AI Safety

A usable safety starter pack should have one framework, one research archive, one evaluation tool, and one red-teaming toolkit. That mix gives people language, examples, executable tests, and a reminder that adversarial work needs its own craft, not just more benchmark rows.

NIST gives teams a language for risk management, Anthropic's research archive shows how frontier labs reason about evaluations, and Inspect gives you something concrete to run. Together they make the work feel operational instead of ceremonial. The hard public questions are about threshold-setting: what evidence should be required before launch, how much outside scrutiny is enough, and when a voluntary practice stops being a sufficient answer. Those arguments are productive when people bring operating context rather than ideology alone.

The tools that keep proving useful usually support evaluation harnesses and benchmark management, policy and review workflows, and incident logging and response tooling without making the underlying work harder to understand. When you bookmark something, write down why it earned the slot.

Three sources worth opening side by side:
- NIST AI Risk Management Framework: nist.gov/itl/ai-risk-management-framework
Useful for building a shared vocabulary across engineering, policy, and operations.
- AI RMF Playbook: airc.nist.gov/AI_RMF_Knowledge_Base/Playbook
The most useful NIST material when a team needs implementation moves, not just principles.
- Inspect source: github.com/UKGovernmentBEIS/inspect_ai
Open source evaluation framework from the UK AI Security Institute.
- Anthropic video archive: youtube.com/@AnthropicAI/videos
Talks and interviews that help connect research language to deployment reality.