AI and deepfakes drive surge in sophisticated investment fraud

Fraudsters are increasingly utilizing artificial intelligence and synthetic media to execute sophisticated investment scams. In 2025, investment scams became the largest fraud-loss category in both Australia and the United States, with losses reaching $837.7 million and $7.9 billion, respectively.

One prevalent method involves deepfake advertisements that impersonate financial professionals to build credibility. These ads direct users to coordinated WhatsApp groups where a fake analyst provides specific trading guidance for small-cap stocks. This tactic facilitates pump-and-dump schemes, where concentrated buying pressure from group members inflates stock prices before operators sell their holdings, leaving victims with significant losses.

Research also indicates that AI chatbots are becoming highly effective tools for “pig butchering” scams—long-term schemes designed to build emotional trust before soliciting investments. In a simulated study, an AI chatbot outperformed human scammers, achieving a 46 percent compliance rate in getting participants to install an app, compared to 18 percent for humans. Participants reported higher levels of trust in the chatbot, which maintained engagement more consistently and recalled personal details more effectively than human counterparts.