Browse press releases
The newest releases come first, each with its full text here and a link to its original. Try the filters to narrow them down.
- OpenAIFilling crucial language learning gapsCompanies: OpenAIGPT-4 deepens the conversation on Duolingo.
- OpenAIPowering virtual education for the classroomCompanies: OpenAIKhan Academy explores the potential for GPT-4 in a limited pilot program.
- OpenAIForecasting potential misuses of language models for disinformation campaigns and how to reduce riskCompanies: OpenAIOpenAI researchers collaborated with Georgetown University’s Center for Security and Emerging Technology and the Stanford Internet Observatory to investigate how large language models might be misused for disinformation purposes. The collaboration included an October 2021 workshop bringing together 30 disinformation researchers, machine learning experts, and policy analysts, and culminated in a co-authored report building on more than a year of research. This report outlines the threats that language models pose to the information environment if used to augment disinformation campaigns and introduces a framework for analyzing potential mitigations. Read the full report here.
- OpenAIThe power of continuous learningCompanies: OpenAILilian Weng works on Applied AI Research at OpenAI.
- OpenAIScaling laws for reward model overoptimizationCompanies: OpenAIAbstract In reinforcement learning from human feedback, it is common to optimize against a reward model trained to predict human preferences. Because the reward model is an imperfect proxy, optimizing its value too much can hinder ground truth performance, in accordance with Goodhart's law. This effect has been frequently observed, but not carefully measured due to the expense of collecting human preference data. In
- OpenAILearning to play Minecraft with Video PreTrainingCompanies: OpenAIWe trained a neural network to play Minecraft by Video PreTraining (VPT) on a massive unlabeled video dataset of human Minecraft play, while using only a small amount of labeled contractor data. With fine-tuning, our model can learn to craft diamond tools, a task that usually takes proficient humans over 20 minutes (24,000 actions). Our model uses the native human interface of keypresses and mouse movements, making it quite general, and represents a step towards general computer-using agents.
- OpenAISolving (some) formal math olympiad problemsCompanies: OpenAIWe built a neural theorem prover for Lean that learned to solve a variety of challenging high-school olympiad problems, including problems from the AMC12 and AIME competitions, as well as two problems adapted from the IMO.
- OpenAISolving math word problemsCompanies: OpenAIWe’ve trained 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.
- OpenAILearning to summarize with human feedbackCompanies: OpenAIWe’ve applied reinforcement learning from human feedback to train language models that are better at summarization.
- OpenAIProcgen and MineRL CompetitionsCompanies: OpenAIWe’re excited to announce that OpenAI is co-organizing two NeurIPS 2020 competitions with AIcrowd, Carnegie Mellon University, and DeepMind, using Procgen Benchmark and MineRL.
- OpenAIOpenAI standardizes on PyTorchCompanies: OpenAIWe are standardizing OpenAI’s deep learning framework on PyTorch.
- OpenAIDota 2 with large scale deep reinforcement learningCompanies: OpenAIAbstract On April 13th, 2019, OpenAI Five became the first AI system to defeat the world champions at an esports game. The game of Dota 2 presents novel challenges for AI systems such as long time horizons, imperfect information, and complex, continuous state-action spaces, all challenges which will become increasingly central to more capable AI systems. OpenAI Five leveraged existing reinforcement learning technique