
As the Pentagon ramps up its AI investments under a recent presidential mandate, industry leaders are watching how speed to deployment plays out against other factors.
“Contractors will face significant pressure to compress timelines, and the question is which testing, documentation and certification baselines survive,” said Jessica Tillipman, assistant dean and professor of law at The George Washington University.
Under the Artificial Intelligence Acceleration Strategy, the Defense Department will invest in AI deployment with a stated goal of making the United States the world’s undisputed AI-enabled fighting force, according to an internal Pentagon strategy memo. That means tapping into current resources, unlocking certain data restrictions, building out more infrastructure and reorienting policies to promote speed and agility.
“We will unleash experimentation, eliminate bureaucratic barriers, focus our investments and demonstrate the execution approach needed to ensure we lead in military AI,” said Secretary of Defense Pete Hegseth. “We will become an ‘AI-first’ warfighting force across all domains.”
The new strategy aims to cut the red tape that can hold back movement to integrate AI across every mission area. But Tillipman said the strategy risks trading oversight for speed.
“The memo expects leading AI models to be deployed within 30 days of public release,” she said. “That is an extraordinary pace for a department that spent years building lifecycle testing and assurance frameworks.”
She noted the memo also establishes a Barrier Removal Board that can waive non-statutory requirements, according to the strategy, but doesn’t mention compensating controls for those waivers.
According to the Pentagon’s public release announcing the strategy, the department is moving decisively with a “wartime approach to delivering capabilities,” emphasizing warfighting, intelligence and enterprise operations. “This approach will strengthen battlefield decision-making, rapidly convert intelligence data and modernize daily workflows, all in direct support of more than 3 million” military personnel, the release states.
Tillipman said contractors should watch for two things:
“First, ‘any lawful use’ as a standard contract term changes the bargaining posture. Oversight becomes something the government must justify adding. That affects audit rights, data rights and testing obligations in every AI procurement. Second, the memo makes ‘model objectivity’ a primary procurement criterion without defining it. Vague criteria can create problems at numerous stages, from solicitation drafting through evaluation to award. I anticipate this will be a source of future protests.”