Drawing from OpenAI’s established safety frameworks, this document highlights our multi-layered approach, including model and product mitigations we’ve implemented to protect against prompt engineering and jailbreaks, protect privacy and security, as w…
Deliberative alignment: reasoning enables safer language models
Introducing our new alignment strategy for o1 models, which are directly taught safety specifications and how to reason over them.
Sora is OpenAI’s video generation model, designed to take text, image, and video inputs and generate a new video as an output. Sora builds on learnings from DALL-E and GPT models, and is designed to give people expanded tools for storytelling and creat…
An update on our safety & security practices
We banned accounts linked to a covert Iranian influence operation using ChatGPT to generate website and social media content focused on multiple topics, including the U.S. presidential campaign. We have seen no indication that this content reached a me…
This report outlines the safety work carried out prior to releasing GPT-4o including external red teaming, frontier risk evaluations according to our Preparedness Framework, and an overview of the mitigations we built in to address key risk areas.
CriticGPT, a model based on GPT-4, writes critiques of ChatGPT responses to help human trainers spot mistakes during RLHF
Exploring the technology behind our text-to-speech model.
Artificial general intelligence has the potential to benefit nearly every aspect of our lives—so it must be developed and deployed responsibly.