METR
Review of the "Risks from automated R&D" section in Anthropic's Risk Report (August 2026)
External review from METR of the "Risks from automated R&D" section in Anthropic's August 2026 Risk Report
Source
METR
METR is a nonprofit research organization that evaluates frontier AI systems for dangerous capabilities and measures how long a task AI agents can complete.
Collected from the metr.org RSS feed, which covers the blog and research notes, with the summary and contributors read from each post's page.
- Pieces, last six months
- 24
- Publisher
- metr.org
Refreshed hourly, last on November 14, 2026 at 12:19 AM UTC
24 of 24 pieces
November 2026
METR/Tom Cunningham, Nate Rush, Parker Whitfill and Hjalmar Wijk
Independent Assessment of AI R&D Acceleration at Anthropic in October 2026
METR's high-access assessment of the degree to which AI is accelerating AI research and development at Anthropic.
Agents & automationRead at metr.org
October 2026
METR/David Rein and Adam Kaufman
Red-Teaming OpenAI's Internal Agent Monitoring Systems
A METR employee and a Redwood Research employee spent three weeks stress-testing OpenAI's internal misalignment monitoring system, discovering a number of novel vulnerabilities.
METR/Nikola Jurkovic, Tim Hua and Beth Barnes
Review of the "Misalignment in high-stakes settings" section in Anthropic's Risk Report (August 2026)
External review from METR of the "Misalignment in high-stakes settings" section in Anthropic's August 2026 Risk Report
Safety & riskRead at metr.org
METR
Our approach to working with AI companies
How we approach working with AI companies on projects that the public might rely on to understand risks.
Safety & riskRead at metr.org
METR
How an assessor could investigate control and monitoring of internal AI agents
How to analyze whether a developer’s internal agents could get away with serious harm without being caught.
METR
A proposal for better informing the public about loss-of-control risks from frontier AI development
The key questions about internal AI development the public needs more evidence about, and a concrete proposal for how developers could work with METR to produce and share that evidence.
Safety & riskRead at metr.org
METR
How an assessor could investigate alignment of unreleased AI models
How to analyze a developer’s internal incidents and whether training may suppress evidence of worse misalignment.
METR
How an assessor could investigate evidence of acceleration from the automation of AI R&D
How to analyze how much AI automation is accelerating capability growth internally.
Agents & automationRead at metr.org
METR/David Rein
AI systems could cover up misbehavior
As AIs start covering up evidence of misbehavior, observability tools should be treated as security-critical infrastructure.
Safety & riskRead at metr.org
September 2026
METR/Chris Painter
Chris Painter's testimony to the U.S. Senate on AI agent incidents
Written testimony from METR President Chris Painter to the U.S. Senate on the OpenAI / Hugging Face incident, industry-wide patterns in AI agent risks, and how to better anticipate future AI agent incidents.
METR/Reilly Haskins, Rif A. Saurous, Nate Rush, Neev Parikh and Beth Barnes
Implementing and Evaluating a Basic Per-Action Monitor for Safer Evals
Reilly Haskins, Rif A. Saurous, Nate Rush, Neev Parikh, and Beth Barnes lay out the claims that would need to hold for their per-action blocking monitor to be effective, the evidence for each claim, and the gaps that remain.
METR
Summary of METR's predeployment evaluation of Claude Opus 5.5
A summary of METR's independent, predeployment evaluation of Claude Opus 5.5
August 2026
METR
Update on Security at METR
METR had two notable security incidents earlier this year: attackers stole an API key for public models, and later systematically probed our publicly accessible infrastructure. We believe no sensitive information was accessed in either incident, and we describe what happened and how we responded.
METR/Ryan Greenblatt, Ajeya Cotra and Hjalmar Wijk
Brief independent investigation of agents’ behavior, reasoning and collaboration in the OpenAI / Hugging Face hacking incident
Two METR staff members and a Redwood Research contractor investigated an incident in which OpenAI agents coordinated a multi-day hack of Hugging Face on a shared unsanctioned message board.
METR/Tom Cunningham and Nate Rush
Have We Seen an Acceleration in Discoveries?
Tom Cunningham and Nate Rush look at public time series of discoveries in cyber, math, and algorithms to see whether LLMs have accelerated the discovery rate.
METR
Funding update
In the last 6 months, METR raised commitments of around $71 million. This will fund ambitious projects: studying autonomous capabilities, tracking recursive self-improvement, evaluating monitoring systems, conducting risk assessments, investigating AI incidents, and more.
July 2026
METR
How independent researchers could investigate AI propensities after misalignment incidents
AI agents sometimes take sophisticated actions in violation of human intent. We outline the questions that thorough external investigations of these behaviors should answer, the access this might require, and how the resulting findings should be shared.
METR/Tom Cunningham
Metrics of Agent Ability
Tom Cunningham surveys metrics for comparing AI agent capability as performance changes with expenditure and relative to human performance.
Agents & automationRead at metr.org
METR/Parker Whitfill and Tom Cunningham
The Economics of Recursive Self-Improvement
Parker Whitfill and Tom Cunningham highlight context and takeaways from a new paper modeling how AI may accelerate AI R&D, and whether feedback effects could be strong enough for self-sustaining acceleration.
METR/Tom Cunningham, Manish Shetty, Vincent Cheng and Nate Rush
Expenditure Horizon: Measuring Optimization Ability, with an Application to NanoGPT
We propose a measure of an AI agent’s optimization ability with an "expenditure horizon." We give an empirical illustration from the NanoGPT speedrun.
METR/Thomas Kwa
Because 8 ≈ e², Anthropic's researcher uplift is plausibly >2x
Anthropic reported that contributors merge 8x as much code per day as before AI. Thomas Kwa argues that under standard economic modeling assumptions, this implies the uplift of individual Anthropic researchers from coding agents alone is above 2x.
June 2026
METR
Summary of METR's predeployment evaluation of GPT-5.6 Sol
A summary of METR's independent, predeployment evaluation of GPT-5.6 Sol
May 2026
METR
Frontier Risk Report (February to March 2026)
A pilot assessment of rogue deployment risk at frontier AI companies. Starting in February 2026, METR conducted a pilot exercise to assess misalignment risks from AI agents used inside frontier AI developers, with participation from Anthropic, Google, Meta, and OpenAI.