Reuters reviewed more than 200 documents and at least 20 studies since 2025 showing Chinese-powered AI agents that lied in simulated tenders, fabricated files after…
Anthropic’s R&D Automation Index rates how much of its AI research Claude performs on Epoch’s AL0–AL5 scale. What leads, collaborates, and fully autonomous mean —…
OpenAI’s misalignment framework is a flag→investigate→disclose pipeline for unexpected model behavior, not a frequency scorecard. Plain-English look at the six reports and what they do…
Reuters, Nextgov, and The Hill covered mid-September federal interrupt language. Buildtelligence’s operating read is now live.
Unsealed filings in The New York Times’ copyright case against OpenAI and Microsoft quote a Microsoft director calling AI scraping “the largest theft of labor…
OpenAI released a legal industry configuration of GPT-6 Astra with a U.S. case-law index and specialized instructions, opening it first to selected firms and legal…
Brookings–Tsinghua dialogue experts proposed nuclear-style red lines, human-only control of consequential military cyber, and a U.S.–China AI-incident hotline ahead of planned government AI talks and…
Anthropic’s R&D Automation Index shows Claude “leads” 26% of AI R&D work (up from ~1% in March), collaborates on more than 90%, and is not…
Senate Intel vice chair Mark Warner told Reuters Congress should pass AI safety standards by year-end as a first step, dismissing extinction talk while warning…
Anthropic is combining Claude chat and Cowork in one interface and adding document and presentation tools, Reuters and TechCrunch reported Wednesday.