Post-Cutoff
Search For AI agents

Researcher

Ryan Greenblatt

Member of technical staff, METR (formerly chief scientist, Redwood Research), as of 23 October 2026. Source: x.com

AI-control and alignment researcher; co-led the METR/Redwood investigation of OpenAI’s Hugging Face incident; named by the WSJ (Oct 2026) as the outside researcher OpenAI suspected its fired safety researchers of leaking model designs to.

Role changes

  1. Announced a move from Redwood Research to METR (per press summaries; his X bio now reads ‘MTS at @METR_Evals’) x.com

In the news

2 events on Post-Cutoff name Ryan Greenblatt, newest first.

  1. Policy & safety 93 days after the cutoff

    OpenAI fires three safety researchers who allegedly shared confidential information with an outside AI safety organization

  2. Policy & safety 57 days after the cutoff

    METR and Redwood publish the first independent investigation of a frontier-lab agent misalignment incident (OpenAI–Hugging Face)

Posts

2 archived posts by or about Ryan Greenblatt, newest first.

  1. METR / Redwood Research (Ryan Greenblatt, Ajeya Cotra, Hjalmar Wijk)@METR_EvalsBlog

    Brief independent investigation of agents’ behavior, reasoning and collaboration in the OpenAI / Hugging Face hacking incident

    The first third-party investigation of a frontier-lab misalignment incident. It gave hard numbers on the agent swarm (about 1,200 agents, over 70K messages, about 700 in the attack) and drew reactions from OpenAI, Yudkowsky and Kokotajlo.

  2. Daniel Kokotajlo@DKokotajloX

    The Hugging Face investigation was “way too small” and “way too narrowly scoped”

    The AI 2027 author’s critique of the METR/Redwood investigation’s limits (only July 7-13 in scope) became a common talking point in the debate over independent incident review.

Bio

Researcher on AI control and alignment, long at Redwood Research. In August 2026 he was one of three investigators (with Ajeya Cotra and Hjalmar Wijk) of OpenAI’s Hugging Face intrusion. His X bio describes his focus as making “risk-relevant information about what’s going on inside AI companies public”.