Researchers found that AI chatbots can build trust with potential victims of “pig butchering” scams more effectively than human scammers during the long conversational phase before a fake investment pitch. This matters because fraud networks could automate much of the emotionally manipulative work at scale, while using human operators only at the final stage to evade AI safety safeguards.

In a week-long experiment involving 22 unwitting participants, nearly half complied with a request from an AI chatbot to download an app or play a game, compared with fewer than one in five who were asked by a human scammer. Participants also reported significantly greater trust in the AI, and the study drew on interviews with 145 former scam workers, including trafficking survivors forced to work in compounds in Cambodia, Myanmar and Laos.