OpenAI Blog

Scaling laws for neural language models

Dota 2 with large scale deep reinforcement learning

Deep double descent

We show that the double descent phenomenon occurs in CNNs, ResNets, and transformers: performance first improves, then…

Procgen Benchmark

We’re releasing Procgen Benchmark, 16 simple-to-use procedurally-generated environments which provide a direct measure…

Safety Gym

We’re releasing Safety Gym, a suite of environments and tools for measuring progress towards reinforcement learning…

Benchmarking safe exploration in deep reinforcement learning

GPT-2: 1.5B release

As the final model release of GPT-2’s staged release, we’re releasing the largest version (1.5B parameters) of GPT-2…

Solving Rubik’s Cube with a robot hand

We’ve trained a pair of neural networks to solve the Rubik’s Cube with a human-like robot hand. The neural networks are…

OpenAI Scholars 2020: Applications open

We are now accepting applications for our third class of OpenAI Scholars.

Fine-tuning GPT-2 from human preferences

We’ve fine-tuned the 774M parameter GPT-2 language model using human feedback for various tasks, successfully matching…

Emergent tool use from multi-agent interaction

We’ve observed agents discovering progressively more complex tool use while playing a simple game of hide-and-seek.…

Testing robustness against unforeseen adversaries

We’ve developed a method to assess whether a neural network classifier can reliably defend against adversarial attacks…

GPT-2: 6-month follow-up

We’re releasing the 774 million parameter GPT-2 language model after the release of our small 124M model in February,…

Learning Day

At OpenAI, each Thursday is Learning Day: a day where employees have the option to self-study technical skills that…

Microsoft invests in and partners with OpenAI to support us building beneficial AGI

Microsoft is investing $1 billion in OpenAI to support us building artificial general intelligence (AGI) with widely…

Why responsible AI development needs cooperation on safety

We’ve written a policy research paper identifying four strategies that can be used today to improve the likelihood of…

OpenAI Robotics Symposium 2019

We hosted the first OpenAI Robotics Symposium on April 27, 2019.

OpenAI Scholars 2019: Final projects

Our second class of OpenAI Scholars has concluded, with all eight scholars producing an exciting final project…

OpenAI Fellows Fall 2018: Final projects

Our second class of OpenAI Fellows has wrapped up, with each Fellow going from a machine learning beginner to core…

Transfer of adversarial robustness between perturbation types

MuseNet

We’ve created MuseNet, a deep neural network that can generate 4-minute musical compositions with 10 different…

Generative modeling with sparse transformers

We’ve developed the Sparse Transformer, a deep neural network which sets new records at predicting what comes next in a…

OpenAI Five defeats Dota 2 world champions

OpenAI Five is the first AI to beat the world champions in an esports game, having won two back-to-back games versus…

OpenAI Five Finals

We’ll be holding our final live event for OpenAI Five at 11:30am PT on April 13.

Implicit generation and generalization methods for energy-based models

We’ve made progress towards stable and scalable training of energy-based models (EBMs) resulting in better sample…

OpenAI Scholars 2019: Meet our Scholars

Our class of eight scholars (out of 550 applicants) brings together collective expertise in literature, philosophy,…

OpenAI LP

We’ve created OpenAI LP, a new “capped-profit” company that allows us to rapidly increase our investments in compute…

Introducing Activation Atlases

We’ve created activation atlases (in collaboration with Google researchers), a new technique for visualizing what…

Neural MMO: A massively multiagent game environment

We’re releasing a Neural MMO, a massively multiagent game environment for reinforcement learning agents. Our platform…

Spinning Up in Deep RL: Workshop review

On February 2, we held our first Spinning Up Workshop as part of our new education initiative at OpenAI.

AI safety needs social scientists

We’ve written a paper arguing that long-term AI safety research needs social scientists to ensure AI alignment…

Better language models and their implications

We’ve trained a large-scale unsupervised language model which generates coherent paragraphs of text, achieves…

Computational limitations in robust classification and win-win results

OpenAI Fellows Summer 2018: Final projects

Our first cohort of OpenAI Fellows has concluded, with each Fellow going from a machine learning beginner to core…

How AI training scales

We’ve discovered that the gradient noise scale, a simple statistical metric, predicts the parallelizability of neural…

Quantifying generalization in reinforcement learning

We’re releasing CoinRun, a training environment which provides a metric for an agent’s ability to transfer its…