The Pentagon announced Friday that it has reached agreements with seven major technology companies to integrate AI systems into classified military networks. Google, Microsoft, Amazon Web Services, Nvidia, OpenAI, Reflection and SpaceX will provide their resources to enhance military decision-making in complex operational environments.
Anthropic 's absence from the list is notable, following its public and legal dispute with the Trump administration over the ethics and safety of using artificial intelligence in warfare. Anthropic had sought assurances that the military technology would not be used in fully autonomous weapons and the surveillance of American citizens , something the Department of Defense refused to guarantee.
See also: Iran names 18 US tech companies as military targets
The Department of Defense has been rapidly accelerating its use of artificial intelligence in recent years. The technology can help the military reduce the time it takes to locate and engage targets on the battlefield, while also helping to organize weapons maintenance and supply lines. According to a report in March by the Brennan Center for Justice , artificial intelligence can turn tasks that used to take months into processes of a few days.
Concerns about the Pentagon's ethical use of AI
Artificial intelligence has already raised concerns about the potential invasion of Americans’ privacy. There are also fears that it could allow machines to select targets on the battlefield. One of the companies working with the Pentagon said its agreement requires human oversight in certain situations. Concerns about the military use of artificial intelligence emerged during Israel against militants in Gaza and Lebanon, with American tech giants subtly empowering Israel to track targets.
Helen Toner , interim executive director of Georgetown University 's Center for Security and Emerging Technology and a former member of the board of directors of OpenAI , points out that the new Pentagon contracts come at a time of concern about the potential overreliance on technology on the battlefield .
“ Much of modern warfare relies on people sitting in command centers behind screens, making complex decisions about confusing, rapidly evolving situations ,” Toner explains .
See also: Major record labels close to big AI licensing deals
Technical capabilities and applications of AI in the military
AI systems can be useful for summarizing information or examining surveillance feeds and trying to identify potential targets. However, questions about appropriate levels of human involvement, risk, and training are still being worked out.
OpenAI had announced a deal with the Pentagon in March to effectively replace Anthropic in classified environments. The company confirmed in a statement Friday that it was the same deal it announced in early March. “ As we said when we first announced our agreement several months ago, we believe that the people who defend the United States should have the best tools in the world ,” the company said.
A company's agreement with the Pentagon included language that said there must be human oversight of any missions in which AI systems act autonomously or semi-autonomously. The language also said that AI tools must be used in ways that are consistent with constitutional rights and civil liberties.
The Pentagon's view on the agreement with AI/tech companies
Emil Michael, the Pentagon's chief technology officer, said it would be irresponsible to rely on just one company (after the problems with Anthropic).
“And when we found out that one partner didn’t really want to work with us the way we wanted to, we went out and made sure we had a lot of different providers,” Michael said.
See also: Tech companies don't care if students use AI agents to copy
Some of the companies, including Amazon and Microsoft, have long worked with the military in classified environments. Others, like Nvidia and Reflection, are new to this kind of work. Both companies build open-source AI models, which Michael sees as an “American alternative” to China’s rapid development of AI systems.
The Pentagon said on Friday that military personnel are already using artificial intelligence capabilities through its official platform, GenAI.mil.
“Warriors, civilians and contractors are putting these capabilities into practice right now, reducing many tasks from months to days,” the Pentagon said, adding that the military’s growing AI capabilities will “give warfighters the tools they need to act with confidence and protect the nation from any threat.”
AI can be used to better predict when a helicopter needs maintenance or to figure out how to efficiently move large amounts of troops and equipment, he said. It can also help determine whether vehicles in a drone's surveillance streams are civilian or military.
But people shouldn't depend too much on it.
"There's a phenomenon called automation bias, where people may be prone to assuming that machines perform better than they actually do," Toner said.
