Knowledge Bases and Evaluation Organizations: MIT AI Risk Repository
Knowledge Bases and Evaluation Organizations · MIT AI Risk Repository. AI risk taxonomy and incident mapping. This source has 7 entries admitted under this category's review policy; each links directly to the original.
Authority tiers in this category: A2–B. Admission policy: All core resources are included. Each secondary source is admitted according to its own database, methodology, versions and leaderboard structure; there is no cross-organization scoring scheme.
MIT AI Risk Repository
Collects risk entries from existing AI risk frameworks and provides a causal taxonomy and a domain taxonomy.
Evidence use: Supports AI risk concepts, taxonomies, source frameworks and cross-framework mappings.
Limitations: The taxonomy emphasizes comprehensiveness rather than risk probability. Database entries cannot replace an assessment of a specific system.
Related concepts: AI Governance and Regulation · Model Evaluation and Benchmarks
Link last checked: 2026-09-24
MIT AI Risk Mitigation Map
Collects risk mitigations and organizes them into categories such as governance, technical safety, operational processes and transparency and accountability.
Evidence use: Supports structured mapping between AI risks and optional mitigations.
Limitations: The effectiveness of mitigations still needs specific evidence. The taxonomy is continuously updated.
Related concepts: AI Governance and Regulation · AI Guardrails
Link last checked: 2026-09-24
MIT AI Risk Initiative data and research hub
Provides open databases and research updates on risks, incidents, governance mapping, mitigations and expert priorities in one place.
Evidence use: The authoritative portal to the MIT AI Risk project's databases, reports and updates.
Limitations: The main portal itself does not support specific risk conclusions. Sub-projects use different methods.
Related concepts: AI Governance and Regulation
Link last checked: 2026-09-24
MIT AI Incident Tracker
Classifies public AI incident reports in a structured way and analyzes risk domains and incident patterns.
Evidence use: Supports observing types of AI risk and how they change through real incidents.
Limitations: The database is continuously updated; statistics should include the access date. Aggregate taxonomies cannot replace a risk assessment of a specific system.
Related concepts: AI Governance and Regulation
Link last checked: 2026-09-24
MIT AI Governance Map
Collects and classifies AI governance policies, principles, standards and frameworks from around the world.
Evidence use: Supports searching coverage of governance instruments, actors, regions and topics.
Limitations: The database is continuously updated; statistics should include the access date. Aggregate taxonomies cannot replace a risk assessment of a specific system.
Related concepts: AI Governance and Regulation
Link last checked: 2026-09-24
AI Incident Tracker: June 2026 update
Explains the incident tracker's data volume, classification approach and the substantive changes in this update.
Evidence use: Supports explaining the current incident database's coverage and version boundaries.
Limitations: The database is continuously updated; statistics should include the access date. Aggregate taxonomies cannot replace a risk assessment of a specific system.
Related concepts: AI Governance and Regulation
Link last checked: 2026-09-24
Mapping the AI Governance Landscape: April 2026 update
Explains the governance map's scope of sources, classification dimensions and findings to date.
Evidence use: Supports explaining the governance database's methodology and current coverage.
Limitations: The database is continuously updated; statistics should include the access date. Aggregate taxonomies cannot replace a risk assessment of a specific system.
Related concepts: AI Governance and Regulation
Link last checked: 2026-09-24