Chris Painter's testimony to the U.S. Senate on AI agent incidents
METR's president testifies to the U.S. Senate on observed frontier AI agent incidents, including an investigation of OpenAI agents' behaviour, to inform policymakers.
Research
METR
Measuring frontier AI capabilities and risks. metr.org
Focus All
Topic Security & misuse
Source METR · 8
Date Any time
1 item
1 item
Five lessons from having helped run an AI-Biology RCT
Reflects on a randomized controlled trial of 153 novices testing whether LLMs give wet-lab biology uplift; found no significant end-to-end gain and draws lessons for AI risk policy.
1 item
Summary of our gpt-oss methodology review
METR reviews OpenAI's malicious fine-tuning evaluations of gpt-oss-120b for bio, chem and cyber dangerous capabilities, issuing 17 recommendations before release.
1 item
Response to OSTP on AI Action Plan
METR's policy response to OSTP recommends US frontier AI governance: security controls, formal capability-measurement standards for CBRN, cyber, autonomy and AI R&D, and reporting or oversight requirements.
1 item
AI models can be dangerous before public deployment
Argues that pre-deployment testing in frontier safety frameworks misses risks from model theft, internal misuse, and misaligned internal agents, and that safety policy must cover internal use.
1 item
The Rogue Replication Threat Model
METR's threat-model post argues there are no decisive barriers to rogue AI agents replicating at scale by earning revenue and acquiring GPUs, and analyses the risks this poses.
1 item
Response to Bureau of Industry and Security’s proposed AI reporting requirements
METR's response to a US Bureau of Industry and Security proposed rule on reporting requirements for advanced AI models and computing clusters, offering red-teaming and security suggestions.
1 item
Response to U.S. AISI Draft “Managing Misuse Risk for Dual-Use Foundation Models”
METR's comments on the U.S. AISI draft on misuse risk for dual-use foundation models, recommending expanded guidance on capability elicitation and robust model safeguards.