Generated Iranian protest videos: When information vacuum meets digital illusion

In January 2026, a video went viral on social media. In the footage, thousands of people holding glowing phones formed a sea of light, marching through dark streets. Overlaid text read: All eyes are on Iran. The video spread rapidly across X, TikTok, Instagram, Facebook, and Reddit, with single posts reaching hundreds of thousands of views and estimated cumulative views totaling tens of millions. It captured the world's gaze on the turmoil within Iran, delivering a powerful emotional impact.

However, this breathtaking scene is not real. An investigation by German ZDFheute revealed that the video is entirely an artificial intelligence creation. Through reverse image search, investigators traced it back to its source—an Instagram user described it as a digital tribute to the current protests in Iran. The creator admitted that the inspiration came from a real video clip showing protesters on the dark streets of Tehran, and they attempted to artistically recreate that scene using AI technology. From a technical perspective, this is a successful digital art creation; from the standpoint of information dissemination, it is a depth charge dropped into the deep waters of public opinion.

This video is just the tip of the iceberg. As Iranian authorities implemented a severe internet blockade to suppress nationwide protests that erupted at the end of the year, the outflow of truthful information became extremely difficult. According to data from the Human Rights Activists News Agency, by mid-December, over [number] protesters had been killed and [number] arrested. It is in this near-total information vacuum that AI-generated videos and images flooded social networks on an unprecedented scale and speed, being utilized by both anti-government and pro-government forces in an attempt to shape conflicting narratives and compete for the attention and perception of a global audience.

Deepfake Wave: A Multi-Faceted Digital Propaganda War

Generated content is not a new phenomenon, but its application in the crisis in Iran has taken on new dimensions and scale. In a report dated [Month] [Year], a U.S. disinformation monitoring agency noted that they had identified at least seven generated videos depicting protests in Iran, which collectively garnered approximately [number] million views across various platforms within a few days. This content does not originate from a single source but rather forms a multi-voiced, and at times even contradictory, digital public opinion sphere.

AI Tools for Anti-Government Narratives. A widely circulated AI video on platform X shows female protesters smashing vehicles belonging to Iran's paramilitary Basij force. The caption reads: Free Iranian women are crushing the machinery of the Basij militia. By God, these women are braver and more honorable than... the militia loyal to the Khamenei regime. The post garnered nearly 720,000 views. Although users later pointed out that the glass fragments in the video appeared abruptly, suggesting it might be AI-generated, its dissemination impact had already been achieved.

More symbolically, some anti-government users, particularly those in the United States, created and disseminated AI-generated videos showing Iranian protesters symbolically renaming streets as Trump Street. One clip depicted a protester changing a street sign to Trump Street, with other demonstrators cheering nearby. Overlaid subtitles read: Iranian protesters are renaming streets after Trump. This clearly echoed and amplified the high-profile statements made by former U.S. President Trump at the time, who expressed strong support for the Iranian protesters. Trump repeatedly stated his intention to assist the Iranian people and mentioned that he had been informed the killings of protesters had stopped, but he would closely monitor the situation and not rule out the possibility of military action. These AI videos combined political symbols with virtual scenarios, aiming to reinforce a narrative that the Iranian people are oriented toward the West.

Digital Counterattack by Pro-Government Forces. Meanwhile, social media users supporting the Tehran regime are also utilizing the same technology. They share AI-generated videos claiming to show large-scale pro-government counter-demonstrations taking place across the Islamic Republic. For instance, some videos attempt to depict hundreds of thousands, rather than the tens of thousands reported by the Associated Press, of pro-government supporters marching in Tehran. This operation aims to counter international attention on the scale of the protests, shape the perception of widespread regime support domestically, and fill the narrative gap left by the scarcity of authentic footage.

From Crafty Substitution to Creation from Nothing: The Evolution. What is even more alarming is that AI generation technology has elevated the production of false information from simple tactics like repackaging old news or altering locations to complete scene hallucination. For instance, a video from several months ago claimed to show protests in Iran, but was actually filmed in Greece in November 2025 (reportedly depicting a conflict scene after a concert in Thessaloniki). Another video, which claimed to depict protesters tearing an Iranian flag, was actually filmed during protests in Nepal last year. These are still misuses based on real footage. However, the aforementioned "Sea of Light" protest march, along with another widely circulated image of a protester standing on a statue waving the old Iranian flag, were generated entirely from scratch. That highly evocative statue image, when tested by Google's SynthID tool, was confirmed to contain the invisible watermark characteristic of AI-generated content. The investigation revealed it was based on a screenshot from a real video, enhanced through AI technology: adding thick smoke, dramatic lighting, more protesters, and a clearer old flag, thereby creating a more impactful and symbolic news image.

The Dramatic Shift in the Information Ecosystem: How a Vacuum Gives Rise to Illusions

Iran's internet blockade has created a nearly perfect information black hole. Ines Chaumunarez, an analyst at NewsGuard, pointed out sharply: There is a lot of news—but due to the internet blockade, it cannot be accessed. This situation has completely changed the dynamics of information dissemination during crisis events.

The Failure of Traditional Verification Mechanisms. In past regional conflicts or social unrest, despite the presence of misinformation, journalists, activists, and ordinary citizens could still transmit some cross-verified visual materials through intermittent network connections, satellite phones, or covert means. These authentic materials formed the basis for the international community's understanding of events. However, when Iranian authorities implemented a comprehensive and severe internet blackout, this information supply chain was fundamentally severed. Professional news organizations and fact-checking groups lost their channels for on-site verification and were forced to rely heavily on limited, difficult-to-verify information sources.

The Filling Instinct of Social Media. Social media platforms' information feeds abhor a vacuum. When users crave to know what is happening in Iran and traditional sources cannot satisfy this demand, the platform's content ecosystem spontaneously generates content to fill the need. The proliferation of AI video generation tools has unprecedentedly lowered the barrier to such production. Any user with relevant software and creativity (or intent) can create seemingly professional, emotionally charged on-scene footage in a short time. Due to their visual authenticity and narrative resonance, such content easily gains algorithmic recommendations and user shares, thereby spreading rapidly.

Emotional Resonance Precedes Fact-Checking. In the rapidly scrolling information flow, users are often emotionally moved first, and only later might (and typically do not) engage in rational verification. The AI-generated video of the light-sea parade garnered tens of millions of views precisely because it accurately captured and externalized the global audience's sympathy and solidarity with the Iranian people. The slogan "All Eyes on Iran" itself serves as an emotional mobilization. In such situations, fact-checking often lags, and even when verification results are announced, their reach is far less extensive than that of the original false content. As the fact-checking department of India's Mathrubhumi newspaper discovered, the light-sea video was judged by AI detection tools to have a 97.9% probability of being AI-generated, but this conclusion has limited corrective effect on the already widespread impression.

Deep Impact: Erosion of Trust and Cognitive Warfare

The proliferation of generated videos in the context of protests in Iran has implications far beyond the spread of a few pieces of false content. It marks a new and more dangerous phase in information warfare and cognitive warfare in the digital age.

A Fundamental Challenge to News Authenticity. The age-old belief that "seeing is believing" has crumbled in the face of deepfake technology. When AI can generate realistic on-site footage, the public's foundational trust in all visual evidence is shaken. What experts call hallucinatory visual content is becoming increasingly common in major news events, often overshadowing genuine images and videos. Over time, this could lead to widespread skepticism, where people may start to distrust any imagery not verified through extremely complex means, or conversely, fall into a post-truth apathy, choosing only to believe narratives that align with their preferences, regardless of their veracity.

A New Tool in Geopolitical Gamesmanship. These AI videos do not exist in isolation; they are embedded within a broader geopolitical narrative. Anti-government content echoes the rhetoric of Western (particularly American) political figures, aiming to create the perception that the Iranian regime has lost popular support and that change is imminent. Pro-government content seeks to reinforce the regime's legitimacy and counter international pressure. This is essentially a cognitive domain operation conducted using cutting-edge digital technology, with the global social media landscape as its battlefield and the goal of influencing international public opinion and potential policy directions. The unsettling scenario depicted in the film "The Summit," where a tech giant's platform sows global chaos and violence through user-generated incendiary videos, finds a disturbing parallel in reality.

Secondary Harm to the Public in Crisis. For people within Iran, struggling under information blackouts and real-world suppression, as well as their relatives and friends overseas, these AI videos that are difficult to distinguish between real and fake may cause confusion, misdirection, or even false hope. They could distort the outside world's understanding of the severity of the situation and might also be used by the authorities as an excuse for further suppression and accusations against foreign media for incitement. Authentic voices are drowned in a wave of digital illusions, and their suffering and demands are, to some extent, sensationalized and instrumentalized.

Where Lies the Path Ahead: Finding an Anchor in the Technological Fog

Facing the challenges brought by generated content, there is no simple solution, but some directions are emerging.

The Arms Race in Technical Defense. Developing more powerful AI detection tools (such as Google's SynthID) is crucial, but this is an ongoing cat-and-mouse game. As generation technology advances, detection technology must develop in sync or even ahead. Platforms need to integrate these tools more deeply into their content moderation workflows and prominently label content suspected to be AI-generated.

Reiteration and Strengthening of Platform Responsibilities. Social media platforms can no longer continue to act as passive, neutral conduits. They need to formulate and strictly enforce policies targeting AI-generated fake news content, especially in high-risk contexts such as armed conflicts, humanitarian crises, and elections. This includes accelerating the verification and handling speed of suspected harmful AI content, as well as adjusting algorithms to avoid excessively promoting highly emotional but dubious content.

Nationwide Upgrade of Media Literacy. Public education has never been more important. It is essential to teach users how to maintain a healthy skepticism towards astonishing visual content, how to seek multiple sources for cross-verification, and how to identify potential subtle flaws in AI-generated content (such as unnatural hand movements, details that defy physical laws, overly perfect lighting and shadows, etc.). News organizations and educational institutions should take on greater responsibility in this regard.

The Irreplaceable Value of Professional Journalism. Amidst the information chaos, the value of professional, rigorous, and ethically-guided news reporting becomes even more prominent. Despite facing immense difficulties, agencies like Reuters, the Associated Press, Agence France-Presse, and their fact-checking departments are still striving to verify information through limited channels and provide validated reports. Supporting independent journalism, especially local journalism operating under high-pressure conditions, is fundamental infrastructure for resisting information pollution.

Iran's internet may not have been restored yet, but the future revealed by AI-generated protest videos has already arrived. We are entering an era where seeing is no longer believing, a time when the information environment can be easily polluted. This digital fog, which began in the dark streets of Tehran, will eventually envelop us all. The solution lies not in nostalgically yearning for the authenticity of the past, but in building a more resilient system of information verification, a framework for platform accountability, and public cognitive capabilities adapted to the new technological reality. In this prolonged battle against digital illusions, maintaining a clear mind may be our most fundamental line of defense.