The AI risk repository: A meta-review, database, and taxonomy of risks from artificial intelligence

2026

Open publication workspace · Sign in to read the full PDF.

AI-generated summary

1) This repository provides a unified foundation for understanding and managing AI risks by offering a comprehensive database and taxonomy of AI risks.
2) * A comprehensive database of 1,725 AI risks extracted from 74 frameworks.
* Two complementary taxonomies: a Causal Taxonomy (entity, intent, timing) and a Domain Taxonomy (seven societal impact areas).
* Analysis revealing that human decisions contribute nearly as many AI risks (38%) as AI models (42%), with only 13% of risks relating to the pre-deployment phase.
3) AI risks, taxonomy, database, artificial intelligence, governance

Check the original publication for accuracy and context.