4. Malicious Actors & Misuse2 - Post-deployment

Defamation/libel/slander

Defamation/libel/slander - Use of a technology system to create, facilitate or amplify false perception(s) about an individual, group, or organisation.

Source: MIT AI Risk Repositorymit952

ENTITY

1 - Human

INTENT

1 - Intentional

TIMING

2 - Post-deployment

Risk ID

mit952

Domain lineage

4. Malicious Actors & Misuse

223 mapped risks

4.3 > Fraud, scams, and targeted manipulation

Mitigation strategy

1. Conduct rigorous pre-deployment red-teaming and model audits to proactively identify and mitigate high-risk outputs, particularly those generating personal profiles, summaries, or answers about individuals and organizations. 2. Implement robust notice-and-takedown protocols to establish rapid-response processes for investigating and correcting demonstrably false outputs once they are reported. 3. Maintain detailed provenance tracking logs of user prompts, model outputs, and moderation actions to create a comprehensive audit trail that can establish good-faith efforts and aid in assessing liability.