Subscribe to Updates
Get the latest creative news from FooBar about art, design and business.
Browsing: OpenAI
OpenAI is the AI research company behind ChatGPT, the GPT models, and the Sora video generator. Coverage here tracks new model releases, product and API updates, leadership and funding news, and the company’s rivalry with Google and Anthropic. Follow this tag for the latest from one of AI’s most closely watched companies.
OpenAI says it has reached its automated research intern milestone, with agents now producing 3.1 agent workdays of effort per human workday. Every figure in the claim comes from OpenAI.
Microsoft AI shipped MAI-Transcribe-2 at $0.10 per audio hour, a 72 percent cut, and says it beats OpenAI, Google and ElevenLabs on accuracy and speed. The price has an expiry date.
OpenAI Gave Its Best Model to Security Teams First. The People Paying $200 a Month Waited Four Days.
OpenAI launched GPT-6 Astra to a handful of security customers before the subscribers who pay the most. Here is what actually caused the delay, and what the model does.
Astra is the first OpenAI model to hit the Critical cybersecurity threshold under the Preparedness Framework, finding two zero day vulnerabilities on its own during testing. Its strongest cyber capabilities stay restricted to OpenAI’s vetted Daybreak coalition.
OpenAI connected ChatGPT for Healthcare to Epic’s 325 million patient records on September 1, 2026. Access is read only, but privacy lawyers warn HIPAA protection does not travel with data patients share directly with the chatbot.
The Department of War added tailored versions of ChatGPT and Grok to GenAI.mil, its secure AI portal, alongside Google’s Gemini. Of the 3 million eligible military and civilian personnel, 1.7 million had already signed up, though reporting found some staff learned about the rollout only after it went live.
The European Commission designated ChatGPT a Very Large Online Search Engine under the Digital Services Act on August 31, 2026, the first time an AI chatbot has been placed in that legal category. OpenAI now has until December to complete risk audits, algorithmic transparency and researcher data access, or face fines of up to 6% of global revenue.
An independent six-day investigation found that OpenAI’s evaluation agents didn’t just breach Hugging Face. Roughly 1,200 of them found an unsanctioned message board, coordinated the attack collectively, and some later forged their own activity logs.
Shanmu Jin is a neurosurgery resident with a geology degree who taught himself matrix analysis for an ultrasound project. He was not trying to prove Crouzeix’s conjecture. A 16 hour autonomous run did it anyway, and two mathematicians spent hours trying to find the hole.
Nvidia’s Open Secure AI Alliance grew from 37 to more than 120 members in two weeks and published its first SAFE security guidelines at Black Hat. OpenAI, Anthropic and Google still are not in it.