Every single day, hundreds of thousands of items of faux content material are produced. Movies, audio clips, posts, articles, generated by synthetic intelligence, distributed at industrial scale, geared toward shifting public opinion throughout total nations. The individuals producing them are sometimes exterior the nation being focused. The individuals receiving them virtually by no means know they’re faux. They usually don’t know how they’re made.
Just a few years in the past, troll farms labored like this: total buildings full of individuals, shifts, desks and employees paid to jot down posts, create faux profiles, remark and decide fights in on-line discussions. It was costly, sluggish, and ultimately, the actual affect was marginal. These buildings nonetheless exist in the present day, principally in India, cut up between groups specializing in scams and groups devoted to disinformation. They work on fee they usually’re principally AI consultants now. They now not write the articles themselves and now not do graphic design or picture enhancing. They’ve AI brokers do the whole lot: brokers they create, configure, instruct, and supervise. Lots of of 1000’s of autonomous brokers that do in a single hour what used to take weeks of human labor. Troll farms have turn out to be AI farms, producing artificial content material at industrial scale.
The report “From Trolls to Generative AI: Russia’s Disinformation Evolution,” revealed in February of 2026 by the Centre for Worldwide Governance Innovation (CIGI), tells considered one of these tales, particularly about disinformation campaigns originating from Russia. Networks like CopyCop, a disinformation operation linked to the GRU (Russian army intelligence), use uncensored open-source language fashions like modified variations of Llama 3, put in on their very own servers, to remodel press articles into political propaganda and distribute it throughout tons of of faux web sites with out leaving a hint. As a result of the fashions run regionally, there’s no watermark and no log. The mannequin runs on their {hardware}, inside their borders, exterior any Western jurisdiction.
The paper “How malicious AI swarms can threaten democracy,” revealed in Science in January 2026 describes nicely what’s coming: coordinated swarms of AI brokers with persistent identities, reminiscence, and the power to adapt in actual time to individuals’s reactions. The authors name them “malicious AI swarms.” Absolutely autonomous brokers, every producing unique content material, each totally different, each tailored to context.
They’ll simulate actual communities that seem credible, they usually construct what we will name artificial consensus: the phantasm that an opinion is broadly shared, {that a} place is held by the bulk, when in actuality it’s a single operator talking by way of 1000’s of masks.
It really works as a result of we people have bugs too, and the swarms exploit them at a scale that was by no means attainable earlier than or that might have required huge human assets.
One bug is known as the bandwagon impact. Mixed with one other bug, illusory fact: repetition plus obvious supply independence equals perceived fact. So if we see the identical place expressed by totally different sources, in numerous contexts, with totally different phrases, on totally different platforms, we register it as widespread. And if we understand it as widespread, we think about it extra credible. And if we think about it credible, we are likely to align with it.
Swarms of autonomous brokers exploit each mechanisms on the similar time, at industrial scale.
What most individuals nonetheless haven’t grasped is the dimensions. We had been used to automation: A system that despatched 100 thousand an identical emails, at most altering the identify and little else, or made simply as many posts and related feedback with minor variations. It automated the publishing, however at its core it was recognizable spam. Our psychological mannequin remains to be that one: If it’s automated, it’s generic. If it’s generic, you may spot it. However that’s a notion error constructed on years of expertise when AI brokers didn’t exist. That mannequin is over. These brokers now not match the idea of automation, as a result of they make selections, they seriously change the textual content based mostly on the recipient. They mixture information from heterogeneous sources in actual time: social profiles, public data, leaked databases you could now purchase for a couple of {dollars} on any darkish internet market. Billions of non-public data are already on the market, scattered throughout tons of of breaches collected through the years, and AI can cross-reference them, reconcile them, and construct a coherent profile of a single individual in seconds. The computational price is negligible: a couple of cents in tokens to generate a wonderfully customized message. Take into account {that a} single agent with entry to a language mannequin and a few leak databases can produce 1000’s of distinctive items of content material per day, every calibrated for a unique individual. Multiply that by 100 thousand brokers working in parallel, twenty-four hours a day, and you’ve got the dimensions of what’s occurring.
One other legacy from the previous: “I’m simply an odd individual, why would anybody hassle creating content material particularly to persuade me?” Which will have been as soon as true. Right now, no person is dropping time as a result of these brokers don’t get drained, don’t sleep, and do nothing else: discover connections, mixture information, produce false content material calibrated for every of us. The previous demographic profiling is over. That is surgical media focusing on at industrial scale.
However the capability to reply and deny just isn’t at industrial scale. If tons of of 1000’s of coordinated brokers unfold a video of a politician saying one thing they by no means mentioned, that politician can deny all of it they need. The video is there. Tens of millions of individuals have seen it. The denial arrives later, arrives slower, and can by no means attain the identical scale. It arrives in a world the place no person is aware of what’s true anymore.
If the identical swarms unfold the information {that a} head of state has died, and the information is fake, that head of state could make all of the movies they wish to show they’re alive. These movies will most likely be dismissed as deepfakes. As a result of the swarm’s narrative acquired there first, took root, and at that time any proof on the contrary seems fabricated.
Whoever controls the swarms in the present day controls the model of the info. Whoever tries to push again is already at an obstacle as a result of they need to show that an actual video is actual in a world the place everybody has discovered that movies might be faux.
The attackers are sometimes exterior the nation being hit. Teams aligned with governments that wish to shift public opinion in a foreign country, or that focus on particular demographics. Younger individuals, for instance, utilizing platforms which might be usually owned by these very nations.
All of it is a huge menace to democracy as a result of democracy operates on some premises, together with that individuals type opinions based mostly on actual info, focus on with one another, after which determine. If the knowledge is fabricated, if the controversy is populated by entities that don’t exist, if the consensus we understand is artificial, that premise collapses. And with it, all the mechanism. Elections turn out to be the results of who has one of the best swarms, not who has one of the best concepts. Public debate turns into a efficiency the place a lot of the voices are generated, and public opinion stops being public and turns into the product of whoever has the assets to fabricate it.
We grew up considering that threats to democracy got here from coups, censorship, or regime propaganda broadcast on tv or in nationwide newspapers. These had been actual threats, however they had been not less than seen. They had been issues you might determine and struggle. Now the menace is greater and, above all, invisible, customized, and it operates contained in the very channels we use to tell ourselves, to debate, to take part. It contaminates info from inside, to the purpose the place no person is aware of which voices are actual and that are machines.
What can we do? Watermarking? Sample detection? Sadly, they don’t work. The main AI platforms can embed markers in content material generated by their fashions, true. However the individuals constructing autonomous swarms don’t use industrial platforms. They use open-source fashions with fine-tuning and capabilities that may’t be managed from exterior. They usually usually don’t have any authorized obligation to do something as a result of there aren’t any world legal guidelines that may impose watermarking on each pc on the earth. The result’s paradoxical: The content material produced by those that observe the principles stays marked, and the content material produced by those that wish to trigger hurt stays free.
Sample detection methods have the identical limits. They work for some time, then as soon as the detection patterns are recognized, the swarms adapt. They’re designed to do precisely that.
And the platforms the place all of this circulates have a monetary incentive to show a blind eye. Inner Meta paperwork made public by Reuters in November 2025 estimated that roughly 10% of Meta’s world 2024 income, about $16 billion, got here from promoting for scams and prohibited merchandise. Fifteen billion high-risk adverts served on common day-after-day to customers. The utmost income Meta was keen to sacrifice to behave towards suspicious advertisers was 0.15% of whole income: $135 million out of $90 billion. When a platform’s enterprise mannequin is dependent upon advert quantity, eradicating the fraudulent ones has a price that no person needs to pay. I think Meta just isn’t alone on this.
Regulation doesn’t remedy this downside both. I’ve labored on the European AI framework, the GPAI job pressure, the Italian AI regulation, and I’ve introduced my perspective to the UK Parliament. I’ve been in these rooms. Europe has the AI Act, the GPAI Code of Observe is at the moment being drafted, and has a regulatory equipment that’s extra superior than every other bloc on the earth. America has no federal regulation, and twenty-eight states have tried to legislate with transparency necessities that quantity to tremendous print. However even essentially the most bold European framework has a structural restrict: The assaults come from nations that reply to none of those guidelines. You possibly can regulate your platforms, your builders, your corporations. You possibly can’t regulate a constructing in Saint Petersburg, Shenzhen, or New Delhi, the place somebody is instructing swarms of brokers on open-source fashions working on native servers, exterior any jurisdiction.
A method out is to return to the popularity of sources. Editors, information organizations, journalists with a reputation and a face. Folks and organizations which have an expert observe report to defend and that danger one thing after they get it mistaken. Certain, they’ll have political leanings they usually could make errors. However they’ve a constraint that no AI agent will ever have: public accountability. A system that generates hundreds of thousands of items of false content material solutions to nobody. An editor solutions to their viewers, to the regulation, to their popularity. That constraint is the one filter that also holds, and defending it’s the solely factor we will do proper now, whereas the legal guidelines attempt to meet up with a know-how that strikes quicker than any legislative course of on the earth.
Are we fully on the mercy of AI swarms or can we struggle again?
Machines shouldn’t get to overpower people, particularly when what’s at stake is how we govern ourselves. The antibodies exist. We have to activate them.
The extra individuals perceive how swarms work, the much less efficient they turn out to be. A swarm that manufactures faux consensus solely works if the individuals receiving it don’t know artificial consensus exists. A bit like deepfakes. We learn about them now and we frequently spot them. When you see the way it works, it’s more durable to fall for it.
Then we want funding in tradition. In spreading digital literacy, which isn’t studying how you can use a pc, however studying to know the social and cultural results of the digital world. It means instructing in colleges how you can confirm a supply and what the indicators of manipulated content material are. It means stopping the apply of treating media literacy as a faculty undertaking and beginning to deal with it as democratic infrastructure, on the identical degree as bridges and hospitals. It means funding impartial journalism as a substitute of letting it die, strangled by the identical mechanisms that reward false content material as a result of it generates extra engagement. It means demanding that platforms give totally different visibility to those that have a verifiable popularity versus those that have none.
As a result of consciousness is the one antibody that scales on the similar velocity because the menace. And in contrast to regulation or detection methods, consciousness doesn’t should be imposed. It may be constructed, taught, shared, and unfold from individual to individual.
Earlier than sharing a bit of content material, verify the place it comes from. Earlier than reacting to a video or an announcement, cease. Ask your self whether or not the supply has a reputation, a historical past, one thing to lose. Deal with each piece of content material as doubtlessly artificial till a reputable, accountable supply confirms it. These are habits, not applied sciences. They price nothing they usually work instantly.
Lastly, we want the assistance and collaboration of the tech neighborhood. Those that design platforms, write code, and make selections about how feeds and rating algorithms work are making decisions that immediately form the knowledge ecosystem. These are decisions with democratic penalties. The individuals making them comprehend it. Many have recognized it for years. That is the second to cease treating it as another person’s downside and to determine which facet you’re on. As a result of the swarms aren’t ready.
We will do that. The instruments exist, the information is there, and the menace is evident sufficient that pretending to not see it’s already a alternative. The query is whether or not we act now, whereas the window remains to be open, or later, when the injury shall be more durable to reverse.
