Over the past six months the number of US Department of Defense employees using commercial AI tools has grown by 1775% — from 80,000 to 1.5 million. This was stated by the Pentagon's chief technology officer Emil Michael at a Hudson Institute event, Fortune writes.

Of the department's total staff of 3.5 million people, about 43% use artificial intelligence.

One of the cases of integrating the technology is the preparation of mandatory reports for Congress. Michael claims that AI can assemble a draft of such a document in five hours instead of 200 hours of work by an entire team of employees.

In April the deputy assistant secretary of defense for science and technology foundations, Jacob Glassman, said that he had instructed an understaffed team to use GenAI.mil to prepare a report. According to him, a week later the team delivered a finished document and called it the best in the last five years.

Later the Pentagon announced partnerships with SpaceX, OpenAI, Google, Nvidia, Reflection, Microsoft, Amazon Web Services and Oracle for the operational use of AI tools.

The use of artificial intelligence in US government structures is developing beyond the Department of Defense as well, journalists write. The authorities introduced initiatives for the use of AI in agencies as early as the 1960s: the first experiments with the technology tried to solve logistical problems.

The 2020 law on the use of AI in government, passed during Donald Trump's first presidency, gave an impetus to practical application. 

Nevertheless, in March 2026 the US Government Accountability Office warned that expanding the use of neural networks in the public sector may increase the risk of generating false information and of access to data outside its intended purpose. 

Recall that in May the media reported that the Pentagon had created a group to deploy hacker AI models.

Источник: ForkLog