Fake AI Content About The Iran War Is All Over X
AI-generated disinformation fills X as Elon Musk’s Grok chatbot spreads false claims about Iranian missile strikes.
San Francisco: When Tal Hagin asked Grok to check a post about Iranian missiles hitting Tel Aviv, Elon Musk’s chatbot gave wrong answers.
The AI first got the video location and date wrong. Hagin said, “Now Grok is replying with AI slop of destruction.” The video came from an Iranian news account. Grok then shared fake images to try proving its point.
Since February 28, X has become full of false information. The US and Israel attacked Iran on that date. Soon after, many accounts posted fake videos using old footage from other events. As fighting continues, these false posts keep growing.
Proliferation means spreading widely – and fake AI images now spread very fast. On March 2, accounts showed fake videos of a burning building in Bahrain. These videos looked real enough that many people believed them. One fake image of a US bomber being shot down was seen over one million times before being removed.
Some videos show Iranian missile factories inside caves. These aren’t real, but still got over one million views. The Iranian government also creates AI posts showing antisemitic ideas – like Orthodox Jews leading American soldiers to war. One fake video made people think they saw young girls walking past President Trump wearing only underwear. This got 6.8 million views before removal.
Hagin said, “What makes this war different is how much AI content I have to check. AI tools now create very convincing fakes.” X announced it will stop paying special verified accounts that post fake war videos without warning labels. Iranian officials had been using these special paid accounts to spread their messages.