MIT AI Risk Repository · Risk Category · 66.09.00

-

Category: Privacy and Security

Open the MIT Risk Navigator All entries from Li2025

Description

"AI systems leaking, reproducing, generating or inferring sensitive, private, hazardous, or secured information"

From A Closer Look at the Existing Risks of Generative AI: Mapping the Who, What, and How of Real-World Incidents (Li2025), as extracted by the MIT AI Risk Repository (CC BY 4.0).

Classification

Domain: —
Subdomain: —
Causal entity: AI
Intent: Other
Timing: Other

Other entries from Li2025

- Risk Category · 5.2
Impersonation / identity theft Risk Sub-Category · 4.3
IP / copyright / personality / rights loss Risk Sub-Category · 4.3
Autonomy / agency loss Risk Sub-Category · 5.2
- Risk Category
Political instability Risk Sub-Category · 6.0
Institutional trust loss Risk Sub-Category · 4.1
Economic manipulation Risk Sub-Category · 4.3
- Risk Category · 3.1
Propagating misconceptions / false beliefs Risk Sub-Category · 3.1
Pollution of information ecosystems Risk Sub-Category · 3.2
Erosion of trust in public information Risk Sub-Category · 3.2