Over the past two years, we’ve learned a lot about how language models can be used and abused—insights we couldn’t have gained without the experience of real-world deployment. In June 2020, we began giving access to developers and researchers to the OpenAI API⁠, an interface for accessing and building applications on top of new AI models developed by OpenAI. Deploying GPT‑3, Codex, and other models in a way that reduces risks of harm has posed various technical and policy challenges.

Large language models are now capable of performing a very wide range of tasks⁠(opens in a new window), often out of the box. Their risk profiles, potential applications, and wider effects on society remain⁠(opens in a new window) poorly⁠(opens in a new window) understood⁠(opens in a new window). As a result, our deployment approach emphasizes continuous iteration, and makes use of the following strategies aimed at maximizing the benefits of deployment while reducing associated risks:

There is no silver bullet for responsible deployment, so we try to learn about and address our models’ limitations, and potential avenues for misuse, at every stage of development and deployment. This approach allows us to learn as much as we can about safety and policy issues at small scale and incorporate those insights prior to launching larger-scale deployments.

Since each stage of intervention has limitations, a holistic approach is necessary.