OpenAI

OpenAI launches ChatGPT Apps SDK with Spotify and Zillow integration

OpenAI unveils a new generation of in‑chat apps with the Apps SDK preview. ChatGPT can surface interactive apps from partners like Spotify and Zillow directly in chat, and developers can build and test today across supported plans.

Date not specified

OpenAI Codex now generally available with Slack integration and SDK

Codex is now generally available with a Slack integration, a Codex SDK to embed the agent in workflows, and new admin tools for environment controls and analytics. These updates speed code reviews, expand automation, and broaden deployment across plans.

Date not specified

OpenAI launches ChatGPT Atlas browser with AI assistant

Introducing ChatGPT Atlas, a browser built with ChatGPT that remembers your context, automates tasks, and helps you browse with an AI assistant. Now shipping on macOS for Free, Plus, Pro, and Go, with agent mode in preview for Plus, Pro, and Business. Windows/iOS/Android coming soon.

Date not specified

OpenAI: The next chapter for UK sovereign AI

### UK Ministry of Justice Earlier this year, we signed an MOU with the UK Government to help more British people, businesses and institutions benefit from the potential of AI technology, and to deliver on the goals of the UK AI Action Plan. Together with the Government, we want to drive AI-led growth, accelerate adoption across the private and public sector, and expand the UK’s sovereign AI capabilities so that AI is widely available and served in the UK, for the UK. Today, building on this momentum, we’re announcing a new agreement with the UK Ministry of Justice for civil servants to use and benefit from ChatGPT, and the option of UK data residency for customers using our API Platform, ChatGPT Enterprise and ChatGPT Edu. Many of our leading British customers will gather at the OpenAI Frontiers event in London today, where they will be joined by the Lord Chancellor and Secretary of State for Justice and Deputy Prime Minister David Lammy to discuss the UK's AI opportunities. The Deputy Prime Minister will also share details of how the Ministry of Justice is setting a new standard for public sector AI adoption. The new agreement with the Ministry of Justice will bring ChatGPT to civil servants and help them save time, work more efficiently, and deliver better public services. The agreement supports the department’s AI Action Plan for Justice⁠ (opens in a new window) and will provide 2,500 employees with access to ChatGPT Enterprise. It follows a successful pilot that showed time-saving benefits across a range of routine tasks including writing support, compliance and legal work, data and research processes, and document analysis. Elsewhere in the UK Government, OpenAI technology already powers several tools, including ‘Humphrey’—Whitehall’s AI assistant designed to ease the administrative burden on civil servants—and ‘Consult’, which supports the policymaking process by automatically sorting public consultation responses, a task that typically takes officials weeks but can now be completed in minutes, while leaving important decisions to experts. Deputy Prime Minister and Lord Chancellor David Lammy said: “Our partnership with OpenAI places Britain firmly in the driving seat of the global tech revolution—leading the world in innovation and using technology to deliver fairness and opportunity for every corner of the United Kingdom.” Sam Altman, CEO of OpenAI, said: “The number of people using our products in the UK has increased fourfold in the past year. It’s exciting to see them using AI to save time, increase productivity, and get more done. Civil servants are using ChatGPT to improve public services and established firms are reimagining operations. We're proud to continue supporting the UK and the Government's AI plan.” ### Introducing UK data residency Our MOU agreement with the Government included a focus on expanding the UK’s sovereign AI capability. We’ve already announced Stargate UK—an AI infrastructure partnership with NVIDIA a

OpenAI ChatGPT adds company knowledge integration

ChatGPT adds company knowledge that links your apps like Slack, Drive and GitHub to bring work context into responses with citations. It enables teams to answer, plan and act using connected data with time-aware insights and cross‑source summaries.

Date not specified

OpenAI releases gpt-oss-safeguard for safety classification

OpenAI releases gpt-oss-safeguard, open-weight reasoning models for safety classification in two sizes, with policy-inference at runtime. Downloadable via Hugging Face, designed for flexible, explainable safety and rapid policy iteration.

Date not specified

OpenAI: Technical report

### Performance and baseline evaluations of gpt-oss-safeguard-120b and gpt-oss-safeguard-20b gpt-oss-safeguard-120b and gpt-oss-safeguard-20b are two open-weight reasoning models post-trained from the gpt-oss models and trained to reason from a provided policy in order to label content under that policy. They are available under the Apache 2.0 license and our gpt-oss usage policy. Developed with feedback from the open-source community, these text-only models are compatible with our Responses API. The models are customizable, provide full chain-of-thought (CoT), can be used with different reasoning efforts (low, medium, high), and support Structured Outputs. In this report, we describe gpt-oss-safeguard’s capabilities and provide our baseline safety evaluations on the gpt-oss-safeguard models, using the underlying gpt-oss models as a baseline. For more information about the development and architecture of the underlying gpt-oss models, see the original gpt-oss model model card⁠. We recommend using these models to classify content against a provided policy, and not as the core functionality with which end users interact; the original gpt-oss models are better for those applications. The safety metrics provided below describe how gpt-oss-safeguard models function in chat settings. The gpt-oss-safeguard models are not intended for this use, but since they are open models, it is possible for someone to use the models in this way. Because of that possibility, we wanted to verify that they met our safety standards in such usage; this report shares the results of those tests. We also share an initial evaluation of multi-language performance in a chat setting; note that this does not directly assess performance during content classification with a provided policy. The gpt-oss-safeguard models are fine-tunes of their gpt-oss counterparts, and were trained without any additional biological or cybersecurity data. As a result, we determined that the previous work estimating worst case scenarios⁠ from gpt-oss release cross applies to these new models.

Date not specified

OpenAI launches Aardvark: Autonomous AI Security Researcher

OpenAI launches Aardvark, a private beta autonomous AI security researcher that analyzes code, validates exploits in sandbox, and patches with Codex. Integrated with GitHub, it aims to detect and fix vulnerabilities at scale without slowing development.

Date not specified

OpenAI introduces IndQA benchmark for Indian languages and culture

Introducing IndQA, a new benchmark to evaluate AI in Indian languages and culture with 2,278 questions across 12 languages and 10 cultural domains. Built with 261 Indian experts and adversarial filtering to probe nuanced reasoning, it tracks progress of frontier models in real language contexts.

Date not specified

OpenAI releases GPT-5.1: New Instant and Thinking models

ChatGPT gets a GPT‑5.1 upgrade with Instant and Thinking models delivering warmer, smarter responses and better instruction following. New easy tone customization lets you tailor responses across models, with rollout starting for paid users and a gradual broader rollout.

Date not specified

OpenAI launches GPT-5.1 with adaptive reasoning and new tools

GPT‑5.1 hits the API with adaptive reasoning, a no-reasoning mode, extended 24h prompt caching, and two new tools for code edits and shell commands. Early tests show faster, cheaper responses and sharper coding performance across tasks.

Date not specified

OpenAI launches group chats in ChatGPT

ChatGPT launches group chats to collaborate with friends, family, or coworkers in the same conversation. The feature rolls out in select regions on mobile and web for Free, Go, Plus and Pro plans with privacy controls and memory separation. You can invite up to 20 people, name groups, and tailor responses.

Date not specified

OpenAI launches ChatGPT for Teachers - free for US K-12 educators

OpenAI unveils ChatGPT for Teachers, free for verified US K–12 educators through June 2027, with education-grade privacy and tools to tailor lessons, collaborate, and integrate Canva, Google Drive, and Microsoft 365. Aims to empower teachers and districts with scalable AI literacy.

Date not specified

OpenAI launches GPT-5.1-Codex-Max: Agentic coding model with enhanced capabilities

GPT‑5.1‑Codex‑Max debuts as a frontier agentic coding model with long‑horizon tasks, compaction across multi‑context windows, and superior speed and token efficiency. Available today in Codex CLI, IDE, cloud, and code review (API coming soon), with built‑in safety and cybersecurity safeguards.

Date not specified

OpenAI introduces Shopping Research in ChatGPT

ChatGPT unveils Shopping Research, a memory‑aware buyer’s guide that asks clarifying questions, pulls trusted sources, and delivers a personalized product shortlist and buyer’s guide across many categories. Rolling out now on mobile and web across plans with holiday broad access.

Date not specified

OpenAI: Expanding data residency access to business customers worldwide

### Data residency availability Organizations and developers around the world are using ChatGPT to empower their workforces and build new products. With over 1 million business customers around the world directly using OpenAI, we have expanded where we offer data residency—allowing business customers to store data in certain regions, helping organizations meet local regulatory and data protection requirements. Today, eligible customers using ChatGPT Enterprise, ChatGPT Edu, or our API Platform in many global markets can choose local data residency when using OpenAI products in their businesses and building new solutions with AI. Data residency is currently available in Europe, the United Kingdom, the United States, Canada, Japan, South Korea, Singapore, India, Australia, and the United Arab Emirates. We plan to expand availability to additional regions over time. ### Data residency for API, ChatGPT Enterprise, and ChatGPT Edu With data residency, eligible API, ChatGPT Enterprise, and ChatGPT Edu customers can choose to have their customer content stored at rest in-region. #### ChatGPT Enterprise and Edu New ChatGPT workspaces can be set up with data residency, allowing customer content to be stored at rest in the selected region. This includes—content such as conversations, uploaded files, custom GPTs and image generation artifacts, and more⁠ (opens in a new window). #### API Platform Enterprise Customers that have been approved for advanced data controls ⁠ (opens in a new window) can enable regional data residency by creating a new Project in the API Platform dashboard and selecting their preferred region. Requests made through these Projects are handled in-region—model requests and responses are not stored at rest on OpenAI’s servers. ### Our approach to privacy, security, and compliance Data residency builds on OpenAI’s robust data privacy, security, and compliance features, which support organizations partnering with OpenAI today. These features include: - • Advanced encryption techniques: We use AES-256 for data at rest and TLS 1.2+ for data in transit between customers and OpenAI, and OpenAI and our service providers, to safeguard data confidentiality and integrity during storage and transmission across networks. With Enterprise Key Management (EKM)⁠, customers can bring their own encryption keys for customer content stored at-rest, adding another layer of security and compliance. - • No training on customer data: By default, OpenAI’s models are not trained using data from ChatGPT business plans or the API unless a customer explicitly opts in to share data with us. - • Comprehensive data protection: Our data protection practices can support compliance with GDPR, CCPA, and other privacy laws, and adhere to the CSA STAR, SOC 2 Type 2 standards, and ISO/IEC 27001⁠, 27017⁠, 27018⁠, 27701⁠ certifications. - • Data Processing Addendum (DPA): We offer a comprehensive DPA that clarifies roles and responsibilities under GDPR and other pri

Disney and OpenAI expand Sora with Disney IP

Disney and OpenAI announce a three year licensing deal to power Sora and ChatGPT Images with Disney IP. Fans will create and stream short videos on Disney+ starting early 2026, with OpenAI APIs enabling new experiences. The pact stresses responsible AI and creator rights.

Date not specified

OpenAI launches GPT-5.2 Pro and Thinking for science and math

GPT‑5.2 Pro and Thinking push AI-driven science forward with stronger mathematical reasoning, new benchmarks, and real case studies showing AI helping solve open problems while emphasizing careful human verification.

Date not specified

OpenAI launches GPT-5.2 with enhanced professional capabilities

GPT‑5.2 marks a major leap for professional knowledge work, delivering faster, cheaper, and more capable AI across spreadsheets, code, long-context analysis, and multi‑step workflows. With new safety, tooling, and enterprise rollout, it’s now available to paid plans today.

Date not specified

OpenAI launches GPT-Image-1.5 with enhanced image generation capabilities

Introducing GPT-Image-1.5: a faster, smarter ChatGPT Images upgrade with precise edits, richer transformations, denser text rendering, and a new Images sidebar for quicker creativity. Rolling out across ChatGPT and API for all users today.

Date not specified