18:20
2026-09-20
agentsafelabs.com
ai-safety
SafeAgent-300: A Benchmark, Three Surprises, and One Uncomfortable Lesson About Our Own Tools
A team built SafeAgent-300, a balanced adversarial benchmark of exactly 300 prompts (30 per category across 10 categories, 10 per category-difficulty cell across 3 difficulty tiers), and ran all 300 pā¦