MIT releases AI Risk Mitigation Taxonomy v3: 2,206 strategies from 47 frameworks
Exp_Mark · x · 2026-11-13
MIT's AI Risk Initiative released version 3 of its AI Risk Mitigation Taxonomy, extracting 2,206 mitigation strategies from 47 frameworks and publishing them as an open database with an interactive explorer.
Key points:
- Two-axis taxonomy: mitigations are organized by domain (what the mitigation aims to achieve) and instrument (how it creates change), plus four additional codings: AI actor, lifecycle stage, risk addressed, and risk management function.
- Answers who should act, when, against which risks, and how.
- The explorer supports cross-filtering, e.g. 161 records for developer-side security & access control, with concrete practices like continuous security testing and account management.
- Includes PRIMER, a shared standard for reporting mitigations, reusable by developers, deployers, policymakers, and researchers.
More from Safety
- Western University reviews $72 AI tutor after student complaints over cost and privacy — Afinetheorem · 2026-11-13
- Families protest in 7 US cities over Google's inability to offer AI-free Search on school Chromebooks — Ars Technica AI · 2026-11-13
- AI safety think tank FLI reportedly 94% foreign-funded, incl. $665M in SHIB from Buterin — csuwildcat · 2026-11-13
- Railway hack postmortem: attacker likely used 'abliterated' AI models with no safeguards — The Pragmatic Engineer · 2026-11-13
- PauseAI UK is organizing what it calls the largest AI safety march in history in London — KatjaGrace · 2026-11-13
- Singapore barred Alice Huang from re-entering after 8-month probe; insufficient evidence of offences — bookwormengr · 2026-11-13