Alignment, interpretability, red-teaming, bias and privacy: the research on what these systems do when they misbehave.
OpenAI is fighting a court order at the demands of The New York Times and plaintiffs, which involves retention of consumer ChatGPT and API user data indefinitely. Learn how we’re working to uphold user privacy, address legal requirements, and stay true to our data protection commitments.
HealthBench is a new evaluation benchmark for AI in healthcare which evaluates models in realistic scenarios. Built with input from 250+ physicians, it aims to provide a shared standard for model performance and safety in health.
Data residency builds on OpenAI’s enterprise-grade data privacy, security, and compliance programs supporting customers worldwide.
An update from the OpenAI board on transitioning its for-profit entity to a Public Benefit Corporation, reinforcing its mission-driven structure under nonprofit oversight while enabling greater impact and long-term alignment with the public good.
This report outlines the safety work carried out prior to releasing deep research 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.
Data residency builds on OpenAI’s enterprise-grade data privacy, security, and compliance programs supporting customers worldwide.
This report outlines the safety work carried out for the OpenAI o3-mini model, including safety evaluations, external red teaming, and Preparedness Framework evaluations.
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 well as details our external red teaming efforts, safety evaluations, and ongoing work to further refine these safeguards.
Trading Inference-Time Compute for Adversarial Robustness
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.
This report outlines the safety work carried out prior to releasing OpenAI o1 and o1-mini, including external red teaming and frontier risk evaluations according to our Preparedness Framework.
We've analyzed how ChatGPT responds to users based on their name, using AI research assistants to protect privacy.
Genmab embraces ChatGPT Enterprise, supported by OpenAI’s commitment to security and privacy
An update on our safety & security practices
Zico Kolter Joins OpenAI’s Board of Directors We’re strengthening our governance with expertise in AI safety and alignment.
We've developed and applied a new method leveraging Rule-Based Rewards (RBRs) that aligns models to behave safely without extensive human data collection.
OpenAI and Los Alamos National Laboratory are working to develop safety evaluations to assess and measure biological capabilities and risks associated with frontier models.
Consistency models are a nascent family of generative models that can sample high quality data in one step without the need for adversarial training.
Nakasone brings cybersecurity experience to growing Board of Directors; will join the Board’s Safety and Security Committee