OpenAI’s Sam Altman announces Pentagon deal with ‘technical safeguards’
Summary
OpenAI announced a deal allowing the Department of Defense to use its AI models on classified networks, following a dispute where rival Anthropic refused to agree to unrestricted military use without safeguards against mass domestic surveillance and fully autonomous weapons. Sam Altman stated that OpenAI's agreement includes technical protections addressing these same concerns, with OpenAI building a 'safety stack' (a set of security and control measures) and deploying engineers to ensure the models behave correctly.
Solution / Mitigation
According to Altman, OpenAI will 'build technical safeguards to ensure our models behave as they should' and will 'deploy engineers with the Pentagon to help with our models and to ensure their safety.' Additionally, the government will allow OpenAI to build its own 'safety stack to prevent misuse' and 'if the model refuses to do a task, then the government would not force OpenAI to make it do that task.'
Classification
Affected Vendors
Related Issues
Original source: https://techcrunch.com/2026/02/28/openais-sam-altman-announces-pentagon-deal-with-technical-safeguards/
First tracked: February 28, 2026 at 03:00 PM
Classified by LLM (prompt v3) · confidence: 92%