Results for "risk assessment"
Central log of AI-related risks.
Privacy risk analysis under GDPR-like laws.
Quantifying financial risk.
AI used in sensitive domains requiring compliance.
Risk of incorrect financial models.
Classifying models by impact level.
Grouping patients by predicted outcomes.
Categorizing AI applications by impact and regulatory risk.
Predicting borrower default risk.
Existential risk from AI systems.
Restricting distribution of powerful models.
European regulation classifying AI systems by risk.
US framework for AI risk governance.
Review process before deployment.
Systematic review of model/data processes to ensure performance, fairness, security, and policy compliance.
Governance of model changes.
AI used without governance approval.
Signals indicating dangerous behavior.
Accelerating safety relative to capabilities.
Minimizing average loss on training data; can overfit when data is limited or biased.
Maximum expected loss under normal conditions.
Risk threatening humanity’s survival.
A robust evaluation technique that trains/evaluates across multiple splits to estimate performance variability.
Models evaluating and improving their own outputs.
Plots true positive rate vs false positive rate across thresholds; summarizes separability.
Asking model to review and improve output.
Internal representation of the agent itself.
Framework for identifying, measuring, and mitigating model risks.
Required human review for high-risk decisions.
International AI risk standard.