An AI-generated image showing Thai police officers dressed in drag for a drug-arrest operation fooled media outlets worldwide before being identified as a hoax [1, 2].
The incident highlights the growing difficulty news organizations face in verifying visual content as generative artificial intelligence becomes more sophisticated. When trusted sources share manipulated media, it can rapidly accelerate the spread of misinformation across international borders.
The image was first posted to a Thai police station's Facebook page on May 21, 2024 [1]. The visual depicted officers in drag, suggesting a tactical disguise used to apprehend suspects during a narcotics operation. Because the image appeared on an official police page, it gained immediate traction and was shared thousands of times on social media [1].
Several news organizations in the U.S. and the United Kingdom shared the image as genuine photography [1]. Some reports described the scene as a creative police tactic, while other global media outlets amplified the story without independent verification [2]. The image continued to circulate as a factual account of Thai law enforcement activity for several days.
Verification efforts eventually revealed the image was not a photograph. On May 27, 2024, the image was officially identified as AI-generated [2]. The creators had used artificial intelligence tools to produce the scene as a joke, but the lack of clear labeling led to it being mistaken for a real-world operation [1, 2].
This cycle of misinformation began with a single social media post and ended only after the image was debunked by digital forensic checks. The incident serves as a case study in how AI-generated content can bypass traditional editorial filters when the source appears authoritative.
“The image was created as a joke/hoax using artificial-intelligence tools.”
This incident demonstrates a critical vulnerability in the modern news cycle: the 'authority bias.' Because the hoax originated on an official police Facebook page, journalists bypassed standard verification protocols, assuming the source was reliable. As AI tools produce more photorealistic images, the reliance on source reputation over technical verification increases the risk of institutional misinformation.
