OpenAI has reported the disruption of ChatGPT usage by cyber scam centers operating out of Cambodia. These centers allegedly integrated the chatbot into various fraudulent schemes, including investment fraud and human trafficking operations, primarily targeting individuals in India.
The scammers leveraged ChatGPT to facilitate investment, romance, gambling, and law enforcement impersonation scams. They also generated deceptive job advertisements aimed at Indian nationals. OpenAI was alerted to this activity by officials at WhatsApp, leading to the banning of accounts associated with the operation.
According to OpenAI, the network utilized their models to create and sustain fake online personas, generate and translate messages sent to scam targets, develop promotional content for their fraudulent schemes, and assist with daily operational tasks. The activity was traced to Poipet, Cambodia, a city known for its connections to scam operations run by Chinese criminal organizations.
The fraudulent schemes involved the creation of fake images of passports, legal notices, stock purchase confirmations, and gambling platform pages using ChatGPT. Beyond direct scamming, the chatbot was also employed to manage the internal operations of the scam centers. This included generating internal announcements, translating communications between staff members, and producing recruitment flyers. These flyers, distributed in India, falsely promised free flights, accommodations, and work visas.
OpenAI's investigation revealed that the operators maintained records of employee debts, salary deductions, disciplinary fines, and loan repayments. They also translated discussions concerning immigration status, work permits, visa overstays, and recruitment incentives. Some conversations even referenced apparent detention, escape attempts, and potential criminal liability for individuals who had been trafficked and coerced into working in these scam operations.
While OpenAI could not precisely quantify the financial losses, they indicated that the operation "may have interacted with hundreds of targets across multiple scam types," with several conversations mentioning victims losing thousands of dollars.
The proliferation of scam centers across Southeast Asia persists despite law enforcement efforts by the U.S. and China that have disrupted numerous such operations this year. Experts have also voiced concerns about the expansion of these centers into South Asia and Africa.
A recent study by INTERPOL, based on a survey of 36 member states, found that over half of all cybercrime in Africa now incorporates artificial intelligence. Law enforcement agencies believe AI is being used to automate every phase of the attack lifecycle, from reconnaissance and phishing to extortion and evasion. One incident in Uganda, where AI-generated audio and video were used to impersonate a public figure promoting a fraudulent investment scheme, resulted in losses exceeding $2 million.
