− HEALTH We are the last generation that will influence what wellbeing means before AI systems optimize that definitio…− HEALTH Abstract There is much discussion of the false outputs that generative AI systems such as ChatGPT, Claude, Ge…± SCIENCE This article presents a Delphi consensus developed by a panel of editors-in-chief of anaesthesiology and pain…+ EDUCATION The study investigated how innovative, AI-supported teaching methods relate to the mathematics problem-solvin…+ EDUCATION As generative AI becomes increasingly integrated into higher education, understanding how students engage wit…+ HEALTH In the complex and multidimensional field of medicine, multimodal data are prevalent and crucial for informed…± SCIENCE This paper explores the evolution of geoscientific inquiry, tracing the progression from traditional physics-…− POLICY This paper conducts a systematic review and interdisciplinary analysis of the ethical challenges of generativ…
TruaceTracing the truth around AIMonday, July 20, 2026
Subscribe
Business·The Trace·Automated dual reading·Published 2026-07-20

AI system risks and governance

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

The risks posed by artificial intelligence (AI) concern academics, auditors, policymakers, AI companies, and the public. Researchers, policymakers, and technology companies discuss AI risks using inconsistent terminology-the same word may describe different problems, while different words describe identical concerns. This fragmentation impedes coordinated responses to AI challenges. We address this by creating the AI Risk Repository: a living database of 1,725 risks extracted from 74 existing taxonomies and fram…

TRV-2026-0344Peer-reviewedPermanent record — cite & verify
Trace impact reading

Contested: both sides are scored from claims and sources, not community votes.

P 70The P score combines the specificity and measured human impact of the grounded problem claim with the strength of this Trace’s cited sources.G 72The G score combines the specificity and measured human impact of the grounded gain claim with the strength of this Trace’s cited sources.
The AI risk repository: A meta-review, database, and taxonomy of risks from artificial intelligence

Application of the Wenger Taxonomy for classifying goods procured by the Federal Government to Commercial off-the-shelf computer hardware equipment by Harrison, William M.. Public domain

The quick read

On March 30 2026, a peer-reviewed paper in Patterns described the AI Risk Repository, a living database of 1,725 risks extracted from 74 existing taxonomies and frameworks. The authors created two complementary systems to organize them: a Causal Taxonomy by origin, intent and timing, and a Domain Taxonomy by effects across seven areas.

The work matters because fragmented language around AI risks makes auditing and governing difficult for academics, companies and policymakers. The repository offers a common reference to coordinate research and oversight, but the source does not report measured reductions in harm or adoption outcomes, only the structure and intent of the database.

Main points
  • Authors extracted 1,725 risks from 74 existing taxonomies and frameworks to build the AI Risk Repository as a living database.
  • Causal Taxonomy classifies risks by origin: which entity causes them (human or AI), whether intentional, and timing before or after deployment.
  • Domain Taxonomy classifies risks by effects across seven areas including discrimination, privacy violations, misinformation and weapons development.
  • Fragmented terminology where same word describes different problems impedes coordinated responses, motivating the shared reference.
Gain

Creation of a living database of 1,725 risks from 74 taxonomies provides a shared reference that enables more coordinated auditing and governing of AI systems.

Problem

AI systems pose risks across seven effect areas ranging from discrimination and privacy violations to misinformation and weapons development that concern auditors, policymakers, companies and the public.

The rundown

By March 30 2026, the authors had compiled 1,725 risks from 74 taxonomies into a living database, noting that inconsistent terminology impedes coordinated responses.

They structured the repository with a Causal Taxonomy focused on entity, intent and timing, and a Domain Taxonomy focused on effects, intended for use in discussing, researching, auditing and governing across sectors and jurisdictions.

Sources

Reader signal

How should this claim be treated?

The debate