Several news outlets reported on a tense internal meeting at OpenAI this week, where CEO Sam Altman tried to calm employees after backlash over the company’s recent agreement with the United States Department of Defense. The meeting came after criticism from staff and users who questioned the company’s growing role in military systems.
Reports from The Wall Street Journal, CNBC, and The New York Times describe a conversation where Altman acknowledged the strain inside the company. According to these accounts, he described the situation as “painful” and admitted the company looked divided during the public debate over the deal.
The contract in question allows the Defense Department to use OpenAI’s technology inside classified systems. The agreement falls under the Pentagon’s Chief Digital and Artificial Intelligence Office (CDAO), which oversees projects that apply artificial intelligence to national defense. Through this program, OpenAI could receive up to $200 million in project funding.
Internal Strife Follows $200M Pentagon Deal and NATO Ambitions
Employees have raised concerns about the ethical limits of such partnerships. Many workers in the AI sector support strict rules on military use. The reaction inside OpenAI reflects a wider debate across the technology industry about how AI should interact with defense and intelligence operations.
Altman’s comments during the meeting suggest the company plans to deepen its role in that space rather than step back. One notable passage reported by The Wall Street Journal said OpenAI is exploring another agreement that would allow its systems to operate across classified networks used by the North Atlantic Treaty Organization (NATO).
Access to NATO’s classified networks carries weight in the technology world. Companies must pass strict security checks before their products can operate inside those systems. Approval signals that a product meets the alliance’s security standards and can handle sensitive information shared by member states.
&imwidth=800&imheight=600&format=webp&quality=medium)
Just last month, Apple announced that some of its devices had received similar approval. The company said the iPhone and iPad could now be used for classified NATO communications. Apple described the approval as the first time consumer devices gained that level of trust from the alliance.
The Strategic Pivot of OpenAI into Classified Military Networks
OpenAI’s situation differs from Apple’s approval. Apple secured permission for its hardware to be used within the network. OpenAI’s deal would allow government agencies to run the company’s AI systems inside classified environments. According to reporting from CNBC, the Defense Department would be able to deploy those systems while keeping operational control.
This distinction matters. It means OpenAI provides the underlying technology while government officials decide how to apply it in defense operations. Altman reportedly stressed this point during the meeting. He told employees that OpenAI would not make operational military decisions.
The push toward defense partnerships also reflects a larger shift in global spending. NATO leaders signaled last year that member countries plan to increase defense budgets over the next decade. That increase has drawn strong interest from technology firms that see artificial intelligence as a core tool in modern security systems.
Some investors describe the trend as the start of an “AI gold rush” in defense. Venture capital firms and startups now compete for government contracts tied to data analysis, autonomous systems, and cyber defense. Governments want tools that process large volumes of information and support faster decision-making.
OpenAI entered this space when it launched its “OpenAI for Government” program in 2025. The initiative aims to provide AI tools to public sector agencies while keeping the company’s safety policies in place. The Pentagon contract marked one of the largest early deals under that program.
However, the debate within OpenAI illustrates the dilemma between business growth and ethical boundaries. For instance, many experts in the field are concerned about the potential impact of the development of the technology on the strategies used in the armed forces. Others argue that the government should partner with companies they trust and not companies whose reliability is in question.
Managing Internal Unity Amidst External Pressure
For the moment, Altman is focused on keeping both parties in view. His message to the employees focused on unity and transparency, as well as the company’s commitment to partnering with the defense institutions.
Whether the balance will be sustained in the future is yet to be seen. With the technology becoming a core aspect in global security, companies like OpenAI will be under scrutiny from their workers, users, and the government.
