Submissions from metr.org

		OpenAI/Hugging Face incident – Independent investigation of agents' behavior [pdf] (metr.org)
		1 point by nns 1 hour ago \| past \| discuss
		Independent investigation of Hugging Face incident - METR (metr.org)
		2 points by Dangeranger 4 days ago \| past \| 1 comment
		METR Report on OpenAI / Hugging Face Hacking Incident (metr.org)
		123 points by stikit 6 days ago \| past \| 106 comments
		Independent investigation of agents' behavior in the Hugging Face incident (metr.org)
		2 points by thunderbong 10 days ago \| past \| discuss
		METR: Funding Update (metr.org)
		1 point by tosh 10 days ago \| past \| discuss
		Have We Seen an Acceleration in Discoveries? (metr.org)
		3 points by tosh 10 days ago \| past \| discuss
		Investigation of agents' behavior in the OpenAI/HuggingFace hacking incident (metr.org)
		9 points by thunderbong 11 days ago \| past \| 1 comment
		Agents' collaboration in the OpenAI / Hugging Face hacking incident (metr.org)
		1 point by dmazin 12 days ago \| past \| discuss
		Investigation of agents in OpenAI / Hugging Face hacking incident (metr.org)
		5 points by giardini 12 days ago \| past \| discuss
		Independent investigation of agents' behavior in OpenAI/Hugging Face incident (metr.org)
		8 points by mellosouls 13 days ago \| past \| 2 comments
		Brief independent investigation of OpenAI / Hugging Face hacking incident (metr.org)
		6 points by Philpax 13 days ago \| past \| 1 comment
		METR's independent investigation of the OpenAI / Hugging Face hacking incident (metr.org)
		5 points by lukaspetersson 13 days ago \| past \| discuss
		Brief independent investigation of agent behavior in OpenAI/Hugging Face hack (metr.org)
		4 points by dwohnitmok 13 days ago \| past \| 1 comment
		The Economics of Recursive Self-Improvement (metr.org)
		1 point by ai2027 48 days ago \| past
		METR introduces Expenditure Horizon metric (metr.org)
		4 points by qntmfred 49 days ago \| past
		We Are Changing Our Developer Productivity Experiment Design (metr.org)
		4 points by Helithumper 53 days ago \| past
		Summary of METR's predeployment evaluation of GPT-5.6 Sol (metr.org)
		10 points by pongogogo 74 days ago \| past \| 6 comments
		AI Cheats [pdf] (metr.org)
		1 point by brian_herman 3 months ago \| past
		Frontier Risk Report (February to March 2026) – METR (metr.org)
		2 points by paraschopra 3 months ago \| past
		Measuring the Self-Reported Impact of Early-2026 AI on Tech Worker Productivity (metr.org)
		3 points by willmarch 3 months ago \| past \| 1 comment
		Task-Completion Time Horizons of Frontier AI Models (metr.org)
		1 point by nsoonhui 4 months ago \| past
		Research note: Fine-tuning experiments on CoT controllability (metr.org)
		1 point by mooreds 4 months ago \| past
		We spent 2 hours working in the future (metr.org)
		3 points by gmays 5 months ago \| past
		Many SWE-bench-Passing PRs would not be merged (metr.org)
		278 points by mustaphah 6 months ago \| past \| 153 comments
		We are changing our developer productivity experiment design (metr.org)
		88 points by ej88 6 months ago \| past \| 61 comments
		Task-Completion Time Horizons of Frontier AI Models (Includes Opus 4.6) (metr.org)
		2 points by admp 6 months ago \| past
		Measuring Time Horizon Using Claude Code and Codex (metr.org)
		1 point by mustaphah 6 months ago \| past
		Task-Completion Time Horizons of Frontier AI Models – METR (metr.org)
		2 points by rootforce 6 months ago \| past
		METR releases Time Horizon 1.1 with 34% more tasks (metr.org)
		1 point by mustaphah 7 months ago \| past
		AI Doubling Time Horizon v1.1 (metr.org)
		1 point by chriskanan 7 months ago \| past
		More