OpenAI has disrupted a coordinated network of ChatGPT accounts originating in Cambodia that supported investment, romance, gambling, and impersonation scams. The investigation, initiated from a lead by WhatsApp, revealed how organized criminal groups opportunistically blend multiple fraud types to deceive victims.
- The network used ChatGPT to create fake personas, generate promotional content, and translate messages for targets on platforms like WhatsApp and Telegram.
- Scammers employed the "ping, zing, sting" pattern: outreach, emotional manipulation, and extracting money through fake investment opportunities or fines.
- Some users generated content suggesting links to human trafficking, including job advertisements in Poipet and records of employee debts and disciplinary actions.
- OpenAI banned the associated accounts, shared threat signals with industry partners and authorities, and implemented measures to prevent these actors from regaining access.
This case highlights that effective disruption requires targeting not just victim-facing scams but also the criminal organizations orchestrating them, as boundaries between fraud and trafficking are often blurred.