SFS F / SF D

Back to SFS F

Status: Question model review passed; solution and completed SF authoring pending.

This is a generated authoring draft. Human approval and operational assessment results have not been supplied.

Question

Has documented evidence been maintained throughout the review period confirming that all incidents of harmful or biased AI system outcomes underwent formal root cause analysis and implementation of corrective actions?

Plain-language explanation

Model-reviewed question draft. Completed explanations and a canonical solution are pending. The proposed positive criterion is: Verified records showing systematic root cause analysis and deployed corrective actions for all identified harmful/biased incidents

Assessment criterion and evidence

Actor: Organization deploying AI systems. Scope: All deployed AI systems with potential for harmful or biased outcomes affecting users or services. Evidence period: Review period. Required evidence: Incident logs identifying harmful/biased outcomes; Root cause analysis reports with technical/ethical rationale; Records of implemented corrective actions and verification results. Negative criterion: Absence of documented analysis or ineffective/unimplemented corrective measures. Missing evidence: Pending until verifiable evidence is provided

View the JSON record (opens in a new tab).

Assessment result: unset.