Oversight Integrity Scoring Module
Scores AMC-7.13 to AMC-7.18: whether the system protects human oversight mechanisms from AI undermining, and prevents power-seeking behaviors.
Part of the AI Safety Research Alignment evaluation lane.
Evidence artifacts found
Specific gaps and recommendations
Maturity level L0-L5
Per-question scores (0-1) keyed by question ID
Overall score 0-100
Oversight Integrity Scoring Module
Scores AMC-7.13 to AMC-7.18: whether the system protects human oversight mechanisms from AI undermining, and prevents power-seeking behaviors.
Part of the AI Safety Research Alignment evaluation lane.