Deepfakes & Fake News: The First Major Casualty of AI Is Truth
AI is going straight at something less visible than our jobs or our creativity. Its first big casualty is nothing less than truth. For now, philosophical and scientific truths are getting off relatively easy. But when it comes to everyday truth, we’ve hit the point where trusting what you see, what you hear and what you’re told keeps getting harder. So you can already start wondering whether this gradual erasure of truth in favor of lies is more or less deliberate. Because when you take a close look at how techno-fascism is advancing, there’s plenty of reason to ask that question very seriously.
Dictatorship thrives in the soil of ignorance. George Orwell
Right now you might not see the problem coming. Maybe you’re comfortable with computers and you spot the fake stuff easily enough. But you’ve probably noticed that lately you have to look twice to keep from falling for it. And maybe you’re not worried because you haven’t read a real deep dive on the subject yet. Which is fair enough, since it isn’t what mainstream media worry about. They’re using and rebroadcasting more and more AI-generated content, because it pleases the people who sign their checks and lets them cut staff at the same time. So yes, access to quality journalism is in serious trouble. But this raises far worse problems, and that’s what we’re going over here.
So what is techno-fascism?
Before we go further, let’s go back over the definition of techno-fascism, because you need it to follow what comes next…
Techno-fascism is what happens when tech power concentrated in a handful of private hands meets the authoritarian drift of political power. Put another way, it’s the moment Big Tech and certain governments stop being separate things and become two sides of the same coin. One brings the muscle of new technology, the other brings legitimacy and the force of law. Together they get to control information and the story told about it. Which lets them do a great deal to shape how people perceive reality. And Orwell had all of it figured out long before the word existed.
Political language is designed to make lies sound truthful and murder respectable, and to give an appearance of solidity to pure wind. George Orwell
What Orwell described with words, AI now does with images, voices, videos and whole texts. Propaganda has always been around, but it kept running into its own technical limits. Those limits are falling one after another. And the worst is still ahead.
Before AI, fakes were hard to make
For a long time we got lucky without really knowing it. The best Photoshop job always gave itself away in the end: a shadow in the wrong place, a suspicious pixel, a blur that didn’t match the rest… Up to then, a trained eye could still tell. Video had almost sacred status, treated as proof you couldn’t argue with. An audio or video recording was admissible in court and could swing a verdict. Those markers shaped how we related to reality for a long time and held a line, imperfect as it was, between what’s true and what’s manufactured.
And it wasn’t only a technical problem. Building a convincing fake took time, seriously specialized skills, expensive software and often a whole team. Cloning a voice, doctoring a video, faking audio credibly enough belonged to the exclusive preserve of Hollywood studios or intelligence services. All that friction naturally slowed the spread of fakes, because manipulation at scale was hard, costly and, above all, traceable.
That era is just about over. And our ability to trust a piece of evidence 100% goes with it. In the near future, an audio or video recording will almost certainly be worthless in court, because anyone can claim it was generated by AI. That’s a time bomb sitting under our justice systems and hardly anybody seems worried about it.
The AI watermark is already an obsolete safeguard
Some people saw AI content tagging as the miracle fix. The idea was simple: stamp an invisible digital signature on every image, video or audio file an AI produces, so you can trace where it came from. On paper it held up.
In reality it’s already game over! With a workstation running two or three decent graphics cards and an open source model, anyone can now generate content that carries no trace of AI at all. The result is fakes that are very hard to detect, made by any bad actor willing to spend money that’s within reach of pretty much everybody. That’s why we’re getting flooded with extremely violent content and nonconsensual porn built from anyone’s face or voice. And sadly, nobody is safe from that kind of manipulation.
Here’s where it gets truly chilling. We’re talking about wrecked lives, reputations destroyed in a few hours and real trauma caused by content that’s 100% manufactured. Women are already victims of porn deepfakes spread without their consent. Minors are being targeted. And the tools for doing it are free, easy to get and getting better all the time.
So what was supposed to be a safety net turned into a sieve. And while institutional circles keep debating how well it works, the lie factory runs at full speed.
Trusting AI like an oracle is a mistake that can cost you dearly
More and more people accept a piece of information because GPT said so. It’s become a reflex, almost automatic. You ask a question, you get an answer delivered with total confidence and you move on. Except that behind that appearance of certainty sits a reality that’s far less reassuring.
Large language models blend sources of wildly uneven quality. So they can spit out a mangled Wikipedia entry because they crossed it with biased material. And above all, they hallucinate! That slightly odd technical word covers a very real thing: the AI makes up facts, sources, entire quotes with the same poise it would use to state an absolute truth. And since the answer always comes out polished, smooth and convincing, nothing in the form tips you off.
But the problem goes way past accidental mistakes. These tools sit in Big Tech’s hands. And the blurrier the line between real and fake gets, the easier it becomes to slip bias, omissions or manipulation into the results without anyone noticing. No need to lie outright. You just frame things, tilt the shading one way and make certain information disappear when it doesn’t serve techno-fascism.
It’s the exact pattern of a toxic partner. At first he’s charming, useful, reassuring and available at any hour. So you trust him. And that trust is precisely what he’ll use later to work you over. By the day you figure out something’s off, you’ll be so used to outsourcing your own thinking that doing without it will be hard. That’s already true for millions of people who think AI is harmless when it’s a tool in very bad hands.
When the disinformation tools belong to libertarians
Let’s talk about Grok, the AI of techno-fascist Elon Musk. His competitors put guardrails in place to limit the worst uses. Musk went the other way. For a while now Grok has allowed deepfake generation, meaning hyper-realistic content built from real people’s faces and voices. And the evil it produces is clearly documented: individuals publicly humiliated through fabricated content, women made the victims of staged porn and clearly established uses for pedophile purposes.
So this isn’t a bug. It’s a deliberate choice! And that choice says something very specific about the man making it. Especially once you know Musk’s name shows up in the Epstein files. We’re not saying that proves guilt. We’re saying it sheds light on how little hurry he’s in to protect the most vulnerable when these tools get designed. Either way, it tells you plenty about the cynical character of this deplorable individual. The guy may be rich in money, but on richness of heart and mind he’s a zero!
Big Tech libertarians see the world in binary. Absolute freedom for them, the consequences of their selfishness for everyone else. And while Grok runs at full throttle, while deepfakes pile up and lives get wrecked, you’re entitled to ask one very simple question. What exactly is Europe doing? Because between the grand declarations on AI regulation and what’s actually happening on the ground, the gap is wide enough to sail an aircraft carrier through. And by the way, why ask what Europe is doing and not America? Simple: on ethics and humanism there’s nothing left to expect from the USA, a country that sank completely into techno-fascism the moment Big Tech took over.
How do authoritarian regimes use AI to impose their propaganda?
Let’s stop talking about future risks and look at what’s already happening right in front of us. Authoritarian regimes didn’t wait for the technology to be perfect before they weaponized it. They’re using it right now, on a huge scale and without the slightest hang-up.
We know that totalitarian propaganda doesn’t need to convince in order to succeed, and that convincing isn’t even its goal. The goal of propaganda is to leave minds discouraged, to persuade every person that they’re powerless to restore the truth around them and that any attempt to fight the spread of lies is useless. George Orwell
Russia turned disinformation into a full weapon of war. Since the invasion of Ukraine, AI-generated content feeding Kremlin propaganda runs into the hundreds of examples. Fake statements attributed to Ukrainian leaders, images of destruction built from nothing to flip the story, doctored videos pushed out at scale on social media… The goal isn’t necessarily to convince everybody. It’s to plant doubt, to drown the true under the false until people no longer know what to believe.
But pointing only at Russia would be a mistake. The United States under Trump crossed red lines too, letting AI-built narratives flourish to lock in a voter base and discredit any opposition. Propaganda no longer belongs to old-school dictatorships. It has settled in comfortably at the heart of facade democracies.
What makes all of this so dangerous is speed. A fake goes around the world in a few hours. The denial always shows up too late, too quietly, and never reaches as many people as the original lie, which was designed to shock. AI handed propaganda firepower nobody’s ever had before. And the regimes that grasped the potential have no intention of stopping. Why would they, when the plan works beautifully and keeps working better? The question’s on the table.
AI and armed conflict, the scenarios that make your blood run cold
What follows, sadly, isn’t science fiction. It’s just the logical extrapolation, and it sits very close to what we’re already seeing today.
One country could invade another, flatten whole cities, slaughter civilians and cover the whole thing in real time with a nonstop feed of AI images showing quiet streets and people going about their day. While the bombs fell, AI would manufacture a clean, comforting parallel reality, pushed out everywhere on social media and swallowed by millions of people with no reason to doubt it. Especially since millions of people are already drowning in tons of contradictory information dumped in among thousands of tons of information and drama that don’t matter in the slightest.
The reverse scenario would be just as chilling. A completely invented attack, built from AI images credible enough to trigger international outrage and used as the pretext for a real invasion. If you doubt it, you don’t have to look far for historical precedent where fabricated pretexts justified wars. Tomorrow those pretexts could be generated by a machine in a few hours. It’s just the next logical step.
Generative AI is an unprecedented gift to conspiracy thinking
Conspiracy theorists have always been around. Until now they cobbled things together with string and a lot of imagination. Now they’ve got very powerful tools that let them make their fantasies solid, with a credibility that was impossible three years ago. Deepfake videos proving a leader said something he never said. AI images supposedly confirming a plot. Anyway, a whole pile of fake documents generated in minutes and blasted out as irrefutable proof.
What makes this stuff dangerous isn’t the technical quality. It’s that it goes all in on emotion. Fear mostly, plus anger, a sense of injustice and the feeling of belonging to an enlightened minority surrounded by sheep. Algorithms love that, because emotion drives engagement. And engagement drives ad money. Conspiracy thinking has become a business model, and Big Tech profits from it handsomely.
And behind most of this content you find very identifiable players. The global far right, feeding brain-dead movements like QAnon with narratives built carefully to reach the most fragile people, the ones looking for simple answers to complicated questions. These movements aren’t spontaneous. They’re built, funded and amplified by networks that understood perfectly that disinformation at scale is a tool of power. Then you have regimes like Russia using conspiracy thinking as a weapon to destabilize democracies by cracking societies open from the inside. The end goal isn’t to get people to sign up for a particular ideology, it’s subtler than that. It’s to wear them out, to disgust them, to make them believe everybody lies. And in the end to push them toward whichever extreme is best equipped to destroy a country from within.
Faced with this avalanche of fakes, the temptation to fight back with the same weapons runs high. That would be a fatal mistake. Answering fake with fake doesn’t win the battle, it gives up the only ground we can still fight on.
Tomorrow, who will decide what’s true and what’s false when it comes to information?
That’s probably the biggest question of the coming decade. And right now nobody is answering it seriously. In the very near future, whose job will it be to help us sort true from false?
Big Tech? Out of the question, they’re judge and defendant at once! They control the platforms, the algorithms and the content generation tools. Handing them the job of certifying truth is asking the fox to guard the henhouse. We’ve already seen how that goes with Meta’s moderation and the arbitrary calls at X under Musk. And Meta recently decided to drop its in-house fact-checkers and replace them with community notes, X-style. In other words, we hand fact-checking to the crowd. The same crowd the algorithms spent years radicalizing.
Politicians? In the democracies still standing, the pull toward an Orwellian ministry of truth is real. But who’d control that ministry? Under what criteria? And what guarantee that official truth wouldn’t simply be the ruling party’s truth? In authoritarian regimes the question doesn’t even come up, we already know the answer.
Sworn experts? Appealing on paper, a lot less so once it hits the ground. Experts contradict each other, they’re fallible, they can be bought, intimidated or simply outrun by the speed information moves at now.
Traditional media? We know their limits, their dependence on advertisers, their concentration in a few big groups’ hands and their taste for the spectacular over the strictly true. Most of them have lost a huge chunk of their credibility and won’t get it back easily.
And independent fact-checkers in all this? The ones doing serious, rigorous, documented work have become targets. They get harassed by hordes of absolute morons and strangled financially, because the work costs a lot and pays little. Meanwhile fake fact-checkers multiply: sites and accounts wearing every marker of serious verification, whose only goal is to validate the disinformation they claim to fight. No need to spell out that cynicism doesn’t get any deeper than this.
Still, the question of a trusted third party is central and urgent. Without a legitimate, independent referee, without a body that can separate true from false with authority most people accept, the strongest will simply impose their will. And right now the strongest is whoever has the biggest disinformation budget.
How do you act right now against AI disinformation?
Waiting for governments to fix this would be a serious mistake. Not that politics has no part to play, it has one and it matters. But legislative time moves slowly, and Big Tech lobbies have plenty of means to stall any regulation or gut it from the inside. Add the speed at which the technology moves. Betting only on a political answer is really very naive.
The real path runs elsewhere. It runs through citizens. And it starts with media literacy. Not as an elective squeezed between two other classes, but as a basic skill on the level of reading and arithmetic. You learn to identify a source, to cross-check, to spot emotional manipulation and to distrust anything that confirms what you already think a little too comfortably. That work has to start in elementary school and keep going for life.
You also have to actively back independent fact-checking networks. We just saw how threatened they are, financially and physically. Backing them isn’t only activism, it’s democratic survival. So subscribe to independent outlets doing serious work. Share what they publish and tell people about them. Every extra supporter is oxygen for organizations that are often hanging on by their fingernails.
And there’s something simpler still: get in the habit of doubting. Not paralyzing doubt or conspiracy doubt, but healthy, methodical doubt. So before you share something, before you believe it, before you repeat it… ask who made it and who benefits.
None of this adds up to a miracle cure. But it’s already a solid start on fighting back against disinformation. And if you’ve got other ideas, drop them in the comments, because we’re really keen to feature them in an upcoming article.
Conclusion: ethics is the key word
The point of this article isn’t to say AI is good or bad. Personally I find it useful enough in science and medicine. And pretty disastrous in content production. Either way, the toothpaste is out of the tube and there’s no getting it back in. So what if the answer fit into one word? The word “ethics.” Individual and collective ethics that would push us to think hard about our relationship with new technology. From that angle the question is simple: do I have to be a victim of technology, or do I make it make my life easier?
The answer looks obvious, putting it into practice much less so. Honestly, I’m not optimistic about what the technological future holds. But if one thing can give me hope back, it’s the thought that most of us, deep down, are fundamentally attached to truth rather than to lies. Even if reality, sadly, is as cruel as this quote:
In a time of universal deceit, telling the truth is a revolutionary act. George Orwell
On our end, sure, we’re a tiny outlet next to the big machines polluting minds on an industrial scale. That’s a fact. But at least we do the job honestly and hold ourselves to as much professionalism as we can manage. That counts for something 🙂 So if you want us to go further, it’s not complicated. Start by sharing this article and back free media. Not out of charity, but to help put uncomfortable truths in the public square. Thanks for reading this far and see you very soon for more adventures. And to close, let’s tip our hat to the talented George Orwell, who kept us company the whole way through.
If you enjoyed this article,
support copyleft and open source
and help us bring our projects to life.
