The Failure First Team
Fifteen specialist AI agents, one methodology. How adversarial AI evaluation scales through Claude Code sessions with distinct roles and standing instructions.
Tag: ai-safety
8 posts
Safety-First Therapeutic AI
Building AI for trauma therapy means the safety architecture has to exist before a single therapeutic feature does. Here's why.
Adversarial Poetry: When Rhyme Bypasses Reason
Reformulating harmful prompts as poetry bypasses safety filters across every major LLM family. A single-turn, universal jailbreak mechanism.
Why Demonstrated Risk Is Ignored
Large organisations rarely fail because risks are unknown. They fail because known risks are structurally difficult to act on.
120 Models, 18,176 Prompts: What We Found
120 models, 18k prompts: supply chain injection at 90–100% attack success, faithfulness gaps in frontier models, and why your benchmark numbers are wrong.
The Cognitive Cage: Humanoid Robot Fatality Risk
A probabilistic risk model for VLA-driven humanoid fatalities projects a 'Danger Zone' between 2027–2029: the mechanism, timeline, and what follows.
Jailbreak Archaeology: 4 Years of Broken Promises
64 jailbreak scenarios across six eras tested on 2026 frontier models. Key finding: 2022 attacks still achieve ~30% success on today's reasoning models.
When AI Systems Talk to Each Other, Safety Breaks Down
Single-agent safety does not compose in multi-agent systems. 1.5M interactions show 46.34% attack success rates and 16-minute median failure windows.