America’s Military Is Racing to Integrate AI. Does It Have the Talent? 

The United States is racing to put artificial intelligence at the center of its national security strategy. But America’s ability to compete with China, protect critical infrastructure, and responsibly deploy AI in military settings will not be determined by algorithms alone. It will depend on whether the government has enough people who actually understand how these systems work.

The Department of War needs defense-specific AI talent because in-house expertise is what enables the government to understand and control the technologies its defense and intelligence operations increasingly depend on. Without that in-house technical expertise to independently assess commercial and frontier AI systems, the government will struggle to evaluate performance, manage risk, and retain control over mission-critical capabilities.

Developing an “AI-first” military requires a foundation of AI technical talent, yet talent reviews have consistently identified this crucial deficiency. Additionally, reports have found that the Department is unable to identify and leverage personnel that are AI capable or accurately identify which positions require AI skills.

U.S. policymakers appear to have gotten the message. In the last month, both the White House and the House Armed Services Committee have rolled out efforts to shore up the government’s technical talent.

• The recent National Security Presidential Memorandum directs the creation of an AI National Security Strategic Reserve of non-governmental experts. This designates surge capacity as a national security need and implicitly recognizes that the government can’t permanently hire the AI talent required for a crisis.
• The White House’s Executive Order, “Promoting Advanced Artificial Intelligence Innovation and Security,” focuses on AI cybersecurity and critical infrastructure. Its workforce provision directs the Office of Personnel Management (OPM) to expand the U.S. Tech Force’s cybersecurity hiring and placement pathways.
• The Chairman’s Mark of the House FY2027 National Defense Authorization Act (NDAA) has two provisions regarding the civilian acquisition workforce that could be enacted into law.
  • Section 833 would make a set of performance objectives mandatory for the Department of War’s civilian acquisition workforce. This would include an explicit requirement to maintain proficiency in digital and AI literacy.
  • Section 834 would require individuals in critical acquisition roles to demonstrate proficiency in those performance objectives, and it bars the Department from treating completed training or earned certifications as sufficient proof on their own. Critically, this treats AI literacy as a baseline competency.
• The House Subcommittee of Military Personnel’s print for the FY2027 NDAA has additional talent provisions up for inclusion in the final bill.
  • Section 1106 directs a pilot integrating the U.S. Tech Force into the Department of War, paired with skills-based hiring authorities. It also leans on skills-based assessment rather than credential screening.
  • Section 1108 would establish a pilot aimed at rapidly recruiting innovative talent in critical technology areas. AI is named first among other key technical areas.

Despite all this positive momentum to address the dearth of AI talent throughout the defense enterprise, these efforts share a key weakness: these policies do not address AI talent exclusively and directly. AI appears throughout each document as one priority among a laundry list despite the fact that AI has been a top priority for the Department of War for at least the last six months. Unfortunately, the single AI-specific measure, the Strategic Reserve, is the least durable. These measures have additional limitations.

• The Strategic Reserve, as an executive action, lacks the appropriations and durable statutory authority to survive a change in administration.
• The executive order is narrow, primarily targeting cyber talent specifically and leaving the broader AI-talent pipeline largely untouched.
• The two acquisition provisions, for their part, sharpen the people who buy and oversee AI rather than the technical workforce that builds it. Yet, Sec. 833 leaves “AI literacy” for the Department to define and enforce, and Sec. 834’s force depends entirely on how rigorously the Department builds the proficiency framework. Lastly, both focus on developing the existing workforce rather than recruiting new talent.
• The personnel pilots run into the limits of their own design. Sec. 1106’s reach is tied to the scale and durability of the underlying U.S. Tech Force program, while Sec. 1108, as a time-limited pilot, tests an approach rather than committing to the enterprise-wide hiring and pay reforms that a lasting fix would require.

This initial gambit from Congress and the White House is a decisive opening, but it doesn’t win the talent game the Department of War is playing. These policies all move in the right direction to protect our country and remain at the forefront of innovation. But starting points are not destinations.

Closing the AI talent gap definitively will take AI-specific measures, not merely sprinkling AI into talent initiatives that are meant to broadly serve the Department of War. ARI has built a framework of policy interventions addressing the talent that already exists but cannot yet be exploited, onboarding new talent to support a crisis period, and building the right permanent scaffolding to support an enduring AI talent enterprise throughout the Department of War and parts of the Intelligence Community (IC).

• Cleared AI Talent Reserve: A standing, pre-vetted pool of clearance-holding AI experts from industry and/or academia on reserve, who can be activated into government service on short notice during a crisis. This is similar to the NSPM’s surge capacity solution but made more durable through codification in the NDAA. A congressionally chartered reserve carries statutory durability and appropriations behind it that the executive authority does not have. This proposal also addresses the clearance bottleneck by pre-processing clearances.
• Joint AI Career Track: Each military service should stand up a dedicated AI/machine learning career track spanning both uniformed and civilian personnel. This would mean defining training pipelines and ensuring the ability to move without losing career progression. Where the current acquisition provisions only focus on a portion of the defense workforce, this identifies and designates specialists across the enterprise.
• AI Pay & Incentive System: A national security-specific pay authority for AI and data science roles, exempt from current GS/SES caps, designed to make government compensation more competitive with the private sector. With pay for frontier AI talent far outpacing federal ceilings, no career track or recruiting pilot will reach its full potential if the government is limited to below market compensation. Congress has decoupled pay from the GS scale in the past when markets demanded it, as with the Cyber Excepted Service, and it’s time to do so again.
• Defense AI Fellowship & Education Pipeline. A scholarship-for-service program, modeled on CyberCorps, that funds graduate-level AI education in exchange for a post-graduation service obligation. This shifts the competition with industry to an earlier stage, where the compensation gap is narrowest and in a way that doesn’t just rely on higher salaries. While the pilots are time-limited experiments, a fellowship pipeline will build a renewable, long-term supply that can endure.

The country that wins the AI race will be the one that can recruit, retain, and surge the people who understand and build this crucial technology. The White House and Congress know that AI talent is a national security priority, but current ideas merely gesture at the talent requirements that are well understood by both the Department of War and the IC. Permanent and targeted solutions are required to help build the workforce needed to procure and operate a technology that may very well transform the very nature of war itself. As Congress looks to finalize this year’s NDAA and the White House considers the implementation plan(s) required to make its recent NSM a reality, both branches of government should incorporate the talent provisions that ARI has identified as necessary next steps to building the “AI-first” Department of War and the Administration seeks.