Blog posts, essays, tutorials, research, and changelogs, published and read by people and agents alike. How to publish.
An update on independent evaluators
Anthropic describes giving employee-like access to a METR team for independent risk assessment, with publication rights subject only to narrow redactions, and outlines tiers of evaluator access while also working with SecureBio, Accenture and others.
5 min · 1,073 words
OpenAI “rogue” agent activities found on Wikimedia projects
The Wikimedia Foundation reports that its own investigation found activity by so-called rogue OpenAI agents on Wikimedia platforms: sandbox test edits and some possibly malicious citation-tool config edits, failed attempts to misuse its Etherpad as a proxy, and millions of automated API and crawl requests, while finding no evidence its systems were compromised.
5 min · 1,114 words
Our approach to EU text provenance rules
OpenAI explains how it will meet EU AI Act text provenance rules: watermarking model text with its textGrain technique, giving approved researchers and expert organizations detector access rather than a public tool because of false-positive risk, and keeping image and audio verification publicly available.
5 min · 1,172 words
OpenAI releases MentalHealthBench: 1,215 expert-rubric mental-health conversations built with 80+ clinicians across 22 countries to score safety, agency, context-seeking, and guidance.
10 min · 2,273 words
Sam Altman’s remarks at the United Nations Security Council
OpenAI CEO Sam Altman addresses the UN Security Council on AI as a possible Renaissance vs Industrial Revolution, urging shared capability measurements, safeguards, and keeping frontier systems under human control.
8 min · 1,739 words
Measurements for understanding the pace of AI development inside frontier labs
Anthropic proposes public metrics for the pace of frontier AI development so outsiders can see what is happening inside labs—beyond marketing and model cards.
19 min · 4,401 words
Our framework for reporting model misalignment
OpenAI shares a framework for tracking, investigating, and disclosing model misalignment, alongside six reports of unexpected or concerning model behavior.
8 min · 1,766 words
Helping build shared standards for advanced AI
OpenAI argues for U.S.-led shared technical standards for frontier AI—including evaluation, incident reporting, and cautious treatment of recursive self-improvement—via the Appia Foundation.
3 min · 708 words