ai-safety 5
- Everyone Is Lying To You For Money
- Perturbation Probing: A New Diagnostic for the Fragility of LLM Safety
- “Sorry, I can’t help with that”: How your guardrails might become the attacker’s best friend
- Claude Code hooks and how Sonar Vortex verifies AI code inside the loop
- Why Fable 5 Still Needs a Second Loop