Capability overhang means too many gaps today between what the models can do and what most people actually do with them. 2026 Prediction: Progress towards AGI will depend as much on helping people use AI well, in ways that directly benefit them, as on progress in frontier models themselves. 2026 will be about frontier research AND about closing this deployment gap — especially in health care, business, and people's daily lives.
About us
OpenAI is an AI research and deployment company dedicated to ensuring that general-purpose artificial intelligence benefits all of humanity. AI is an extremely powerful tool that must be created with safety and human needs at its core. OpenAI is dedicated to putting that alignment of interests first — ahead of profit. To achieve our mission, we must encompass and value the many different perspectives, voices, and experiences that form the full spectrum of humanity. Our investment in diversity, equity, and inclusion is ongoing, executed through a wide range of initiatives, and championed and supported by leadership. At OpenAI, we believe artificial intelligence has the potential to help people solve immense global challenges, and we want the upside of AI to be widely shared. Join us in shaping the future of technology.
- Website
-
https://openai.com/
External link for OpenAI
- Industry
- Research Services
- Company size
- 201-500 employees
- Headquarters
- San Francisco, CA
- Type
- Partnership
- Specialties
- artificial intelligence and machine learning
Locations
-
Primary
Get directions
San Francisco, CA 94110, US
Employees at OpenAI
Updates
-
Your Year with ChatGPT! Now rolling out to everyone in the US, UK, Canada, New Zealand, and Australia who have reference saved memory and reference chat history turned on. Just make sure your app is updated. If you're in one of the countries above, check back throughout the day to see if "Your Year with ChatGPT" has rolled out to you. You can also try adding the "Your Year with ChatGPT" app by tapping the + sign and asking Chat, "show me my year with ChatGPT."
-
-
When we think about this year at OpenAI, we think about what changed across the enterprise landscape. We watched 1 million businesses move from experimenting with AI to standardizing on it. OpenAI became core infrastructure as companies consolidated on a single AI platform to build, operate, and compete at scale. Customers rolled out ChatGPT across their organizations to help with writing, coding, research, data analysis, design, and the hundreds of small tasks that add up every day. They built agents to automate workflows that used to take hours. Developers used Codex to move faster and tackle bigger problems. And companies built entirely new products on our API, working across voice, video, images, and other modalities. When we asked customers what AI enabled them to do, 75% told us they were completing tasks they had never been able to do before. We’re grateful to the companies leading this transition and setting the pace for what enterprise AI becomes next. https://lnkd.in/gmaaMyWc
-
OpenAI reposted this
At this week’s OpenAI Forum, Kevin Weil (OpenAI for Science) sat down with Brian Spears (Lawrence Livermore National Laboratory; DOE’s Genesis Mission) to talk about how frontier reasoning models are moving beyond “chat” and becoming practical tools for working scientists. Watch the recording here https://lnkd.in/eNDuskWJ Spears shared his “o1 moment”: starting from a one‑megabar shock hitting stainless steel, the model progressed from explaining the physics to writing the core conservation equations used in real simulations—and even referred to a complex radiation‑hydrodynamics system at his lab, a tool that typically takes years to master. The breakthrough was the model’s ability to reason over complex subject matter in the same language scientists use. He also described what’s possible now as models combine reasoning with execution: iterative analysis that converges on better starting points for expensive high‑performance computing runs, plus agent-style workflows that set up simulations, plan large parameter sweeps, and automate post‑processing—while keeping humans in the loop for verification and high‑consequence judgment. Watch the recording to hear the full conversation, including how this connects to fusion milestones and the Genesis Mission’s platform approach to accelerating discovery. Read the full breakdown here https://lnkd.in/eTRSizMP
-
Introducing ChatGPT Images, powered by our flagship new image generation model. - Stronger instruction following - Precise editing - Detail preservation - 4x faster than before We’re also introducing a new Images surface within ChatGPT. Just tap “Images” in the sidebar and have fun. Rolling out today in ChatGPT for all users, and in the API as GPT-Image-1.5. https://lnkd.in/gtsE-MiH
-
Accelerating scientific progress is one of the most impactful ways AI can benefit society. Models can already help researchers reason through hard problems — but doing this well means testing models on tougher evaluations and in real scientific workflows grounded in experiments. We’re releasing a new eval to measure expert-level scientific reasoning: FrontierScience. This benchmark measures PhD-level scientific reasoning across physics, chemistry, and biology. It contains hard, expert-written questions (both olympiad-style problems and longer research-style tasks) designed to reveal where models succeed and where they fall short. https://lnkd.in/gP3qDKbg
-
We shipped Sora for Android in 28 days using the exact same assistant available to any team: Codex. OpenAI engineers Patrick Hum and Rj Marsan unpack how they used Codex for planning, running parallel coding workflows, generating potential failure points, testing, maintenance, and more to nimbly deliver a production app from scratch to #1 in the Play Store. Read more in our latest Engineering Blog: https://lnkd.in/d6sX9mN9
-
GPT-5.2 is now rolling out to everyone. This is our most capable model series yet for professional knowledge work. We designed GPT-5.2 to unlock even more economic value for people. It’s better at creating spreadsheets, building presentations, writing code, perceiving images, understanding long contexts, using tools, and handling complex, multi-step projects. Overall, GPT-5.2 brings significant improvements in general intelligence, long-context understanding, agentic tool-calling, and vision—making it better at executing complex, real-world tasks end-to-end than any previous model. In ChatGPT, GPT-5.2 Instant, Thinking, and Pro will roll out today to all users. In the OpenAI API, they are available now to all developers. https://lnkd.in/g_bYbUB9
-
As our models grow more capable in cybersecurity, we’re taking a deliberate, long-term approach to safety and preparedness. We’re strengthening safeguards and working closely with global experts as we prepare for upcoming models to reach ‘High’ capability under our Preparedness Framework. Cybersecurity is inherently dual-use, so we design for defenders first. We’re investing in layered protections—from training and access controls to monitoring, detection, and enforcement. At the same time, we’re expanding ecosystem efforts like trusted access programs, Aardvark, expert advisory councils, and sharing threat modeling with other frontier labs. These partnerships ensure safeguards reflect real-world needs and that insights move quickly across the security community. We see this work as a sustained, long-term investment in giving defenders an advantage and continually strengthening the security posture of the critical infrastructure across the broader ecosystem. https://lnkd.in/gSF-7s7k