AI advances in science, healthcare, and education
OpenAI is offering U.S. servicemembers and veterans within 12 months of retirement or separation a free year of ChatGPT Plus to support their transition to civilian life. The tools can help with resumes, interviews, education, and planning for what鈥檚 next.
OpenAI and Sur Energy are exploring Argentina鈥檚 first Stargate project鈥攁n AI and clean energy collaboration that could make Argentina a Latin American leader in artificial intelligence, sustainable infrastructure, and digital innovation.
Applications are now open for OpenAI鈥檚 People-First AI Fund, a $50M initiative supporting U.S. nonprofits advancing education, community innovation, and economic opportunity. Apply by October 8, 2025, for unrestricted grants that help communities shape AI for the public good.
As part of our DALL路E 2 research preview, more than 3,000 artists from more than 118 countries have incorporated DALL路E into their creative workflows. The artists in our early access group have helped us discover new uses for DALL路E and have served as key voices as we鈥檝e made decisions about DALL路E鈥檚聽features.
We built a neural theorem prover for聽Lean聽that learned to solve a variety of challenging high-school olympiad problems, including problems from the聽AMC12聽and聽AIME聽competitions, as well as two problems adapted from the聽IMO.
We鈥檝e trained聽a system that solves grade school math problems with nearly twice the accuracy of a fine-tuned GPT-3 model. It solves about 90% as many problems as real kids: a small sample of 9-12 year olds scored 60% on a test from our dataset, while our system scored 55% on those same problems.
OpenAI Five is the first AI to beat the world champions in an esports game, having won two back-to-back games versus the world champion Dota 2 team,聽OG, at聽Finals聽this weekend. Both OpenAI Five and DeepMind鈥檚 AlphaStar had previously beaten good pros privately but lost their live pro matches, making this also the first time an AI has beaten esports pros on聽livestream.
We鈥檝e developed a simple meta-learning algorithm called Reptile which works by repeatedly sampling a task, performing stochastic gradient descent on it, and updating the initial parameters towards the final parameters learned on that task. Reptile is the application of the Shortest Descent algorithm to the meta-learning setting, and is mathematically similar to first-order MAML (which is a version of the well-known MAML algorithm) that only needs black-box access to an optimizer such as SGD or Adam, with similar computational efficiency and performance.
We鈥檙e releasing eight simulated robotics environments and a Baselines implementation of Hindsight Experience Replay, all developed for our research over the past year. We鈥檝e used these environments to train models which work on physical robots. We鈥檙e also releasing a set of requests for robotics research.
Deep learning is an empirical science, and the quality of a group鈥檚 infrastructure is a multiplier on progress. Fortunately, today鈥檚 open-source ecosystem makes it possible for anyone to build great deep learning infrastructure.