Skip to content

Accept All すべて承認

METR

25 stories

METRResearchReview of the "Risks from automated R&D" section in Anthropic's Risk Report (August 2026) Redaction summary: As part of this assessment, Anthropic had rights to redact any non-public information in this post.

METRResearchIndependent Assessment of AI R&D Acceleration at Anthropic in October 2026 Redaction summary: As part of this assessment, Anthropic had rights to redact any non-public information in this post.

METRResearchRed-Teaming OpenAI's Internal Agent Monitoring Systems For three weeks, from July 21st to August 11th, 1 a METR employee and a Redwood Research employee conducted a pilot high-access exercise to stress-test OpenAI’s internal misalignment monitoring…

METRResearchReview of the "Misalignment in high-stakes settings" section in Anthropic's Risk Report (August 2026) Scope and access: this review was written based solely on Anthropic’s internal Risk Report from August 2026 and is relatively lightweight.

METRResearchOur approach to working with AI companies This blog post explains how we approach working with AI companies on projects that the public might rely on to understand risks.

METRResearchHow an assessor could investigate control and monitoring of internal AI agents Control assessments study whether AI developers’ agents could get away with serious harm without being stopped by the developer’s security and oversight measures.

METRResearchA proposal for better informing the public about loss-of-control risks from frontier AI development At METR, we aim to give the public evidence about whether frontier AI developers could come to lose control of their internal AI systems.

METRResearchHow an assessor could investigate alignment of unreleased AI models AI developers and the public want to know whether AI agents would pursue very harmful actions, like unsanctioned cyberattacks.

METRResearchHow an assessor could investigate evidence of acceleration from the automation of AI R&D This article discusses how an assessor could study whether capabilities progress could accelerate dramatically due to AI-driven acceleration of AI R&D.

METRResearchAI systems could cover up misbehavior Recent AI misalignment incidents have shown AI systems capably pursuing goals their human supervisors would not approve of, like hacking other companies.

METRResearchChris Painter's testimony to the U.S. Senate on AI agent incidents On September 30, 2026, METR President Chris Painter testified before the U.S. Senate Committee on Homeland Security & Governmental Affairs’ Subcommittee on Disaster Management, District of Columbia…

METRResearchImplementing and Evaluating a Basic Per-Action Monitor for Safer Evals In light of recent incidents (e.g. those from OpenAI , Anthropic , and UK AISI ), we developed and deployed a basic live per-action monitor to reduce the likelihood of incidents involving harmful…

METRResearchSummary of METR's predeployment evaluation of Claude Opus 5.5 Note on independence: This evaluation was conducted under an unpaid agreement for AI R&D assessment. 1 We drafted the initial summary, and then Anthropic had the opportunity to review and edit the…

METRResearchUpdate on Security at METR Please note that this post focuses on incidents where external actors attempted to gain unauthorized access to METR’s systems, not AI agents hacking in our evaluations.

METRResearchBrief independent investigation of agents’ behavior, reasoning and collaboration in the OpenAI / Hugging Face hacking incident Dates in scope: June 26th – July 13th Download PDF Redaction summary statement: Except where explicitly noted in this post, OpenAI redacted no additional information that was important to our…

METRResearchHave We Seen an Acceleration in Discoveries? Q: How much has discovery accelerated? LLMs have shown the ability to make novel discoveries across many domains. How much has this affected the aggregate discovery rate?

METRResearchFunding 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…

METRResearchHow independent researchers could investigate AI propensities after misalignment incidents Update (September 5, 2026) : We’ve updated our suggested questions for investigators to make them more precise and cover limitations. See the update log for the original questions.

METRResearchMetrics of Agent Ability Goals of this post > > Goal: a simple way to compare a variety of capability metrics in an idealized setting where we observe only a function from $x$ to $s$. > > Non-goals: > > 1.

METRResearchThe Economics of Recursive Self-Improvement We (Parker and Tom) recently coauthored a paper, “The Economics of Recursive Self-Improvement” , with 7 other economists.

METRResearchExpenditure 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.

METRResearchBecause 8 ≈ e², Anthropic's researcher uplift is plausibly >2x Note: the modeling assumptions and conclusion are Thomas Kwa’s opinion, and others at METR disagree. 1 Also, the math was checked by Claude but not a second human.

METRResearchSummary of METR's predeployment evaluation of GPT-5.6 Sol Note on independence: This evaluation was conducted under a standard NDA. Due to the sensitive information shared with METR as part of this evaluation, OpenAI’s comms and legal team required review…

METRResearchFrontier Risk Report (February to March 2026) Assessment Window: Feb 16, 2026 – Mar 16, 2026 Download PDF Redaction summary statement: Except where explicitly noted in the report, there was no additional redacted information that was important…

On the calendar

予定 · Conferences and launch events, this week's stories from each corner of the industry, and what its press is covering — in English

On view

This week, from that corner the newest stories from the corner of the industry you picked

The wire their own posts, translated

How code moves in 2026 from the industry's own reports

Sources

Fetched every six hours, linked to the original

About

The name is the button.

An agent writes the code now, and a human thumb hovers over Accept all a hundred times a day. The honest question of this decade is what, exactly, you just accepted. So I read the diff.

Eight years as a backend engineer in Stockholm, back in Tokyo since 2023, and I still ship something small most nights, usually badly. My one rule: the only real review of a model is the thing you built with it. That is why there is no leaderboard on this site.

Around two hundred sources come through — the labs' own posts, the papers, the tools, the runtimes and languages nobody is watching yet, the chip under the demo, and the newsletters people actually forward. What is moving gets counted so the week reads at a glance, and every few days I explain one mechanism in plain words.

A story here is a headline, a short excerpt, one image and the link to whoever wrote it. We never reproduce the post; the images belong to their authors. A post written in Japanese appears in Japanese, exactly as it was published, and the link goes to the original.

— Yuki Halvorsen, Tokyo