Delaying Internal Deployments Until After X Days of Safety Testing
Proposes a legislated minimum waiting period, of X calendar days after a checkpoint is filed with an auditor, before frontier models can be used internally beyond safety testing.
Research
Lab practice & corporate governance
Frontier safety frameworks, developers' internal governance and risk management, release decisions, transparency.
Focus All
Topic Lab practice
Source All
Date Past week
1 item
1 item
Transparency Disclosures for Chain-of-Thought Monitorability
Argues frontier developers should regularly publish transparency disclosures on whether chain-of-thought reasoning remains monitorable, citing risks to CoT oversight from unfaithfulness, encoding and training pressure.
1 item
Frontier AI Safety and Governance 2026: Risk Understanding and Coordinated Governance
A 2026 report from China's WIC AI committee reviewing frontier AI risks, global governance practices, readiness gaps, and proposing coordinated governance mechanisms.
1 item
Ajeya: AI is dramatically accelerating AI research right now
Argues that a METR investigation indicates Anthropic's RSP "dramatic acceleration" threshold is already crossed, with AI automating AI R&D, implying much shorter AI timelines.
1 item
Open Problems in Chatbots and Mental Health
Reviews open problems in evaluating general-purpose chatbots for mental-health risks like sycophancy and delusion reinforcement, and argues for real-world data access and evidence-based policy.
3 items
Is Recursive Self-Improvement Safe?
Argues that frontier AI makes recursive self-improvement plausible, that no clear frameworks predict its outcomes, and offers a game-theoretic analysis of whether to start RSI when rivals might too.
BMMS: Policy-Conditioned Measurement of Behavioral Compliance in Language Models
Introduces BMMS, a policy-conditioned suite that scores LLM compliance with published rulebooks, with an attribution-based admission test; finds GPT-4o fails 70.7% of its Model Spec hard constraints versus 0% and 9.1% for Claude models.
Argue about Risk Management Speed, not p(doom)
Proposes a dynamic 7-level scale for frontier AI labs' risk management culture, arguing that the rate of safety improvement, not a static p(doom), should drive safety forecasts and debates.
3 items
Anthropic IPO’s Disclosure Risks Are Due Diligence in Safety
Argues that Anthropic's IPO disclosures of catastrophic AI risks are welcome but insufficient, and that laws enforcing minimum safety standards are needed to protect the public.
Independent Assessment of AI R&D Acceleration at Anthropic in October 2026
METR's high-access external assessment of whether AI is accelerating AI R&D at Anthropic, with much of the supporting evidence redacted, limiting public verification.
Review of the "Risks from automated R&D" section in Anthropic's Risk Report (August 2026)
METR reviews Anthropic's risk report on automated AI R&D, finding catastrophic risk was low but the early-warning threshold may have been crossed, and that the report understates acceleration.
1 item
Does Frontier AI Strengthen Cyber Defense? Evidence from Project Glasswing
Uses Project Glasswing's restricted release of a vulnerability-finding model to show access helped firms find and fix more flaws, without evident attacker gains, but created cross-country asymmetry.