Skip to content
Submit incident
AII-1001-51651
ConfirmedA court, a regulator or the organization itself has stated the core facts, or the record cites a court or regulator source.Supporting sources are pending verification.

Japanese town withdraws bear alert based on AI-generated image

A Japanese town issued a public bear alert based on an AI-generated photograph and subsequently apologized after discovering the image was synthetic. The incident reflects wider concerns about false AI-generated bear videos circulating on social media in Japan during a year of record bear attacks.

27 November 2025 Β· Onagawa Β· Misinformation and synthetic media
27 November 2025
AII-1001-51651
Quick facts
Organization
Onagawa (deployer)
Category
Misinformation and synthetic media
Country
Japan
Date
27 November 2025
Drafted by
Raidu Incident Detection

What happened

The town of Onagawa in northern Japan issued a public alert on social media warning of a bear near a kindergarten, based on a photograph showing the animal. The town later removed the alert after being contacted by the person who created the image, discovering the photograph had been generated by artificial intelligence.

Reported impact

The town apologized for causing anguish and confusion among residents. Journalists from the Yomiuri Shimbun found that approximately 60 of 100 bear-related videos on TikTok analyzed were AI-generated, with examples including a scene of an elderly woman feeding apples to a bear.

Index analysis

The town prioritized informing residents to prevent danger, given that authentic bear reports had occurred in another district the previous weekend and actual bears have been entering homes, lurking near schools, and damaging supermarkets in rural northern regions. The town had no significant doubts about the image at the time of posting, as bear sightings were occurring in the area.

Classification

Organization
Onagawa (deployer)
Category
Misinformation and synthetic media
Industry
Media, social platforms and marketing
Country
Japan
Harm status
Actual harm
Intent
Unintentional
Timing
Post-deployment
Harm types
Psychological
Harmed
General public
Failure modes
Deepfake image, Misidentification
Consequences
Content removal, Public apology
AI systems
Unknown AI image generation system

Sources and evidence

Sources not yet published

This record was researched and written by the Index from original reporting. Its sources are still being verified by an editor, and they are listed here once that check is complete. Until then the record carries no reviewer.

How sources are verifiedSuggest a source for this incident