When a magnitude-7.1 earthquake struck Japan’s Kumamoto Prefecture, the devastation was immediate, but a secondary disaster emerged almost simultaneously in the digital realm. As rescue teams scrambled to reach those in need, social media platforms were flooded with a toxic mix of AI-generated fake news, fabricated cries for help, and malicious conspiracy theories. This deluge of misinformation did more than just confuse the public; it actively hampered disaster response efforts, turning social media—a tool that should be a lifeline—into a chaotic environment that threatened to drown out legitimate appeals for assistance. Local authorities were forced to pivot their focus from solely saving lives to fighting a war against digital deception, pleading with the public to halt the indiscriminate sharing of unverified claims that were actively obstructing real-world rescue operations.
The extent of the deception was as alarming as it was sophisticated. On X, formerly Twitter, a single post claiming, “Please help me. I’m buried alive,” garnered over 300,000 views before it was exposed as a hoax—the address provided simply did not exist. Beyond these individual pleas, bad actors leveraged generative AI to create deepfake news broadcasts that mimicked reputable organizations, gaining tens of thousands of views and lending a veneer of professional legitimacy to outright lies. Other posts recycled harrowing footage from the 2024 Noto Peninsula earthquake, reframing the older clips as current damage to incite fear. Perhaps most disturbingly, xenophobic rumors claiming that foreigners were setting fires began to circulate, weaponizing the chaos to stir up social discord at a time when collective unity was needed most.
The phenomenon reached a surreal peak with the emergence of conspiracy theories claiming the earthquake was “artificial” or had been predicted in advance. One particular post, which featured a doctored map to “prove” the author had anticipated the seismic event, racked up over 3.4 million views. By feeding into a human desire for answers and control during a terrifying event, these posters were able to manipulate public sentiment and amplify misinformation at a scale that traditional news outlets struggled to counter. This digital pollution didn’t just misinform; it created a distorted reality where residents were unsure of what to believe, whom to trust, or where to seek safety, effectively paralyzing the community’s ability to respond rationally to the unfolding crisis.
Experts have pointed out that this is not an entirely new problem for Japan, but it has certainly reached a dangerous new plateau. History shows a recurring pattern of misinformation during disasters, ranging from the absurd, such as the 2016 rumor about a lion escaping a zoo, to the genuinely harmful, like the fake rescue requests that plagued the Noto Peninsula response earlier this year. However, the rise of generative AI has changed the rules of engagement. Because AI can now produce images, videos, and news-style reports that look almost indistinguishable from reality, the burden on the average citizen to detect fraud has become exponentially higher. The speed at which sensational content travels has outpaced our collective ability to verify it, making the digital ecosystem during a crisis increasingly hostile.
In response to this digital tide, professionals like Furuta Daisuke, editor-in-chief of the Japan Fact-check Center, are calling for a fundamental shift in how we consume information during emergencies. The message is simple but vital: pause, verify, and verify again. When confronted with sensational footage or dramatic pleas for help, individuals must resist the impulse to hit the “share” button. Instead, they should cross-reference information exclusively with official announcements from government bodies, emergency services, and established, verified news organizations. The act of sharing an unverified post—no matter how well-intentioned—can have real-world consequences, potentially diverting resources away from genuine victims and placing an unnecessary burden on an already strained rescue infrastructure.
Ultimately, the lesson from the Kumamoto earthquake is that we must become more responsible stewards of information in the digital age. As AI tools become more advanced, the “truth” will become harder to find, but the tools to identify it remain the same: patience, critical thinking, and reliance on trusted sources. We must recognize that every click and share is a vote for what information survives in the chaos of a disaster. By refusing to engage with unverified or sensational content, the public can play a crucial role in preventing secondary harm. Protecting our disaster response systems now requires not just physical aid and rescue teams, but a digital discipline that ensures truth remains a casualty of neither the earthquake nor the internet.