**Sen. Elizabeth Warren Demands Transparency on AI Contracts with Department of Defense**
A recent letter from Senator Elizabeth Warren (D-Mass.) has spotlighted concerns over the Defense Department’s expanding use of artificial intelligence (AI) in military operations. In her correspondence with the DoD and seven major technology companies, Warren is pushing for disclosure of the terms governing AI deployment on the department’s classified networks.
The Defense Department entered into agreements in May with SpaceX, OpenAI, Google, NVIDIA, Microsoft, Amazon Web Services, Reflection, and Oracle to integrate AI capabilities into its Impact Level 6 and Impact Level 7 network environments. These environments handle information classified up to the secret level and highly restricted data, respectively. However, Warren argues that the department has not provided sufficient transparency regarding the contractual safeguards in place to prevent misuse.
“Ultimately, it is impossible to assess any safeguards and prohibitions that may exist in your company’s agreement with DoD without seeing the full contract, which neither DoD nor your company have made available,” Warren stated. She added that, despite previous Congressional requests, the DoD has not disclosed details about the scope or substance of these agreements beyond allowing technology use on classified systems for all “lawful operational use.”
Warren’s letter to the companies highlights specific concerns that the contracts could grant the DoD and Trump administration officials “free rein to engage in domestic surveillance – including spying on U.S. citizens exercising their legal rights — or build autonomous weapon systems that have enormous power to make decisions about targeting without human intervention.”
Referencing OpenAI’s partially released contract with the DoD, Warren noted that existing language such as “all lawful purposes” and “applicable law” does little to constrain potential misuse. Experts have similarly pointed out that terms related to mass surveillance are so broad they could allow significant latitude. Warren also raised questions about the Defense Department’s decision to award a contract to Reflection AI, a startup backed by 1789 Capital, where President Donald Trump’s son, Donald Trump Jr., is a partner.
When announcing the agreements, DoD officials stated they would “prevent AI vendor lock and ensure long-term flexibility for the Joint Force.” However, previous reporting from *The New York Times* suggests the Defense Department may have used negotiations to pressure Anthropic into dropping safety concerns about military AI usage. Anthropic was designated a “supply chain risk” after resisting demands to remove restrictions on how the Pentagon could use its models.
“These agreements come in the midst of DoD’s ongoing litigation against Anthropic after the company objected to their models being used for all lawful use cases, which would have removed preexisting safeguards preventing its technology from being used to develop fully autonomous weapons and engage in mass domestic surveillance,” Warren said.
The senator is advocating for clearer guardrails, echoing efforts from other Democratic lawmakers. Bills introduced by Sen. Elissa Slotkin (D-Mich.) and Sen. Kirsten Gillibrand (D-N.Y.) seek to prohibit the use of autonomous weapons without human authorization and ban AI-enabled mass surveillance and nuclear weapons launches. The Senate Armed Services Committee has also incorporated similar restrictions into its version of the 2027 defense policy bill.
Warren has requested responses from the Defense Department and the companies in an unclassified format by July 20.
—
*Original Article Source:* [Federal News Network](https://federalnewsnetwork.com) — “It’s impossible to assess any safeguards that may exist in your company’s agreement with DoD without seeing the full contract,” July 8, 2026.
