Artificial Intelligence Policy Tracker

MIT AI Risk Repository · Additional evidence · 17.02.01.b

Compromising privacy by leaking private infiormation

Category: Information Hazards

On this page 3 sections

Description

—

From Ethical and social risks of harm from language models (Weidinger2021), as extracted by the MIT AI Risk Repository (CC BY 4.0).

Classification

Domain
—
Subdomain
—
Causal entity
—
Intent
—
Timing
—

Other entries from Weidinger2021