OpenAI National Security Principles and Government Partnerships

OpenAI has published its National Security Principles to establish a transparent framework for how its frontier AI systems are deployed in government and national security contexts. This framework is designed to ensure that AI is used to protect people, defend critical infrastructure, and respond to emerging threats while reinforcing democratic accountability and the rule of law.

National Security Principles Framework

OpenAI developed these principles through a cross-company effort involving research, safety, policy, and government partnership teams. The process was facilitated by national security expert David Kris to provide independent judgment and included internal listening sessions to ensure a comprehensive approach to the use of technology in advancing national security and law enforcement.

Defensive AI Applications and Global Partnerships

OpenAI is expanding its work with the U.S. government and allied partners, focusing specifically on critical defensive areas such as cyber and biosecurity.

Cyber Defense

Through the "Daybreak" cyber defense program, OpenAI has established "Trusted Access for Cyber" partnerships with the following entities:

• Australia
• Canada
• Japan
• Republic of Korea
• France
• Germany
• Poland
• The Netherlands
• EU institutions, including ENISA

Additionally, the company maintains a partnership with the UK government focusing on cyber, testing, and evaluation.

Biosecurity

OpenAI is applying a similar partnership approach to biosecurity. The company has provided expanded trusted access to the GPT-Rosalind model for select U.S. government and allied partners supporting biodefense missions and public health.

Restrictions on High-Risk AI Use

OpenAI has established specific contractual restrictions to prevent the high-risk use of its technology. These restrictions apply to all current and future national security and law enforcement partnerships, including existing work with the Department of War.

Key prohibitions include:

• No use of OpenAI technology for mass domestic surveillance.
• No use of OpenAI technology to direct autonomous weapons systems.
• No use technology for high-stakes automated decisions.

Democratic Process and Legislative Support

OpenAI states that the company does not believe AI companies should make decisions on the most consequential questions regarding AI in government alone. Instead, the company supports legislative efforts to establish safeguards around high-risk military uses of AI, specifically in the contexts of autonomous weapons systems, domestic surveillance, and other uses that directly affect U.S. national security.