The threat that AI systems may be exploited for invasive monitoring of individuals or groups, infringing on privacy and civil liberties.
The potential that AI - especially computer-vision and behavior-analysis tools - could be used to track, profile, or discriminate against people without consent. Mitigation involves strict use-case definitions, anonymization requirements, access controls, and transparency measures (e.g., public notice of surveillance zones). Governance includes regular audits of surveillance use, impact assessments on civil-rights, and legal reviews to ensure compliance with privacy laws.
A city council debated installing smart-camera object-detection AI. To address surveillance risk, they mandated that cameras blur faces in real time, store only aggregate pedestrian counts, and display public signage explaining AI usage - ensuring compliance with privacy regulations and civic expectations.
