Japan’s release of AI-generated videos depicting a dramatic eruption of Mount Fuji has stirred mixed reactions. Is this forward-thinking disaster preparedness — or unnecessary alarm that risks eroding public trust?

For the first time in more than 300 years, Mount Fuji is “erupting” — not in reality, but on screens across Japan. As part of Volcanic Disaster Preparedness Day, the Tokyo Metropolitan Government has released AI-generated videos simulating a violent eruption of Japan’s most iconic peak, a mountain that last erupted in 1707.

The videos, produced with computer modelling and artificial intelligence, show an unsettling scenario: ash clouds spreading over Tokyo within hours, blanketing roads, paralysing trains and planes, cutting off food and power, and leaving residents with long-term respiratory problems. The message at the end is stark: “We need to arm ourselves with facts and prepare for disaster in our daily lives.” Viewers are reminded to keep pantries stocked with canned food, water, and first-aid supplies.

Officials insist there are no signs at all that Fuji is stirring. “The simulation is designed to equip residents with accurate knowledge and preparedness measures,” the Tokyo government explained. But the videos, vivid and cinematic, have unsettled some viewers. One Tokyo resident asked pointedly: “Are there actually any signs of eruption? Why are we suddenly hearing about ashfall in Tokyo?” Another said the frightening images pushed her to stock up on emergency supplies — which may well have been the intended effect.

So is this smart public safety messaging, or unnecessary fearmongering? Experts note that Japan is no stranger to disaster simulations, given its location on the seismically volatile Ring of Fire. Risk communication professor Naoya Sekiya of the University of Tokyo stressed that Fuji is not showing signs of imminent eruption, adding, “There’s no particular significance to the timing.” Yet, with memories of earthquakes, typhoons, and tsunamis never far away, Japanese authorities see preparedness as a national duty.

What’s less clear is whether AI-driven hyper-realistic disaster “trailers” help citizens calmly prepare, or simply add another layer of anxiety. Disaster planning is vital — but some wonder if Mount Fuji’s imagined eruption risks crossing the line into disaster-porn.

Ultimately, the debate raises a bigger question: is this the best use of AI by governments? Used wisely, AI can help educate, simulate, and even save lives. But if the technology tips into spectacle, it risks undermining the very trust and calm that effective disaster preparedness requires.