Fake AI-generated images and videos about the arrest of Venezuelan president Nicolas Maduro have flooded social media, receiving millions of views. Experts warn that visual cues are no longer a reliable indicator of artificial intelligence.
Within hours of Donald Trump’s announcement that Maduro had been captured, AI-generated images of the arrest and outdated footage claiming to show the military operation in Caracas were widely shared. Elon Musk reposted an AI-generated video of Venezuelans crying on their knees, thanking Trump and the US for freeing them from Maduro. The footage, originally published by an account on X called Wall Street Apes, has been viewed 5.7 million times on the platform.
Research by Shayan Sardarizadeh, a senior journalist for BBC Verify, revealed that the video was originally posted on TikTok by an account called Curious Mind, which regularly shares AI-generated videos. A community note underneath the reposted video described it as “AI generated and is currently being presented as a factual statement intended to mislead people.” Multiple errors can be spotted in the video, from incorrect flag patterns and disappearing objects to missing teeth. The video has since been taken down on TikTok but remains on X.
Musk has gone on to share multiple “AI slop” videos, including one deepfake of Maduro breakdancing with President Trump, and another of the Venezuelan president in prison with Sean “Diddy” Combs. Vince Lago, mayor of Coral Gables, Florida, reposted a hyper-realistic, AI-generated photo of Maduro being led off a plane by US law enforcement agents on Instagram. Analysis by Google’s SynthID detection tool revealed an invisible watermark on the content that proves it was generated or edited using AI software.
The creator of the now-viral, AI-generated photo of Maduro later came forward as a Spanish-based X user with fewer than 100 followers. A self-described “AI video art enthusiast”, who goes by the name Ian Weber, told AFP that he had never expected it to become so widely shared. He created the fake with Google Gemini’s Nano Banana Pro to post within 20 minutes of Trump’s announcement of the operation on Truth Social.
NewsGuard, which provides services such as misinformation tracking, said that seven of the misleading photos and videos it identified have now garnered more than 14 million views on X alone. Benjamin Dubow, a democratic resilience fellow at the Center for European Policy Analysis (CEPA), explained to The Independent that the AI-generated content came about as social media users attempted to fill an information void in the hours after Maduro’s capture was announced.
