65-Year-Old Woman Used AI Deepfake to Romance Scam Shanghai Man of 170,000 Yuan
A 65-year-old woman in China used AI technology to create over 7,000 videos and photos of a fake persona as a 30-year-old female doctor to conduct romance scams with multiple men. She defrauded a Shanghai man of 170,000 yuan over five months by impersonating a young medical professional who could provide construction projects, maintaining the deception despite living in the same city and never meeting in person.
What happened
A woman born in 1961 created an account claiming to be Yu Xiuzhen, born in 1992, a doctor at a tertiary hospital, using AI to generate over 7,000 videos and photos of a young woman. The account accumulated over 10,000 followers and published daily life content to establish credibility.
She engaged in romantic relationship with a Shanghai man named Liu Hui over five months, maintaining that she could provide hospital construction projects. During this period, she never met him in person despite both living in Shanghai, and refused video calls, claiming her mother was preventing contact.
Over five months, Yu Xiuzhen requested 170,000 yuan from Liu Hui through multiple transfer requests, citing project deposits and personal expenses. She claimed her mother's opposition required a 300,000 yuan bride price and Shanghai property purchase.
When promised construction projects did not materialize, Liu Hui reported the case to police. Authorities discovered the perpetrator was a 65-year-old woman born in 1961, without medical credentials, formerly employed as janitorial staff at a hospital. She had a prior 2009 conviction for fraud with a 4.5-year sentence.
Reported impact
Liu Hui lost 170,000 yuan in the fraud scheme and suffered psychological harm from the extended deception conducted across five months by someone he believed was a legitimate romantic and business contact.
The incident raised public concerns about the authenticity of online video content and the difficulty of verifying identities when AI-generated content can be indistinguishable from genuine material.
Index analysis
The perpetrator used AI video synthesis to lower the barrier for identity fraud, generating bulk video content with skin smoothing and scene composition to create a convincing false persona. Some videos included small text disclaimers stating 'this work contains AI-generated content,' which the victim did not notice.
The fraudster simultaneously maintained romantic relationships with multiple victims using the same fake identity, demonstrating scalability of the scam through synthetic media creation.
Governance controls
Social media platforms included small-font disclaimers on AI-generated content, though these were not sufficiently prominent to alert users to synthetic material.
Classification
Sources and evidence
This record was researched and written by the Index, based on the following original reporting.