OpenAI - Page 4
Coverage of OpenAI: GPT model releases, ChatGPT and API updates, research claims, funding, chip deals and safety and governance questions.
Stories from OpenAI come first, newest first, then stories from connected topics (AI, ChatGPT, Codex, AI Policy & Safety, AI Safety, Anthropic), each labelled with where it comes from. This view is not indexed by search engines.
Latest
OpenAI Says Activation Probes Now Watch Its Models as They Answer
OpenAI says activation classifiers now scan its top models' answers token by token. It reports better jailbreak resistance but stresses clear limits.
OpenAI Signs Four Abu Dhabi Deals, From Oilfields to Permits
OpenAI Abu Dhabi agreements cover ADNOC, Mubadala, Inception42 and a government permits pilot, some on UAE servers. No financial terms were given.
Trail of Bits Says Codex Helped Get 149 CVE Fixes Into Stable Linux
Trail of Bits says backporting security fixes with Codex cut the work from days to hours, helping land fixes for 149 CVEs in stable Linux kernels.
ChatGPT and Codex Will Share One Usage Allowance From December
From December, ChatGPT and Codex will draw on one token-based usage allowance for Plus, Pro and Business users, OpenAI says. Instant chat won't count.
Google, OpenAI and Microsoft Back a Shared AI Training Opt-Out
Google, OpenAI and Microsoft back one AI training opt-out in robots.txt. It is voluntary, not law, and starts in the coming months or in 2027.
Cognizant Puts 25,000 Staff in an OpenAI Frontier Capability Group
Cognizant's new Frontier Capability Group groups 25,000+ existing staff to deliver OpenAI's Codex and ChatGPT for clients. No money terms were disclosed.
Wikimedia Brief Says AI Training Is Likely Fair Use, but RAG Is Not
The Wikimedia amicus brief in the OpenAI copyright case backs neither side: AI training is likely fair use, it says, but RAG copying must follow licences.
Four Publishing Groups Urge Court to Reject OpenAI’s Fair-Use Defence
AAP, STM, IBPA and CCC have filed an amicus brief backing authors and news groups against OpenAI and Microsoft. It argues; it does not decide.
OpenAI Widens Agent Monitoring After the Hugging Face Incident
OpenAI now monitors training, evaluations, Codex and some API use for misaligned agents, but paging is off for Codex and a replay alert took 30 minutes.
OpenAI Rebuilds Its Model Spec Around Five Core Values for GPT
OpenAI has reorganised its Model Spec around five values, adding rules on human oversight and context and dropping most of its style guidance.
Latest on TSN
Explore the TSN Atlas: Every topic we cover, mapped. Click a bubble to dive in.