OpenAI banned a coordinated network of ChatGPT accounts that likely originated in Cambodia’s Preah Sihanouk province, a region reports have linked to online scam compounds and human trafficking operations.
The network used the company’s models to create and manage fake online personas, generate and translate messages sent to scam targets, produce promotional content for fraudulent schemes, and support daily operations.
Earlier this year, a lead from WhatsApp prompted OpenAI to investigate and disrupt a Cambodia-based scam operation that used ChatGPT to support investment, romance, gambling, and law enforcement impersonation scams.
“The full scale of financial losses associated with the network is unknown, but based on the scammers’ own communications, the operation may have interacted with hundreds of targets across multiple scam types. User conversations referenced individual victims losing thousands of dollars, although we are unable to independently verify those claims,” OpenAI said.
How the operation worked
The operation ran multiple scams at the same time, often combining elements from different schemes. In interactions with targets, scammers relied on what the company describes as the “ping” (initial outreach), the “zing” (generating emotion), and the “sting” (extracting money).
AI-generated images created by scammers in the network to advertise jobs in Cambodia on social media. Redactions added by OpenAI. (Source: OpenAI)
The scammers used ChatGPT to translate messages and generate conversations with people on platforms including WhatsApp and Telegram. They created social media posts and researched dating profiles and related content to make fake identities appear more credible.
Their messages were designed to build trust and create a sense of urgency. They promised guaranteed returns and “risk-free” investments, used romantic language, encouraged targets to keep conversations private, and pressured them to act before fake offers or bonuses supposedly expired.
Once victims were engaged, the scammers instructed them to send money to unlock fake rewards, pay activation fees, or settle fabricated fines. They then asked victims to provide screenshots of payments or share account information as proof that the transfers had been completed.
Links to human trafficking
Some accounts in the network generated content indicating links to human trafficking and forced criminality associated with organized crime groups in East Asia. Public reporting has documented that these groups recruit people with promises of legitimate employment before trapping them in debt bondage or coercive working conditions.
ChatGPT was also used for administrative tasks, including drafting internal announcements, translating messages between staff, and documenting matters related to recruitment, immigration status, working conditions, and employee discipline.
The network also used OpenAI’s models to create social media advertisements offering flights, accommodation, meals, visas, and work permits. Other activity involved managing workers by documenting employee debts, salary deductions, fines, loan repayments, visa overstays, work permits, immigration status, and recruitment incentives.
Some conversations suggested that people involved may not have been there voluntarily. They referred to detention, attempts to escape, and concerns about being held criminally responsible. While these conversations do not establish what happened to any individual, they are consistent with public reporting on organized crime groups that traffic people and force them to work in online scam operations.
