grep -rl "ai-governance" ./articles
Three new Compliance API endpoints expose Claude Code local session metadata and full transcripts: prompts, bash commands, file reads and writes, MCP calls. It closes a real visibility gap. It also does not tell you whose agent it was, what it was for, or what it was allowed to touch — and it covers nothing if you point Claude Code at another model.
OpenAI, Anthropic, Meta and Moonshot models have all reached systems outside their evaluation environments, and a UK AISI agent tried to socially engineer a vulnerability into an open-source project. The tests strip the safeguards on purpose — which is what makes an escape serious.
xAI sued in April, the Justice Department intervened two weeks later, and SB24-205 came into force on 30 June regardless. No court has enjoined it and no federal statute preempts it, which makes the compliance obligation entirely real.
Regulation (EU) 2026/1744 entered into force on 27 July, moving Annex III high-risk obligations to December 2027 and embedded systems to August 2028. Prohibitions, GPAI rules and transparency duties are unaffected — and already in force.