how Agentforce can support a customer service model, showing the intricacies of how the platform can deal with even super-specific customer queires.

It's all very next-level, promising a new age of customer services - all powered by Agentforce.

Benioff returns to the stage, full of praise for Agentforce.

"You can see why we're so excited," he says. "It's not just what is possible, but what you're going to make possible."

"Our goal is simple, for you to augment your employees, create better experiences, and deliver better business results for your companies."

Agentforce performs OpenAI on Microsoft Azure, Benioff claims, challenging users to benchmark the platform and see the cost and time differences.

Benioff calls up Parker Harris, co-founder of Salesforce, from the audience, to talk about AI customer success over the past 25 years.

"It's been a journey - our job was just to make it easy," Harris says.

"The Salesforce Platform has evolved," Benioff notes, "it has transformed, it's evolved, it's gone forward - but it's still the simple, easy-to-use platform that's inside all your companies already."

"Our job is to take you to the next place."

Now we're moving on to Data Cloud.

Benioff notes customers have "islands of disconnected data" holding them back - and Salesforce wants to help them utilize this.

Companies are now processing 767 trillion records every month on Data Cloud, Benioff notes.

"We were preparing you for the next step," he notes.

Benioff welcomes Sanjna Parulekar, VP of Product Marketing, to the stage, to talk us through the latest Data Cloud upgrades.

Top of the list is harmonizing all kinds of data, across a number of apps and services, to build a singular customer profile.

AI may need data to understand your business, but this doesn't need to be a tough journey - prompts can teach an LLM about your business.

Benioff returns, and now it's time to focus on Customer 360.

Salesforce is the world's #1 CRM, he notes, which is why its AI can be so accurate - with tools such as Sales Cloud, Marketing Cloud and Service Cloud all contributing to boosting success.

"Every cloud and every agent, on one trusted platform," is the goal, Benioff says.

Tableau is also getting an Einstein-powered upgrade - and Slack is also seeing even more interoperability - meaning you can have just one platform, for your entire company.

There will also be over a hundred industry-specific prompts for Agentforce, from Automative to Media to Education to Financial Services.

"The idea that every customer can have this next generation and to connect with their customers in a whole new way...is just too hard - but we can do it in Agentforce," Benioff declares.

"This is an unbelievable opportunity," Benioff says, thanking the keynote team for what he says has been the best Dreamforce start ever.

Customers can get going on Agentforce today, he says, with the new Salesforce Foundations giving companies a free upgrade to several major platforms, to give you an AI boost.

And that's a wrap! A bumper keynote indeed, lots to consider and cover if you're a Salesforce user.

We're off to digest all the news, and will be back shortly with more coverage from Dreamforce 2024.

And we're back (already) as Marc Benioff has come to the press lounge for a Q&A...

After a quick intro, Benioff opens it up to questions from the floor.

First up, a question about competitors also announcing AI agents - is Salesforce just following suit?

Benioff says Salesforce is far ahead of others, and urges us to "dig into the code" and really investigate what these other firms are doing - to talk to customers and hear their experiences.

He adds Salesforce's aim of having a billion customers interacting with Agentforce by this time next year.

Next up is a question about words - specifically trust.

When it comes to a major pivot such as Agentforce, how does Salesforce deal with acceptable levels of hallucinations?

Benioff calls in some extra help, but notes tuning and nudging will help fine-tune things as time goes by.

Next up, a question on the amount of training that will be needed for workers to get up to speed on Agentforce.

"Very little," Benioff states - noting the platforms are low- and no-code. The vats majority of users won't be AI experts, but Salesforce platform experts - often knowing little about agents or AI.

"Our goal is to empower them," he says.

Next, a question on trust - why should customers pick Salesforce over Microsoft or other competitors?

Benioff notes with a laugh that Microsoft has always been mean to him, but the company only cares about customer success.

"All that other shit - we don't fucking care," he says, tongue firmly in cheek (we think).

"Over and over, we have shown our approach is better."

Benioff has referred to comparisons between tech firms as a "bake-off" three times now...is he a secret Paul Hollywood fan?

"I hope we're on the right side of history here...we realize we're dealing with the most avant-garde technology, but how many of us really have the ability to get our hands into the soil?" Benioff asks.

We've been in the generative AI world for three years now, but there's still a huge possibility to come, he says.

Next up, a more personal question - what was Benioff's "Eureka" moment when it comes to AI?

He replies in San Francisco, such moments are common - the opportunity now is spotting the next big thing.

"It's a great place to do business," he notes.

He adds that this morning's Dreamforce keynote was, in effect, a Eureka moment, with all the content completely rewritten in recent weeks following customer breakthroughs.

Benioff also notes he recently spoke to X founder Elon Musk about the potential AI and the future - sadly he doesn't reveal too much about exactly what they discussed.

"And I'm not sure what's going to happen...but we believe we have to do this," he declares.

"This is really important - it's not just hype."

What's next after Agents? A challenging question - Benioff mentions prompt engineering and deep learning.

Deploying AI agents in the physical world is going to be a major step, as will the next generation of robotics...

Next, a question about Salesforce's relationship with Microsoft - what does he think of them?

"I think they're great!" Benioff replies, very tongue in cheek, referring to a recent EU report on Microsoft's recent troubles regarding Slack and Microsoft Teams...

"I can't really comment any more on that," he says, sadly.

Is Salesforce doing too many different clouds, in regards to its investments?

The focus has shifted now to "more core", Benioff notes - and that the company believes it needs to do this, "in order to end up with the AI that we want".

It's not totally done - but its well on the way, he says.

"For us to deliver the AI vision, we have to get the loose threads tied up together," he notes.

A question about jobs next - especially with AI and recent cutbacks in customer service and entry-level jobs.

Benioff thinks back to a visit to Gucci, with its high-level customer service, noting that those agents were actually faster and more efficient - so the end results may really depend on the business using it!

"AI can go in many directions," he notes, "there's going to be two ways to look at it" - either efficiency boosts will lead to cutbacks, but could also allieve pressures on some staff.

And with that, it's a wrap - thanks to Benioff for dropping by, we're off to get some lunch!

Good morning from day two of Dreamforce!

After a packed first day, we're ready to hit the floor once again. Sadly the weather is not quite as sunny as yesterday - but we're sure it'll brighten up soon.

Our first port of call is the Agentforce keynote - we're set to hear more about Salesforce's major leap forward in AI.

CEO Marc Benioff described Agentforce in yesterday's keynote as, "the way AI is supposed to be" - so let's hear exactly what it can do for businesses and customers alike...

Einstein and Astro are here - but only Astro is in his AI-boosted new look...what could that mean?

Or have we just not had enough coffee yet...

After a brief intro video, (accompanied by some weirdly off-key clapping) the lights go down and Adam Evans, SVP Product at Salesforce, takes to the stage.

"This is such a moment...there aren't many moments like this," he declares.

We're now in the third wave of AI, Evans says, repeating Marc Benioff's mantra from yesterday.

The first wave was predictive AI, then generative AI - now it's time for the third wave...

Workforces are being overwhelmed with tasks, and don't have enough time to do it in - AI can be the answer.

"We are in the era of agents," Evans says. Agentforce is built on the Salesforce platform, so you can get up to speed easily.

Agents will combine your company's data with customer information, backed up by the Atlas reasoning engine in Agentforce to create all-new insights.

He runs through a whole page of possible use cases, from patient care to customer service and online shopping order manhement.

Agentforce is about bringing humans, AI, data and actions together, Evans says, turning things over to Gary Brandeleer, Sr Director of Product Management, for some demos.

"Everyone is chasing the same dream," when it comes to enteprise AI, Brandeleer says - and being told to do it yourself is not really the answer.

What is? Surprise surprise - Agentforce, which works with your employees, agents, data and (Salesforce) CRM to create and build a new way of doing things - without needing to learn a whole new family of technologies.

Agentforce doesn't just assist, Brandeleer says - it takes action on your behalf, and can easily hand over to humans when needed.

He gives an example of a customer who had 140 intents in their chatbot, but moving to Agentforce, simplified this down to just two.

He names a range of possible use cases across multiple industries, noting, "the beauty is, the possibilities are truly endless."

It's time for a customer use case - Brandeleer outlines how educational publishing firm Wiley has used Agentforce to deal with seasonal surges (as seen in back-to-school rushes).

The company used agents that were customized to its specific policies, handing over to a human when needed, taking actions based on instructions dictated by the firm.

All of this on a platform your IT team will be familiar with, due to being built on existing Salesforce infrastructure.

It's time to talk about data - Avanthika Ramesh, director of product management, is here to talk about trust.

Trusted agents need trusted data, she notes - "your AI is only as good as your data".

"It's all about bringing the right data to the right prompt, at the right time."

As mentioned, Agentforce can use Data Cloud to make the most of your data - another advantage of everything on the Salesforce Platform being so connected.

The company has announced Prompt Builder, a new tool designed to help simplify the process.

We're joined by Ezcater, a meal-delivery company using Agentforce to solve its customer service hurdles, for example when ordering specific amounts or ingredients.

After a lenghty demo, Evans returns, and now we're turning to Actions.

Claire Cheng, VP software engineering, takes to the stage.

"The true value here is about getting work done," she notes, "agents being able to take actions is what empowers employees and drives success."

The Atlas reasoning engine gets another big shout-out, giving companies the power to get any task done, so long as they can describe it.

Ultimate customization is the goal, Cheng notes - and the new Agent Builder can help this.

The platform uses AI-powered tools to describe and build any agent, which can then be connected to prompts, flows and more.

"We are truly bringing together humans and AI agents to drive customer success," she notes.

We're then given a demo of exactly how to build a customer agent using the new platform, from set-up to launch.

The platform offers a number of pre-set solution, or you can establish one from scratch. Just a few simple prompts are all that's needed, with the platform then suggesting topics , tweaks and possible expansions to your initial idea.

And with that, it's a wrap - we've learned a lot more about Agentforce, and how it can help businesses in every industry.

Evans returns to wrap things up, noting many of the solutions will be available within the next few weeks, and that customers can pilot Agentforce now.

We're off for some coffee, and will be back soon!

Next up, we're at the Yerba Buena Theater to hear Salesforce CEO Marc Benioff talk to US DHS Secretary Alejandro N. Mayorkas for a deep dive on cybersecurity and AI. Learn about urgent challenges, cutting-edge threats, & bold strategies needed to secure our digital future.

Matthew McConaghey is here...maybe he's a big cybersecurity fan?

Benioff gives him a shout-out - apparently he's the brains behind the line of "this is what AI was meant to be".

We're right into things, with a question about AI - Sen. Mayorkas highlights the "Incredible opportunity" AI presents.

"We want to create the space for its potential to be realized," he says, "I want to demonstrate government can move at the speed of business, that we can harness AI to advance our mission."

Mayorkas outlines the role of the AI Safety and Security Board, where he says the key is to develop a framework that includes and accounts for the entire ecosystem - developers, cloud compute, the entities using AI.

"What we're trying to do is capture the entire ecosystem with a framework that defines roles and responsibilities where everyone has a role to play."

"The magic is bringing everyone together - there's a strength in collaboration and collective action," Mayorkas adds. "I think it's very important to get everyone around the table, working towards a common goal."

Harking back to his childhood in Cuba, Mayorkas said organizations need to take responsibilites towards change.

"I'm not interested in the static, I'm interested in what tomorrow can bring, and how we can redefine ourselves for the future."

Mayorkas has worked under three seperate Presidents (Clinton, Obama and Biden, and Benioff asks him about the differences in leadership, and what lessons he took from them.

Mayorkas says empowerment was a key factor, with the Presidents giving the freedom for their appointees to carry out their policies.

Benioff turns now to the recent political situation in the US, particularly the assassination attempts against former President Trump.

He asks Mayorkas about his insight onto "what is really going on".

"We are in an intensely heightened threat environment," Mayorkas says - not just from foreign terrorism, but increasingly "home-grown" extremists too.

The levels of hate in the US right now is a major worry, he adds - "it's an incredibly challenging landscape".

"A divide speaks of a chasm - and nature abhors a vacuum, and it must be filled," he adds, harking to the recent controversy surrounding false claims against Haitian immigrants in Springfield, Ohio.

"Actions are more instructive than words," Mayorkas adds, highlighting the tremendous work done by the Secret Service.

We have to learn to disagree, Benioff notes, so what can Secretary Mayorkas tell us about how to look, or who to turn to?

Mayorkas says he draws inspiration from his family, his faith - but also the next generation, who are growing up in a more progressive society.

It's about building bridges, Mayorkas highlights, the same as with AI - "it's not 'they'...but it's me, it's I, it's you, it's we, it's us - we own it".

"It's on all of us, in whatever role we play."

Mayorkas turns the table a litte, highlighting how Benioff and Salesforce can reach a huge potential audience, not just through its software, but through corporate responsibility.

Benioff harks back to his work with President Bush, and particularly cybersecurity - which was only just becoming a potential threat back then.

Funding had been an issue, and remains so today, so Benioff asks why is it difficult for tech figures or firms to provide the right kind of advice for government?

Mayorkas notes the situation is very different from redirecting funding within a tech business - the political landscape is very different from governance, so finding the right processes and agreements together is often extremely difficult.

"The mosaic is very different," he says - but there are people who can bridge that divide.

And with that, it's a wrap - a fascinating insight all round, look out for our full write-up soon.

Next up, we have a Future of Work AI panel - made up of Mick Costigan, SF Ventures, Denise Dresser, Natalie Scardino, CPO Salesforce, Ben Kus, CTO Box, who are going to be talking about all things AI in the workplace.

Asked about the State of Work today, Dresser notes, "it's pretty challenged today."

"The work of work" is a big drain/worry, she adds, as enterprise apps aren't connected, and data is disjointed.

"The future of work will have to be a platform that brings all your humans, all your data, and all your workflows, in one place"

Asked about the effect of AI on skills/jobs, Scardino notes things are more challenging, as different demands are affecting the needs of all kinds of businesses.

The panel is asked about the start-up feel of both Slack, and Box - how does AI impact that?

Dresser says AI going to open up markets we don't even know about - freeing up time for skills and learning, so companies should embrace it.

Kus adds efficiency and productivity boosts are bound to come from AI, which is helping everybody within his business.

"Human-centric" AI has become a big theme at Dreamforce, with Scardino asked about the influence of Salesforce's workers on the company's releases.

She notes working closely with the company's CTO to influence certain products, but the use cases are always crucial - Salesforce has annonced 120-odd agents recently to help workers in all areas of the business.

This isn't about removing humans, but adding scale, Costigan adds. Scardino adds guard rails are also crucial, with certain tasks always better done by a human.

The panel is asked about advice they would give to companies considering AI adoption.

Kus says finding a starting point is crucial - especially with something as complex as AI. The technology is changing so rapidly, having an anchoring point is vital - as is giving people access earlier, so they can figure out exactly how best to use it - they might come up with use cases you never considered.

Scardino says bringing employees on the AI journey is important, giving them insight into the benefits the technology can bring.

Finally, Dresser says thinking of how AI can be a leadership totem, with bosses able to push through advances that will bring better working environments and habits for everyone.

Time for some Q&A. Dresser is asked about the challenges of using AI at work.

Exposure, education and trust are crucial considerations, she says - Slack wants to help onboard and get users up to speed as quickly as possible.

Scardino adds the action piece is also important - even with summarizations on calls you may have missed, these reports can help empower workers.

Next is a question about entry-level jobs - what will the effect on them be if AI is taking all the basic roles?

Kus says he has seen an embracing of AI among young people - they are used to having the technology help them do tasks, and this helps them be more effective in a work environment.

Scardino adds the importance of getting the workforce ready for the next evolution - Salesforce has invested heavily in getting support into educational institutions to boost just this.

There are going to be new jobs, Dresser adds - the job market will evolve!

What about the risks of workers "going rogue" with AI, such as using their own personal accounts at work?

Dresser notes Slack's own research has found having clear guidelines on AI usage is "critically important". Being responsible and transparent is a human character, she adds - so everyone needs to step up.

Scardino says she's asked the question a lot - and that Salesforce is looking to address that, both in-house, and through customer conversations and consultations.

Finally, a question on productivity and "how busy you look" - how can AI help create new KPIs and overcome this barrier and help management understand it better?

Kus says the advent of AI means you can end up with the output being more important than what is put in - so it's all about AI helping us do more.

And that's a wrap - lots to digest!

The end of day two, and that's a wrap on our Dreamforce 2024 live coverage.

We've got plenty more to come though, including briefings and features with some very senior tech figures, so stay tuned to TechRadar Pro for more soon!