Become a TechRadar Insider
We hear it a lot: there’s a trust crisis because we no longer know whether what we see or hear online is real.
A crisis is a sudden rupture in the ordinary workings of life, during which the mechanisms that keep society coherent are exposed as inadequate or deliberately corrupted.
Co-founder at VerifyLabs.AI.
But the thing about crises is that you must know you’re in one to call it that. There’s no such thing as an unknowing crisis, unless you count midlife, and then only with the benefit of hindsight.
Right now, most of us are too wrapped up in the demands of our daily lives to second-guess our news feeds, social media timelines, or that voice note from a “work colleague.” Unless we’ve been directly targeted by deepfake technology, we aren’t aware of danger, nor of how quickly that danger is cannibalizing the social contexts we rely on. That unawareness is the point.
The Ship Was Unsinkable, After All
Humans don’t traditionally react until we are gripped by consequence. Consider the Titanic after the iceberg struck: it was business as usual for many on board until the water started pouring in. The Titanic took two hours and forty minutes to sink, and even when the danger had become undeniable, numerous passengers reportedly carried on as normal. They had been told the ship was unsinkable, so it had to be true.
The deepfake threat follows a similar pattern. The technology generating synthetic video, audio, and imagery has evolved far beyond what the public imagines. The deepfakes being produced today are qualitatively different from what existed six months ago, and the pace of that change is accelerating.
Detection capabilities, whether human or technical, are struggling to keep up. Most people are not trained media forensics experts and cannot reasonably be expected to become them. That gap leaves individuals, businesses, and institutions genuinely exposed, whether or not they’ve noticed the water rising yet.
One of the more unsettling features of sophisticated synthetic media is that false content can be more convincing than the truth. A well-constructed deepfake carries none of the visual noise or contextual gaps that might make authentic footage look uncertain.
In early 2024, a finance worker in Hong Kong transferred $25 million to fraudsters after attending what appeared to be a video conferencing call with his company’s CFO and several colleagues. Every face on that call was fake. No one questioned what they were seeing. The ship, after all, was unsinkable.
Doublethink in the Age of Synthetic Media
In George Orwell’s 1984, the Party engineers a crisis by forcing citizens to hold two contradictory ideas at once. Today, anyone paying attention to the growth of AI deepfake attacks is doing a version of the same thing: wanting and needing, as humans, to trust that what they see and hear is real, while simultaneously suspecting it probably isn’t. That cognitive dissonance has a cost. It erodes the basic social contract that allows us to function, to transact, to believe one another.
Governments are not oblivious to this. They talk quietly among themselves about a crisis but rarely use that language publicly. If you call a crisis, you risk a panic. So instead, the conversation happens in policy rooms and industry forums, while ordinary people navigate an information environment that is shifting beneath their feet without anyone telling them so.
What makes the deepfake threat genuinely unprecedented is this: even the worst dictatorships of the twentieth century, those that gaslit entire populations into observing different realities, could not remove the human prerogative of personal perception. Yet AI deepfake technology has managed something those regimes could not. It has instilled doubt into belief itself.
Detection as a Foundation, Not a Fix
Detection technology is no longer optional. In the current digital media ecosystem, it functions as a baseline requirement for any organization operating at scale. As synthetic media grows more sophisticated, human judgement alone cannot be relied upon as a defense.
The important distinction is that detection software is an enabler of trust, not a substitute for human judgement. The goal is not to remove people from the process of verification but to give them something solid to stand on. Trust begins with reliable identification of synthetic content, but genuine confidence grows when people can understand and act on what the technology reveals. A result returned without explanation, a confidence score without context, asks users to place blind faith in a system rather than develop their own informed understanding. That is a fragile foundation.
Transparent, explainable detection encourages healthy skepticism rather than blanket distrust, and the distinction matters. Blanket distrust of all media is paralyzing. Healthy skepticism is empowering. Well-designed tools turn verification into a learning experience, gradually building the kind of pattern recognition that makes future judgments sharper and less dependent on any single piece of software.
Verification as an Everyday Habit
Media literacy has always evolved in response to new formats and new threats. The arrival of synthetic media is the latest chapter in that evolution, and it demands a meaningful update to what digital literacy actually means. The ability to question and verify the authenticity of content is now foundational, not specialist—and it applies even when content seems trustworthy.
The shift that matters is turning verification from an occasional task into an everyday habit. When detection tools are accessible, intuitive and widely available, they empower ordinary users to fact check, rather than outsourcing their judgement to platforms or institutions whose interests may not align with theirs. Widespread access to verification technology strengthens critical thinking at scale and supports well-informed public discourse. Confidence replaces uncertainty, and that confidence is grounded in something real.
From Dialogue to Action
The imperative now is to move from commentary to consequence.
This means immediate legislation that proactively protects people in the age of AI tools. It means a willingness to put humans before profit, to penalize and shut down the companies enabling the exponential growth of the deepfake threat. It means shared responsibility across platforms, organizations and individuals, because no single actor can address this alone.
Trust can be rebuilt, but only intentionally. It will not recover passively as attention moves elsewhere. It requires deliberate investment in the tools and skills that help people to safely navigate the consequences of their own beliefs.
Trust is a skill to be supported, not assumed. The expectation that people will simply know what is real is no longer reasonable. What is reasonable is giving people the means to find out and then making sure those means are genuinely within reach.
Let’s stop commenting on the trust crisis and start helping people survive it.
We've reviewed the best Identity Theft Protection.
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