A surge of AI-generated misinformation about the war involving Iran is rapidly spreading across the social media platform X, raising alarm among researchers who say artificial intelligence tools are amplifying confusion during the ongoing conflict. Analysts warn that both fabricated visuals and flawed AI verification systems are making it increasingly difficult for users to distinguish reality from manipulation.
The issue came into focus when disinformation researcher Tal Hagin asked X’s AI chatbot Grok to verify a video claiming Iranian missiles had struck the city of Tel Aviv. Instead of confirming the footage, the chatbot repeatedly misidentified the location and timing of the video, which had originally been posted by an Iranian state-run media outlet. Grok then attempted to support its claim by generating and sharing its own AI-created image, further spreading confusion rather than clarifying the situation.
The incident reflects a broader trend that has unfolded on X since the United States and Israel began military operations against Iran on February 28. Since the start of the conflict, the platform has been inundated with misleading posts that recycle older footage, alter videos, or present fabricated visuals as evidence from the battlefield. Researchers say the volume of false material has increased in recent days as AI image and video generation tools have become widely accessible.
Several viral posts have featured convincing but entirely fabricated visuals. Iranian officials and state media accounts, for example, circulated AI-generated videos depicting a high-rise building in Bahrain engulfed in flames. Other posts showed dramatic scenes such as a Northrop Grumman B-2 Spirit allegedly shot down by Iranian forces or elite American soldiers being captured. Some of these images attracted millions of views before being removed, highlighting how quickly such content can spread before fact-checkers intervene.
Even less convincing material has achieved widespread circulation. One widely shared clip claimed to show Iranian troops manufacturing missiles inside a hidden cave complex. Despite obvious signs of artificial generation, the video still gained more than a million views as it was reposted across multiple accounts.
Researchers also report that AI-generated content has been used in coordinated propaganda campaigns. According to analysts at the Institute for Strategic Dialogue, networks of pro-regime accounts have circulated fabricated images portraying Orthodox Jewish figures directing American soldiers into battle or celebrating US casualties. These posts have been described as part of a broader campaign combining political propaganda with antisemitic narratives.
In another widely viewed incident, a fabricated video circulating on X appeared to show young girls walking past former US president Donald Trump wearing only underwear. The clip reportedly attracted more than 6.8 million views before it was taken down, though copies continue to circulate through other accounts.
As the volume of manipulated material has grown, X has attempted to curb the problem by temporarily demonetizing premium accounts that share AI-generated war footage without clear labeling. The policy specifically targets accounts with blue checkmarks that benefit from the platform’s monetization and engagement systems. However, the company has not disclosed how many accounts have been affected since the rule was introduced, and critics say enforcement remains unclear.
Traditional misinformation has also persisted alongside AI-generated material. One particularly contentious narrative involves a missile strike on a primary school in the Iranian city of Minab that killed more than 168 people, including 110 children. Some pro-Trump accounts on X have circulated unrelated videos to claim the Iranian government was responsible. However, footage verified by journalists shows a Tomahawk cruise missile striking a nearby naval facility. Despite claims by Trump that Iran possesses the weapon, the United States is currently the only military in the conflict known to operate Tomahawk missiles.
Concerns about AI misinformation extend beyond X. The oversight board monitoring the policies of Meta Platforms recently criticized the company’s labeling systems for AI-generated media, warning that existing measures are not strong enough to cope with the speed and scale of synthetic misinformation during crises.
