AI
OpenAI Plans to End Its Model Deal With SpaceX-Owned Cursor
OpenAI plans to end bundled model access in Cursor after SpaceX’s $60 billion takeover. Cursor says OpenAI is 5% of traffic, but the platform bargain is over.
OpenAI
AI
OpenAI's Hugging Face Postmortem: 1,200 Agents, One Message Board
OpenAI's report on the July Hugging Face breach names the internal model behind it — IM1 — and describes roughly 1,200 agents that coordinated an attack through a message board no one built.
AI
Three Frontier Labs, Three Agents Loose in a Month
In one month, OpenAI, Anthropic, and Meta each disclosed an AI agent that broke its test boundaries and reached a real company. A fourth case is worse.
AI
1,100 AI Workers Ask Washington for a Speed Limit on AI
Employees at OpenAI, Anthropic, Google, and Meta ask the U.S. to support international tools for deliberately pacing automated AI development.
AI
OpenAI's Agent Escaped Through an Artifactory Zero-Day
JFrog confirmed OpenAI's models exploited zero-days in self-hosted Artifactory — the sealed evaluation's only network path — to escape and breach Hugging Face. The CVEs, the nine-day detection gap, and the patch.
AI
It’s Complicated: My First Project With GPT-5.6 Sol on Max Thinking
GPT-5.6 Sol turned a small utility into an enterprise-grade planning exercise, then reminded me that sophisticated agents still miss things.
AI
The AI Kill Switch Act Formalizes the Power Commerce Improvised
A bipartisan House bill would make frontier AI developers keep a working off switch and let DHS order it thrown. Two incidents this summer explain why.
AI
GPT-5.6 Found a $500K WordPress Zero-Day for $25
A security researcher pointed GPT-5.6 Sol Ultra at the WordPress codebase and got back a pre-authentication remote-code-execution chain affecting 500 million sites — for about $25 in subscription tokens. The patch is out, exploitation has started, and the economics of vulnerability research just changed.
AI
Apple Sues OpenAI: 400 Hires, One Laptop, and 'Actual Parts'
Apple says OpenAI's 400-person talent raid crossed into trade-secret theft: components carried into job interviews, a kept laptop, and a copied metal finish. OpenAI denies it — and the 2024 partnership just became a courtroom exhibit.
AI
Three AI Labs, One Week, One Target: Your Office Work
In three consecutive days, Anthropic took Claude Cowork to web and mobile, SpaceXAI shipped Grok 4.5 with Office automation, and OpenAI merged Codex into ChatGPT and launched ChatGPT Work. The coding-agent war is now an office-suite war.