arXivPiercosma Bisconti, Matteo Prandi, Federico Pierucci, Federico Sartore, Enrico Panai, Laura Caroli, Yue Zhu, Adam Leon Smith, Luca Nannini, Marcello Galisai, Susanna Cifani, Francesco Giarrusso, Marcantonio Bracale Syrnikov, Daniele NardiThu, May 21, 2026, 8:50 AM PDT
score 14.7
New benchmark tests if AI agents fall for gradual manipulation attacks
Original: Boiling the Frog: A Multi-Turn Benchmark for Agentic Safety
Source: arxiv.org ↗
Writing ELI5 summary…