The company behind ChatGPT has cut off a cluster of accounts it ties to scam operations in Cambodia. OpenAI traced the network to Preah Sihanouk province, where scam compounds and human trafficking are well documented, and began investigating after a WhatsApp lead surfaced earlier this year.
OpenAI models powered the scheme’s fake personas, drafted and translated the messages sent to targets, generated promotional material, and handled day-to-day logistics. Scammers ran investment, romance, gambling and law enforcement impersonation schemes on platforms including WhatsApp and Telegram.
OpenAI says the full scale of financial losses is unknown, but the scammers’ own communications referenced hundreds of targets and individual victims losing thousands of dollars. Messages built trust and urgency, promised risk-free returns, and pushed targets to send money for fake rewards, activation fees or fabricated fines.
Content from certain accounts pointed to human trafficking and coerced criminal activity connected to organized crime networks in East Asia. ChatGPT was used for administrative work too, including drafting internal announcements and documenting employee debts, salary deductions and visa overstays. Conversations referenced detention and attempts to escape, consistent with public reporting on trafficked workers forced into scam operations.
OpenAI says the takedown is part of a broader push to document how criminals abuse generative AI and to build detection for coordinated abuse at scale.
