The Pentagon Turns to AI to Draft Congressional Reports

The US Department of Defense is using generative AI tools to help draft congressionally required reports, with Pentagon Chief Technology Officer Emil Michael pointing to the process as an example of how the agency has adopted AI through its GenAI.mil platform.

Michael discussed the approach during a June 12 event hosted by the Hudson Institute in Washington, DC. He said the department has used AI to reduce the amount of staff time needed to prepare reports for Congress, a recurring obligation that covers a wide range of national security topics. “I have to report to Congress every year on this thing,” Michael said. “Let me load all the papers onto it and have it draft me a congressional report that would otherwise take 200 hours of staffing time and do it in five hours.”

The Pentagon has made generative AI tools available to personnel across all six military branches through GenAI.mil since December 2025, beginning with Google Cloud’s Gemini for Government. Michael said usage has grown from 80,000 Department of Defense personnel in December 2025 to 1.5 million in June 2026. The department’s total workforce is about 3.5 million.

Another Pentagon official has described a similar use case. Jacob Glassman, deputy assistant secretary of defense for science and technology foundations, said at the Box Federal Summit on April 23 that he told a short-staffed team working on a congressionally mandated report to “use GenAI.mil, do the best you can.”

According to DefenseScoop, Glassman said the team returned a week later and called the AI-generated document “the best report we’ve written in the past five years.” The specific report was not identified.

The use of AI comes as the Pentagon continues to face a large reporting burden. The US Government Accountability Office found that the number of mandated Defense Department reports rose from just over 500 in 2000 to more than 1,400 by 2020. A 2023 GAO report also found that identifying reporting requirements and assigning them to the proper teams could take three to six months.

The shift raises questions about review and oversight. The article says it is unclear what processes the Pentagon uses to verify the accuracy of AI-generated reports before they are sent to Congress. Those reports are part of congressional oversight of the military, including how taxpayer money is used.

The broader risks of AI-generated writing have already surfaced in other industries. The article cites KPMG’s decision to pull a report on AI in business after GPTZero identified case studies containing AI-generated errors and false claims, according to the Financial Times.

Generative AI is also being used elsewhere inside the military. A Small Wars Journal article said service members have used AI tools to draft personnel evaluations, commendation medal citations and counseling statements.

The Defense Department is expanding its AI access beyond unclassified tools. On May 1, it announced agreements with eight frontier AI companies to deploy AI systems on classified networks for lawful operational use. The companies listed were SpaceX, OpenAI, Google, Nvidia, Reflection AI, Microsoft, Amazon Web Services and Oracle.

The government has not disclosed how much it is paying under those contracts. The article also notes that Anthropic was not included after being blacklisted by the Trump administration over its restrictions on Claude’s use for autonomous warfare and mass surveillance.

This analysis is based on reporting from ars technica.

Image courtesy of The New York Times.

This article was generated with AI assistance and reviewed for accuracy and quality.