Skip to content
Submit incident
AII-1000-76731
ConfirmedA court, a regulator or the organization itself has stated the core facts, or the record cites a court or regulator source.

AI-Generated Deepfake Video Falsifies Behavior After Tainan Policewoman's Death

An AI-generated deepfake video falsely depicted a female university student's behavior after a fatal motorcycle-to-pedestrian traffic accident in Tainan, Taiwan that killed a policewoman. The synthetic video, which accumulated over 300 million views, distorted her conduct to fuel online harassment, though authorities confirmed possession of authentic footage and a lawyer publicly identified the deepfake.

28 April 2026 · Tainan Police Department · Misinformation and synthetic media
28 April 2026
AII-1000-767318 sources ↓
Quick facts
Category
Misinformation and synthetic media
Country
Taiwan
Source
Liberty Times Net
Date
28 April 2026
Drafted by
Raidu Incident Detection

What happened

On April 28, 2026, Zheng Yongxin, a policewoman in Tainan, was fatally struck by a tour bus after being hit by a motorcycle ridden by a female university student on An'an Road in Annan District. A video subsequently circulated online showing the female student repeatedly examining her motorcycle's damage, which attracted over 300 million views and thousands of likes.

Lawyer Lin Zhiqun identified the widely-shared video as AI-generated deepfake content that did not accurately depict the actual incident. Real footage from a dashcam showed the female student did briefly examine her motorcycle damage and spoke with the tour bus driver, but the AI video falsely exaggerated her behavior as repeatedly and obsessively checking the damage.

Reported impact

The deepfake video fueled public anger and harassment against the female student, who was subjected to online mob searching and personal attacks. The false portrayal contributed to widespread condemnation despite the student's actual culpability in the accident.

Public discussion highlighted concerns about the difficulty in distinguishing authentic from synthetic video content, with observers noting that the AI-generated video had near-perfect visual fidelity and was nearly impossible to identify as fake without careful examination or exposure of the genuine footage.

Index analysis

The incident demonstrates how synthetic media can amplify existing reputational harms by misrepresenting factual events. Although the female student bore responsibility for the initial accident, the AI-generated content exaggerated and distorted her behavior to maximize public outrage.

The deepfake video was disseminated through what some observers characterized as a possible cognitive warfare operation, with allegations that the content originated from external actors aiming to exploit and inflame public sentiment.

Governance controls

The Tainan Police Department confirmed possession of the original, unaltered video footage and stated they would not be influenced by the circulating false version. However, police noted that without a formal complaint from the affected parties, they could not initiate an investigation into those spreading the deepfake video.

The Tainan District Prosecutor's Office indicated they were collecting and reviewing related information and continuing their investigation into the underlying traffic incident, though no specific action regarding the synthetic video was documented.

Classification

Category
Misinformation and synthetic media
Industry
Media, social platforms and marketing
Country
Taiwan
Harm status
Actual harm
Intent
Unknown
Timing
Unknown
Harm types
Reputational, Psychological, Public interest, Human or fundamental rights, Physical death
Harmed
General public
Failure modes
Deepfake video, Manipulation or dark pattern
AI systems
AI video synthesis system
Regulations
GDPR (cited)

Sources and evidence

This record was researched and written by the Index, based on the following original reporting.

Liberty Times Net
消費女警車禍悲劇 AI變造肇事者影片吸「仇恨」流量 - 臺南市 - 自由時報電子報
6 May 2026
壹蘋新聞網
女警車禍亡|女大生「查看車損」傳出AI變造片!來源竟是中國|壹蘋新聞網
7 May 2026
新頭殼 Newtalk
認知作戰?害死女警女大生反覆查看車損引撻伐!律師:AI假影片 | 社會 | Newtalk新聞
7 May 2026
UDN
台南女警車禍影片疑經AI變造?台南警:已掌握原始檔案 | 聯合新聞網
7 May 2026
中時新聞網
撞女警女大生「反覆看車損」影片 吸300萬瀏覽 律師:AI做的 - 生活
7 May 2026
Central News Agency
台南女警案影片疑經AI變造 警已掌握原始檔案 | 社會 | 中央社 CNA
7 May 2026
mnews.tw
驚傳畫面被AI動手腳!台南女大生追撞女警「事後影片疑變造」 警方回應了
7 May 2026
三立新聞
逆風!撞死女警「女大生冷血看車」是AI假影片? 律師:又是中國搞的! | 社會 | 三立新聞網 SETN.COM
7 May 2026