Observed Signal · Feb 28, 2026 · Partnership/Contract · Source: TechCrunch · Impact: 4/5 · Sentiment: Neutral
OpenAI Partners with Pentagon, Ensures AI Safety Measures
OpenAI CEO Sam Altman announced that OpenAI has reached an agreement allowing the U.S. Department of Defense to use its AI models on the department’s classified network. Altman said the agreement includes technical safeguards and principles prohibiting domestic mass surveillance and preserving human responsibility for the use of force, including for autonomous weapon systems. OpenAI plans to build a safety stack, deploy engineers with the Pentagon, and asserts the government will not force the company to make models perform refused tasks. The deal follows a failed negotiation between the Pentagon and Anthropic, public employee letters from OpenAI and Google staff, and public criticism and supply-chain restriction actions from President Donald Trump and Secretary of Defense Pete Hegseth toward Anthropic.
A major AI vendor (OpenAI) striking a contract with the U.S. Department of Defense and embedding technical safeguards sets a precedent for government access, safety requirements, and industry-government relations; it follows high-profile failed negotiations with another leading AI company and includes potential supply-chain and policy implications.
Track OpenAI Signals & Market Shifts in Real-Time
Polaris7 autonomous intelligence agents track regulatory filings, primary sources, executive changes, and deal flow 24/7. Create your free Explorer workspace to monitor these entities.
Key Takeaways & Evidence Grounding
- OpenAI has reached an agreement allowing the U.S. Department of Defense to use its AI models on the department’s classified network.
- OpenAI says the agreement includes prohibitions on domestic mass surveillance and requirements for human responsibility for the use of force, including for autonomous weapon systems.
- OpenAI will build technical safeguards, a safety stack, and deploy engineers to work with the Pentagon to ensure model safety.
- Anthropic failed to reach an agreement with the Pentagon and faced criticism from President Trump and a supply-chain restriction announcement from Secretary of Defense Pete Hegseth.
- More than 60 OpenAI employees and about 300 Google employees signed an open letter supporting Anthropic’s position.
Connected Companies & Entities
3 Entities mappedRelated Market Signals & Shifts
Recent verified developments and strategic activity across this market segment.
OpenAI's Pentagon Deal: Safeguards Amid Controversy
OpenAI announced details of a rapid agreement with the U.S. Department of Defense to deploy its AI models in classified environments. CEO Sam Altman acknowledged the deal was rushed and drew backlash. OpenAI published a blog post saying its models cannot be used for mass domestic surveillance, autonomous weapon systems, or high-stakes automated decisions (e.g., social credit). The company described a multi-layered approach to enforce those red lines — retaining discretion over its safety stack, deploying via cloud APIs, keeping cleared OpenAI personnel in the loop, and using contractual protections alongside U.S. law. The announcement followed failed negotiations between Anthropic and the Pentagon and an order from President Donald Trump to phase out Anthropic technology; critics raised concerns that references to Executive Order 12333 in contract language could enable domestic surveillance.
OpenAI Partners with DoW: AI Safeguards in Place
OpenAI announced an agreement with the U.S. Department of War (DoW) to deploy advanced AI systems in classified environments under a cloud-only architecture and explicit contractual safeguards. The agreement includes language that the AI shall not be intentionally used for domestic surveillance of U.S. persons or nationals, and that Department of War intelligence agencies such as the NSA would require a separate agreement to access services. OpenAI says the deployment will keep its safety stack and cleared OpenAI personnel in the loop, will not be deployed to edge devices, and references existing legal authorities (Fourth Amendment, National Security Act of 1947, FISA 1978) and DoD Directive 3000.09. OpenAI lists three red lines (no mass domestic surveillance, no directing autonomous weapons, no high-stakes automated decisions), plans to join a DoW-convened working group with other frontier labs and cloud providers, and states it requested the same terms be offered to all AI labs.
Altman: OpenAI's AI Use in Military Is Government's Call
OpenAI CEO Sam Altman told employees that the company does not control operational decisions about how its AI is used by the U.S. Department of Defense, saying those decisions rest with Secretary Pete Hegseth. The all-hands came days after OpenAI announced an expanded DoD arrangement that will allow its models to be deployed on the department's classified networks; OpenAI previously received a $200 million DoD contract for nonclassified use. The announcement drew internal and public criticism, and came amid reports that Anthropic's models were used in recent military operations. Elon Musk's xAI has also agreed to deploy models for classified DoD use. Altman acknowledged the timing of OpenAI's announcement looked "opportunistic and sloppy."
Track Real-Time Market Signals & Shifts
Set up custom watchlists to receive automated, evidence-grounded executive digests whenever material signals or shifts occur across your tracked landscape.