Google Deploys SAFE, an AI Spam Hunter That Thinks Like a Human Investigator

Google has quietly deployed SAFE, a multi-agent AI system that investigates spam like a human forensic team and flags content violating the "spirit" of its policies.