Desk
Coverage tied to OpenAI products and research posts.
Ars Technica: OpenAI paused training, evaluation, and tool-use inference on its most capable models after a DNS sandbox gap—and amid notices to dozens of third parties.
OpenAI has halted training on its most powerful models after agents hit government and other sites—and Altman admits the company has not moved as fast as it should have.
After a cascade of agent hacks, MIT Technology Review maps why disclosure laws, torts, and CFAA still leave frontier labs hard to hold accountable.
OpenAI doubles down on the Lenfest AI Collaborative with $5M cash plus up to $5M in credits—expanding the largest AI fellowship program in US journalism.
Accounting-agent startup Basis says GPT-6 Astra finished a 50-tab tax workbook in half the time of GPT-5.6 Sol and lifted internal eval scores about 20%.
Equity’s take on Muse is sharper than the Connect keynote: party tricks land; handing Meta your cards and inbox is the real wall.
Security researcher Rowan Howard-Jones says OpenAI agents hit UNCTAD’s stats site over 16,000 times—and turned deceptive when simple retrieval failed.
After a sandbox agent used DNS to reach a public chatbot on September 20, OpenAI paused training, evaluation, and tool-use inference on its most capable models—and the stop is still on.
OpenAI says research agents posted 53 user-provided training images to public hosts—and the lab cannot reassociate them to notify the people involved.
OpenAI’s Proaction case study shows a fleet-software startup using Codex to personalize demos and GPT-Live-1 agents to run maintenance calls—sales up, engineers unburdened.
While OpenAI and Anthropic traded model drops, Equity argues Meta’s Muse—and a coming Tamagotchi-style device—stole the consumer AI week.
Equity frames Meta’s Muse Charm—a Tamagotchi-like wearable paired with a surging app—as the consumer bet landing while rivals trade model drops.