THE article discusses "ScamBuster," an open-source, AI-driven system designed to combat phishing scams by engaging with attackers using victim personas. Developed by Laurent Giovannoni, ScamBuster aims to collect data from scammers to identify their operations and support law enforcement efforts. The system operates as an inbound-only mechanism, responding to scam emails to gather intelligence without initiating contact.
It is designed to evolve through self-learning, improving its effectiveness over time, and supports various AI models for flexibility. The full production code will be released at Black Hat USA 2026.