This page is still being polished. If you have thoughts, please share them via the feedback form.
Data on this page is preliminary and may change. Please do not share or cite these figures publicly.
Unclassifiable mitigations.
Stage: Containment and Mitigation; Stakeholder: Third Party Researchers; Additional information: AI developers, external researchers and AISIs should prioritise safety and alignment measures, including by building validated safety cases. If risk assessment and prevention are not prioritised, competitive pressures and geopolitical rivalries may push developers to continue deploying potentially misaligned AI models (Mitre & Predd 2025). Developers should collaborate with external researchers, evaluators and AISIs to build on emerging safety case projects for AI development, training and deployment (Irving 2024). Future efforts around safety cases may include independent verification of model characteristics and alignment, evaluation of unintended capabilities, and assessment of worst-case failure modes.
Reasoning
Insufficient detail to identify focal activity; "technical assistance" lacks specificity on mechanism, implementation, or stakeholder role.
Monitor critical capability levels
2.2.2 Testing & EvaluationIdentify early warning signs and emergent capabilities
2.2.1 Risk AssessmentEstablish standardised benchmarks and reporting
3.2.1 Benchmarks & EvaluationImplement compute monitoring and anomaly detection
1.2.3 Monitoring & DetectionEnhance hardware and supply chain oversight
2.3.3 Monitoring & LoggingLead efforts to establish shared criteria for AI LOC
3.2.2 Technical StandardsStrengthening Emergency Preparedness and Response for AI Loss of Control Incidents
Somani, Elika; Friedman, Anjay; Wu, Henry; Lu, Marianne; Byrd, Christopher; van Soest, Henri; Zakaria, Sana (2025)
As artificial intelligence (AI) systems become increasingly embedded in essential infrastructure and services, the risks associated with unintended failures rise. Developing comprehensive emergency response protocols could help mitigate these significant risks. This report focuses on understanding and addressing AI loss of control (LOC) scenarios where human oversight fails to adequately constrain an autonomous, general-purpose AI.
Operate and Monitor
Running, maintaining, and monitoring the AI system post-deployment
Other
Actor type not captured by the standard categories
Manage
Prioritising, responding to, and mitigating AI risks
