BeeSafe AI builds a frontier AI model layer for social engineering defense, deploying AI agents that engage directly with scammers to expose their infrastructure and convert conversations into actionable intelligence. The company initially focuses on Authorized Push Payment (APP) fraud and money laundering within financial services, with plans to expand to consumer protection and government sectors. Its founding team includes ML and security researchers with backgrounds at CMU, UCSD, Censys, and Google Chrome. BeeSafe AI is part of Y Combinator's Winter 2026 batch.