Pro-Iran groups use artificial intelligence to create anti-Trump memes and shape the war narrative
Artificial intelligence has become a weapon of war — and we are not talking about drones or missiles.
While armed conflicts dominate headlines around the world, a parallel battle is playing out on the phone screens of millions of people. Groups linked to the Iranian government are using AI to produce sophisticated memes in English, with one clear goal: shape the conflict narrative against the United States and Israel, and stir up popular opposition in the West. 🎯
Analysts say these memes appear to come from groups connected to the Tehran government and are part of a strategy to maximize limited resources and inflict damage on the US, even if indirectly. That logic is similar to the way Iran uses attacks and threats to control the flow of traffic through the Strait of Hormuz and keep pressure on the global economy.
And the most surprising part of all this is not just the technology behind the content — it is how well they understand American culture from the inside.
References to US domestic politics, pop culture, even the visual style of the Lego movies. All used with a precision that has researchers on high alert.
These creations are racking up millions of views on social media, although it is still unclear how much real influence they have on public opinion. The question nobody can answer with confidence is: how far can this kind of digital propaganda actually go?
How AI is being used to create propaganda in meme format
For a long time, the idea of digital propaganda conjured up images of poorly translated text, obvious grammatical errors, and low-quality images that anyone could spot as foreign-made content. Those days are over. Generative artificial intelligence tools have completely changed the game, and what digital security researchers are documenting now is something very different: visual and textual pieces so well calibrated for American audiences that they easily pass as organic content created by US citizens themselves.
Neil Lavie-Driver, an AI researcher at the University of Cambridge, left no room for doubt when classifying what is happening. According to him, referring to Iran, this is a propaganda war for them. The goal, he said, is to sow enough discord around the conflict to eventually force the West to back down — and that is why this front matters so much to Tehran.
What stands out in the identified materials is the layer of cultural knowledge embedded in every detail. It is not enough to just speak fluent English — you need to understand the rhythm of American humor, the references that resonate with different demographics, the right timing to post certain types of content, and even which visual aesthetic will perform best on each platform. And that is exactly what is being delivered by AI systems that have been trained or guided to mimic American culture on social media with surgical precision.
Nancy Snow, a scholar who has written more than a dozen books on propaganda, summed up the strategy in a straightforward way: they are using popular culture against the number one country of pop culture, the United States. 😮
Receive the best innovation content in your email.
All the news, tips, trends, and resources you're looking for, delivered to your inbox.
By subscribing to the newsletter, you agree to receive communications from Método Viral. We are committed to always protecting and respecting your privacy.
The memes go deep into American culture
The memes are not just fluent in English — they are fluent in the American cultural universe and in the art of digital trolling. Published across multiple social platforms, these pieces have racked up millions of views and demonstrate an impressive command of the visual and political vocabulary of the US.
The content portrayed President Donald Trump as aged, disconnected, and internationally isolated. They referenced bruises on the back of Trump’s right hand that sparked speculation about his health, internal disputes within the MAGA base, and even the contentious confirmation hearing for Defense Secretary Pete Hegseth.
Among the most notable examples is a series of animations in the style of the Lego movies. In one of them, an Iranian military commander raps something like you thought you ruled the world sitting on your throne, now we are turning every base to stone, while Trump falls into a target built with the so-called Epstein files — the US government investigative records on financier Jeffrey Epstein.
The level of sophistication in these productions did not go unnoticed. Mahsa Alimardani, director of WITNESS, a human rights organization that works with AI-generated video evidence, pointed out that the animations demonstrate levels of sophistication and internet access that indicate ties to government offices. According to her, if you can have the bandwidth needed to generate content like this and upload it, you are officially or unofficially cooperating with the regime — a reference to the severe internet restrictions Iran has imposed as part of a crackdown on national protests.
Who is behind the memes and what is the connection to the Iranian government
Iranian state media reposted some of these memes, including material from the account behind the Lego-style videos, called Akhbar Enfejari, which means Explosive News.
When contacted by the Associated Press through the Telegram messaging app, the Akhbar Enfejari group described itself as an independent collective of Iranians with no connection to the government. According to the group, they receive no funding whatsoever — they are just a group of friends working voluntarily, paying for their own internet, using their own laptops and computers, and doing everything on their own.
The group said it produces and publishes content from inside Iran to try to break decades of Western dominance over media narratives. In their own words, they have dominated the media scene for a long time and, through that power, imposed narratives on many nations. But this time something feels different. This time, we interrupted the game. This time, we are doing it better.
Beyond the memes coming from pro-Iran groups, official Iranian government accounts have also jumped into the digital trolling. On Wednesday, the Iranian embassy in South Africa published a post featuring a photo of the Iranian flag and the phrase say hello to the new world superpower. Both the US and Iran declared victory after agreeing to a ceasefire. ⚡
Analysts point out that this deep command of American politics and culture is not solely a product of AI, but also comes from more traditional propaganda methods: an Iranian government program that has existed for decades and is dedicated to promoting narratives against the US and Israel.
As Alimardani explained, this meme war comes from institutions that are very aware of what the American public knows and of pop culture references that can attract that audience.
What the United States and Israel are doing on this front
Analysts point out that the US and Israel do not appear to be engaged in the same kind of campaign targeting the Iranian public — and given the restrictions Iran has imposed on internet access, getting that kind of message to ordinary Iranians would be very difficult.
At the start of the conflict that began on February 28 with joint US-Israel strikes, Israeli Prime Minister Benjamin Netanyahu released a video that used AI to simulate him speaking in Farsi, in which he urged Iranians to overthrow their government. The White House, for its part, has been publishing a steady stream of memes, but those are aimed at the American public and use clips from US TV shows and sports.
Voice of America, the US government news service that for decades has broadcast reporting to countries without a tradition of a free press, still broadcasts in Farsi. However, it has been operating with a skeleton crew since Trump ordered its shutdown.
Nancy Snow made an observation that sums up the geopolitical moment nicely: this world order is really changing overnight, and the US is not necessarily going to end up as the state that everyone listens to.
This is not the first time memes have been used in conflicts
It is worth remembering that the use of memes in conflicts is nothing new. The practice has evolved in recent years to include images generated by artificial intelligence. In 2022, after the Russian invasion of Ukraine, AI-produced images bombarded Ukrainians. Last year, the term AI slop became popular to describe the flood of imperfect images published online during the Israel-Iran conflict related to the nuclear program.
What sets the current moment apart is the quality and sophistication. Generative AI tools have advanced to the point where distinguishing fabricated content from authentic content has become much harder. And when we are talking about memes — a format that already circulates without source verification by nature — that difficulty multiplies exponentially. 🔍
Why memes are the perfect vehicle for this kind of conflict
The meme as a format has a characteristic that no other type of content can replicate with the same efficiency: it spreads on its own. People share memes because they find them funny, because they agree with the message, or because they want to be part of a larger cultural conversation. Nobody stops to verify the source of a meme before hitting the share button. And it is exactly that dynamic that makes the format so powerful — and so dangerous — when put to work for an organized propaganda operation within a geopolitical conflict.
In the context of an international conflict, the meme works like a cultural precision missile. It does not need to convince everyone — it just needs to create doubt, amplify divisions that already exist, and make certain groups feel that their beliefs are being validated by others around them. When you see a meme that reflects exactly what you think about an issue, the tendency is to assume it came from someone like you. It is a natural cognitive bias, and it is that bias that sophisticated disinformation operations exploit masterfully.
With artificial intelligence generating this content at industrial scale, the volume of material circulating completely changes the equation. We are not talking about a few dozen posts a week — we are talking about continuous, adaptable, and customizable production, capable of responding to events in real time and adjusting the tone based on the political temperature of the moment. This creates a propaganda ecosystem that is much harder to track, identify, and fight than any previous disinformation campaign.
What researchers and platforms are doing about it
Detecting AI-generated content in service of foreign influence operations is still a rapidly developing field. Digital security companies, disinformation research labs, and social media platforms themselves are investing in tools capable of identifying patterns that indicate automated production at scale — things like publishing speed, similarity between accounts, repetitive linguistic patterns, and coordinated distribution behavior.
But the artificial intelligence used to create this content is also evolving fast, which makes the technological arms race between creators and detectors extremely dynamic with no clear winner on the horizon.
Organizations like the Stanford Internet Observatory, the Atlantic Council’s Digital Forensic Research Lab, and independent fact-checking groups have been documenting increasingly sophisticated influence operations over the past few years. What changes now is the speed and quality at which this content is being produced. Before, it was possible to identify an Iranian or Russian influence operation by the linguistic patterns in the text, cultural errors, or the awkwardness of the references. With well-configured generative AI tools, those markers are disappearing, and fabricated content is becoming indistinguishable from the real thing for most people.
Platforms, for their part, have adopted removal and reach-reduction policies for accounts identified as part of coordinated influence operations. Meta, X (formerly Twitter), and YouTube have already published transparency reports describing the takedown of account networks linked to foreign governments. The challenge, however, remains the same: identifying these networks before the content has already circulated enough to have a real impact on public opinion. And at the pace artificial intelligence is advancing, that gap between creation and detection remains one of the great dilemmas of modern digital security. ⚠️
What stays after the meme disappears from the screen
Perhaps the most concerning aspect of this new era of digital propaganda is not the content itself, but the cumulative effect it produces over time. A single meme does not change anyone’s mind. But millions of memes, distributed strategically, at the right moments, to specific audiences, within platforms that already favor content that generates emotional engagement — that creates an environment where certain narratives become familiar, comfortable, and eventually accepted as truth by large segments of the population.
In the specific case of materials identified as part of operations linked to the Iranian government, the focus has been on amplifying American opposition to military interventions in the Middle East, creating distrust toward US allies in the region, and reinforcing internal political divisions that already exist organically in American society. None of this is made up — these are real tensions, exploited with a new skill that comes directly from the ability of artificial intelligence tools to understand and imitate American culture in depth.
What changes, at the end of the day, is the scale and precision. Propaganda has always existed. The use of narratives to influence populations during times of conflict is as old as war itself. What is new here is the ability to do it with unprecedented efficiency, at an extremely low cost, without needing large teams or physical resources — just access to the right tools and the knowledge of how to use them. And that knowledge, unfortunately, is becoming more accessible every day. 🌐
Understanding how this process works is already an important first step. Knowing that a funny meme about international politics might have come from a very different place than it appears is the kind of digital awareness that makes a difference when deciding what to share — and what to question.
