AI-generated war footage, satellite images and a fake Epstein video flood social media during the 2026 Iran war

AI-Generated Synthetic Media Overwhelms Verification During the 2026 Iran War

The opening weeks of the 2026 Iran war witnessed an unprecedented surge in synthetic media, including AI-generated battle footage, doctored satellite images, and Gemini-produced images of captured soldiers. These fabricated visuals spread across social media and messaging platforms faster than fact-checkers and platform moderators could verify them, establishing generative video as a primary vector for wartime misinformation. The conflict marked the first major war where such technology played a central role in distorting public perception and amplifying disinformation.

The rapid dissemination of these materials overwhelmed traditional verification processes. Fact-checking organizations and social media platforms struggled to keep pace, allowing false narratives to dominate discussions about the conflict. The sheer volume and speed of AI-generated content created an environment where misinformation could spread uncontrollably, undermining trust in both digital and traditional media sources.

Platforms like X (Twitter), Facebook, and TikTok became key channels for the distribution of these synthetic materials. Elon Musk’s X platform, in particular, faced criticism for allowing users to generate and share AI images of war zones, despite its own content policies. The platform’s relaxed moderation during this period contributed to the proliferation of fabricated imagery, including claims of American soldiers being captured by the Iranian Revolutionary Guard Corps (IRGC).

Key Examples of AI-Generated Disinformation

• AI-Generated Battle Footage: Purported videos showing nonexistent wartime scenes, such as explosions and destroyed city streets, circulated widely. These videos were designed to manipulate public opinion by exaggerating the scale of destruction or misrepresenting the conflict’s progression. The OECD AI Incidents and Hazards Monitor classified these incidents under the domain of Misinformation and Malicious Actors & Misuse, highlighting their role in distorting wartime narratives.
• Doctored Satellite Images: A manipulated satellite-style image claimed to show U.S. radar equipment in Qatar destroyed by an Iranian strike. The image was later debunked as a fabrication derived from older Google Earth imagery of a U.S. site in Bahrain, not Qatar. This false claim gained millions of views and was amplified by Iranian state-aligned media outlets like Tehran Times, further eroding trust in open-source intelligence and fact-checking efforts. The incident was categorized under Epistemic Integrity and Wartime Information Ecosystem harms.
• Gemini-Generated Captive Soldier Images: Images purportedly generated by Google’s Gemini AI depicted U.S. Delta Force soldiers captured by the IRGC. These images contained detectable markers and visual inconsistencies, yet they spread rapidly across X and Facebook in multiple languages. The Associated Press (AFP) confirmed their AI origin, noting that their circulation risked distorting public understanding of the conflict. The amplification of these images by pro-Iran accounts, including political figures like Jill Stein, underscored the role of coordinated influence operations in spreading disinformation.
• AI-Generated Jeffrey Epstein Video: Pro-Iran accounts on X, including HDX News and GPX News, circulated a fabricated video featuring Jeffrey Epstein while falsely claiming it was real. The video was used to suggest that U.S.-Israeli strikes on Iran were launched as a distraction from Epstein-related controversies. The posts garnered over 11.5 million views combined before the accounts were suspended. This incident was linked to Discrimination & Toxicity and Epistemic Integrity, as it targeted specific groups and reinforced conspiracy theories to undermine trust in geopolitical narratives.

Mechanisms of Dissemination and Harm

The rapid spread of AI-generated disinformation was facilitated by several factors, including the accessibility of generative AI tools, the lack of robust verification protocols, and the strategic use of social media algorithms. State-aligned and pro-Iran accounts exploited these weaknesses to amplify false narratives, often tailoring content to resonate with specific audiences across different languages and regions.

Social media platforms played a dual role in this crisis. Additionally, the use of AI-generated content in wartime created a feedback loop: false claims gained traction, prompting fact-checkers to debunk them, which in turn fueled further skepticism toward legitimate reporting.

The harm extended beyond immediate misinformation. The erosion of epistemic integrity*—the trust in information itself—had long-term consequences for democratic discourse and national security. Open-source intelligence researchers and fact-checkers were also impacted, as their ability to verify information was compromised by the volume of AI-generated content.

The incidents also revealed vulnerabilities in AI detection tools. While some synthetic media contained detectable artifacts (e.g., Gemini markers), others were designed to evade scrutiny. This cat-and-mouse dynamic between creators of disinformation and those tasked with verification underscored the need for more advanced detection methodologies, particularly in high-stakes environments like wartime.

Actors and Motivations Behind the Disinformation Campaigns

The primary actors behind these AI-generated disinformation efforts included state-aligned media outlets, pro-Iran information manipulation networks, and anonymous accounts on platforms like X. These groups leveraged generative AI to create content that aligned with their geopolitical agendas, often targeting audiences in the Middle East, the U.S., and Europe.

Key entities involved in the dissemination of synthetic media included:

• Tehran Times and Pro-Iran Accounts: These outlets amplified doctored satellite images and AI-generated videos to portray Iran as victorious or to justify its actions. Their content often framed the conflict in ways that served Iranian narratives, such as claiming U.S. military defeats or internal divisions.
• Their posts were designed to distract from Iranian military setbacks or to exploit sensitivities around Western figures.
• Information Manipulation Actors: Coordinated networks exploited social media algorithms to maximize the reach of fabricated content. By reposting AI-generated materials in multiple languages and formats, they ensured that false claims reached diverse audiences, including those with limited access to fact-checking resources.

In others, the objective was to sow confusion among civilian populations, undermining trust in both governments and media outlets.

Challenges for Fact-Checkers and Platforms

Fact-checking organizations faced significant challenges in verifying AI-generated content during the 2026 Iran war. The speed at which synthetic media spread often outpaced their ability to analyze and debunk claims, leaving many false narratives unchallenged for extended periods. Additionally, the visual realism of AI-generated images and videos made them difficult to distinguish from authentic footage without specialized tools or expertise.

Social media platforms were similarly overwhelmed. While companies like Meta and X had policies in place to combat misinformation, these were not adequately adapted to handle the scale or sophistication of AI-generated disinformation. This delay allowed misinformation to take root, particularly in regions where social media was a primary source of news.

For example, Facebook’s anti-spam filters were found to incorrectly block legitimate posts about COVID-19 in past incidents, suggesting a broader issue with over-reliance on automated systems that lack contextual understanding. In the context of wartime disinformation, such errors could have compounded the spread of false information by creating confusion about which sources to trust.

Long-Term Implications for AI and Wartime Disinformation

The 2026 Iran war incidents served as a case study in the dangers of unchecked AI-generated disinformation, particularly in high-stakes environments like conflict zones. The rapid evolution of generative AI tools outpaced the development of countermeasures, leaving societies vulnerable to manipulation.

• Platform Accountability: Holding social media companies responsible for the spread of AI-generated disinformation by enforcing stricter moderation policies and investing in verification infrastructure.
• Media literacy initiatives must evolve to address the unique challenges posed by synthetic media, particularly in regions with limited access to fact-checking resources.
• The global nature of social media means that misinformation campaigns can originate from one region and have immediate impacts elsewhere, necessitating a coordinated response.

While generative AI has legitimate applications in fields like journalism, entertainment, and education, its potential for misuse in influence operations cannot be ignored. Addressing these risks requires not only technical solutions but also a broader societal dialogue about the responsible development and deployment of AI systems.

Sources

• incidentdatabase.aihttps://incidentdatabase.ai/cite/1406/
• incidentdatabase.aihttps://incidentdatabase.ai/cite/1494/
• incidentdatabase.aihttps://incidentdatabase.ai/cite/1414/
• incidentdatabase.aihttps://incidentdatabase.ai/cite/1496/

Incident summary from the AI Incident Database, used under CC BY-SA 4.0.