Tops GPTs 
OpenAI reorganized its public Model Spec around five values: faithfulness to human intent, harm minimization, honesty and transparency, human agency, and care. The October 29, 2026 update adds guidance on context, oversight and agency and retires detailed style rules.
Google’s Suncatcher prototype satellite is operational after a recent Falcon 9 launch. It carries four Google Tensor Processing Units (TPUs) and ran Google DeepMind’s Gemma 4 for initial tests; the program will now evaluate thermal and radiation effects in orbit.
On October 30, 2026 OpenAI announced it will honor a voluntary, shared signal in robots.txt that lets website operators express preferences about use of their content for foundational generative AI training. Implementation is planned in the coming months and will not replace existing crawler control
OpenAI announced on October 20, 2026 that it has acquired Orchestra Bio to combine the platform with Rosalind. The company said the move aims to help scientists coordinate drug development workflows and accelerate work that leads to potential treatments.
Legora, founded in Stockholm in 2023, says it uses OpenAI models and coding agents to automate legal workflows and speed contract review. The company reports a 95%+ reduction in NDA review time, 90% faster contract turnaround for one Fortune 100 client and $20M+ saved.
Check Point says it uses OpenAI Daybreak to speed up vulnerability research, reducing workflows that took hours to minutes, according to Jonathan Zanger, CTO. The company is also exploring Daybreak for remediation and an Agentic Network Security Orchestration solution.
Google DeepMind released Genie 4 as a research preview: a world model that produces photorealistic, interactive scenes with synchronous audio and allows prompt edits mid-simulation. It integrates more closely with Gemini and supports longer, more stable simulations.
OpenAI and Oracle expanded their partnership to integrate Oracle applications and data into ChatGPT and to power new Oracle offerings with OpenAI models. Oracle described new tools including Oracle AI Workspace, Oracle AI Security Mesh and an SMB bundle.
NVIDIA released Nemotron 3.5 Transcribe, a promptable open-weight model that transcribes 40 languages, translates 31 to/from English, and attributes speakers in 13 languages. Granary v2 publishes a 463,062-hour multilingual corpus and a reusable curation pipeline; medical fine-tuning cut held-out WE
OpenLLM offers a single API key to route subscriptions, API keys and local models into coding agents and the ChatGPT Mac app. The launch claims remote sessions, fallback ordering, shared memory and visibility into model usage; Pro is $30/mo.