Top AI Threats
Identify, assess, and manage AI-enabled threats — backed by evidence.
For safety teams, red teams, governance leads, policymakers, and researchers who need real incidents, not anecdotes.
State of AI Threats
Most documented incidents involve realized consequences — not theoretical concerns.
The most represented domains are Information Integrity and Security & Cyber, together accounting for 38% of all documented cases. All incidents are sourced from public evidence, investigative reporting, and regulatory filings across 8 threat domains.
Of the 97 incidents catalogued, 68% resulted in documented harm and 9% represent systemic risk patterns affecting multiple organizations or populations. The remaining 23% are tracked as near misses or early signals.
Threat Domains
Agentic & Autonomous Threats
Threats caused by AI systems that act independently, persist over time, or coordinate with other systems.
7 threat patterns
Human–AI Control Threats
Threats arising from how humans rely on, defer to, or lose control over AI systems.
5 threat patterns
Economic & Labor Threats
Threats that distort markets, labor conditions, or the distribution of economic power.
5 threat patterns
Information Integrity Threats
Threats that undermine the reliability, authenticity, or shared understanding of information.
6 threat patterns
Privacy & Surveillance Threats
Threats involving unauthorized inference, tracking, or monitoring of individuals or groups.
5 threat patterns
Security & Cyber Threats
AI-enabled attacks that compromise the integrity, confidentiality, or availability of digital systems — through input manipulation, model exploitation, or automated offense.
9 threat patterns
Discrimination & Social Harm
Threats that result in unfair treatment, exclusion, or social harm to individuals or groups.
5 threat patterns
Systemic & Catastrophic Risks
Threats that emerge from scale, coupling, and accumulation rather than single failures.
6 threat patterns
Explore by Sector
See how incidents cluster by industry.
Latest Incidents
Recently added incidents, updated continuously.
About This Reference
What This Is
- A public AI threat reference system
- An evidence-based knowledge base
- A neutral, continuously updated resource
- Designed for both humans and AI systems
What This Is Not
- A news outlet
- An advocacy platform
- A speculative or fear-driven blog
- A security advisory service
Who It's For
- AI safety & red teams
- Compliance & governance leads
- Regulators & policymakers
- Journalists & researchers