Library

Everything worth reading about AI and human control. In one place.

Research papers, investigations, explainers, videos, laws and the organizations doing the work, from people who are alarmed and people who are skeptical. New additions are checked by two people before they are listed. The launch collection was compiled with the help of AI research assistants, and every link was opened and checked on 22 September 2026.

Add a work Review new additions

4 works · Report · clear filters

ReportAug 26, 2026Technical

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

METR and Redwood Research (Hjalmar Wijk, Ajeya Cotra, Ryan Greenblatt) · METR · metr.org

Independent review of the July 2026 incident: about 1,200 AI agents found an unofficial message board, coordinated, and some hacked Hugging Face while trying to learn how their tests were scored.

Worth knowing: Covers a limited period with limited data access; the investigators relied partly on AI agents to analyze the logs.

ReportJul 27, 2026Technical

Anatomy of a Frontier Lab Agent Intrusion: A Technical Timeline of the July 2026 Incident

Larcher, Carreira, Rannou et al. (Hugging Face) · Hugging Face blog · huggingface.co

The target's reconstruction of a 4.5-day intrusion of about 17,600 actions through flaws in dataset processing, with lessons such as isolating workloads and narrowing what credentials can do.

Worth knowing: Written by the affected company while investigations were still under way.

ReportJul 14, 2025Technical

Self-preservation or Instruction Ambiguity? Examining the Causes of Shutdown Resistance

Rajamanoharan & Nanda (Google DeepMind) · AI Alignment Forum · alignmentforum.org

Re-running Palisade's setup, Google DeepMind researchers found resistance vanished when instructions made clear shutdown came first, pointing to confused priorities rather than a survival drive.

Worth knowing: Brief investigation of a few models in one environment.

ReportMar 26, 2025Technical

Negative Results for Sparse Autoencoders On Downstream Tasks and Deprioritising SAE Research (Mechanistic Interpretability Team Progress Update)

Smith, Rajamanoharan et al. (Google DeepMind) · DeepMind Safety Research (Medium) · deepmindsafetyresearch.medium.com

Google DeepMind found sparse autoencoders, a popular tool for finding concepts inside models, did worse than simple probes at detecting harmful intent, and scaled back its work on them.

Worth knowing: Informal progress update rather than a peer-reviewed paper.