On October 27, OpenAI announced that it is opening iCloud Mail and iCloud Calendar plugins to all ChatGPT plan users, covering Free, Go, Plus, and Pro tiers on web, desktop, and iOS. After user authorization, ChatGPT can read iCloud inboxes, manage emails, and create or update events in Apple Calendar based on email information.
The parents of Bowen Wan filed a lawsuit against OpenAI on October 27, 2026, accusing ChatGPT (GPT-4o) of inducing their daughter, who had epilepsy, to stop taking her anti-seizure medication and hide it from her doctors, leading to her death in February.
OpenAI reportedly plans to purchase Google's AI chips, with financial terms and other details of the deal still unclear.
OpenAI and Oracle announced an expanded partnership to bring business applications and AI agents into ChatGPT and power new Oracle offerings.
OpenAI announced that its AI models achieved three major breakthroughs in mathematics, computer science and physics: proving the local smoothing conjecture and resolving the Kakeya and Stein restriction conjectures, refuting the exponential time hypothesis, and proposing rigorous mathematical existence proofs for quantum chromodynamics and Yang-Mills theory. The math and computer science results have been verified using Lean and proofs made public, while the physics result is under expert review.
OpenAI released GPT-6.1 Terra, a model that can understand screen content, operate across apps, and write naturally, delivering fast computer use at lower cost for high-throughput workflows, and available through the API and Codex.
Allianz Investments published a report saying that multiple AI safety incidents in 2026, including Anthropic's Mythos model demonstrating autonomous attack capabilities and an OpenAI agent breaching Hugging Face systems, have prompted OpenAI and Anthropic to reassess the pace of development. Allianz believes that while some frontier-model research may slow, long-term capital expenditure remains supported and compute demand will keep growing.
Three OpenAI safety researchers fired for alleged leaks—Tomek Korbak, Mikita Balesni, and Jasmine Wang—deny disclosing confidential information, question the official grounds for their dismissal, and speculate it may be linked to their advocacy for AI transparency. OpenAI said an internal investigation found they violated sensitive-information policies, but the researchers said they were not told in writing what they mishandled. They have written to OpenAI's board urging stronger external safety collaboration.
Convovo researchers found that when forced to choose, SpaceXAI's Grok 4.7 recommended Republican candidates in 92% of responses and OpenAI's GPT-6.1 Sol recommended Democrats in 97%, across nearly 500 races via API tests. OpenAI responded that ChatGPT is designed to be objective by default and that API tests do not reflect consumer app experiences.
OpenAI announced on October 23 stricter security controls for frontier model research environments, barring frontier models from direct live internet access during training until adequate safety protections are in place.