Viral AI Hurricane Photo Highlights Misinformation Dangers

AI-Generated Image

A Viral Deception Amidst a Real Disaster

As the Caribbean grappled with Hurricane Melissa, one of the most powerful storms in recorded history with sustained winds of 185 mph, a bizarre and completely fake image of the storm began to circulate online. While sharing altered photos after natural disasters is a long-standing issue, the rise of generative AI is making crisis misinformation more severe than ever before.

One of the most prominent pieces of disinformation surrounding Hurricane Melissa was an AI-generated picture showing a massive hurricane eye from high above, with a flock of birds inexplicably circling safely within the calm.

Experts Debunk The Impossible Image

Meteorological experts quickly pointed out the scientific absurdities of the viral image. Rich Grumm, a retired meteorologist, explained to Yale’s Climate Connections blog that the scale was completely wrong. Hurricane Melissa’s eye was reported to be around 10 miles wide, which would make the birds in the image “larger than football fields.”

Lee Grenci, a former meteorology professor, added that for the image to be real, “these birds would have to have been flying at altitudes well above the summit of Mount Everest.” At such heights, the air temperature and density are far too low for birds to fly. The fabricated image stood in stark contrast to the real, jaw-dropping footage captured by hurricane hunters who bravely flew directly into the storm's eye.

How The Fake Image Spread Like Wildfire

The fake image appears to have originated on X (formerly Twitter) before spreading rapidly across Facebook, Instagram, TikTok, and Threads, gathering tens of thousands of reactions. The widespread posting of the identical image suggests the involvement of a coordinated bot farm designed to amplify its reach. The deception was effective, with one user on X commenting, “I’ve never seen the ‘stadium effect’ so pronounced and well-defined... This will be in meteorology textbooks.”

The Broader Danger of AI Disinformation

This incident was not isolated. Another piece of AI-generated content that went viral appeared to show a Jamaican hospital completely destroyed by the hurricane. Fact-checkers were able to identify it as fake thanks to a SynthID stamp, an invisible watermark used by Google’s AI tools.

Despite the debunking, the potential for harm is significant. Such misinformation could have easily convinced people seeking medical help that a functional hospital was out of service. These events are a stark reminder of the dangerous rise in AI-generated misinformation and the challenge of containing its impact in our increasingly digital world.

More on AI: Slop Farmer Boasts About Using AI to Flood Social Media