While adversaries field new capabilities in months, U.S. forces often wait years for similar tools to navigate our acquisition system. This technology gap represents one of the most pressing challenges facing American defense today. With major Defense Department acquisitions taking an average of 11 years to deliver initial capabilities, the slow pace of government acquisition and procurement processes risks leaving warfighters equipped with yesterday’s tools.
The July 2025 AI Action Plan explicitly identifies artificial intelligence as an enabling technology that will “transform both the warfighting and back-office operations of the DoD, “emphasizing that “the United States must aggressively adopt AI within its Armed Forces.” Additionally, Executive Order 14179 calls for removing barriers to American AI leadership, reflecting a growing consensus that the government must rethink how it acquires and deploys emerging technologies.
While the challenge is not new, the advent of AI may offer a long-awaited solution. By helping to accelerate procurement workflows, AI can begin to close the gap between the pace of acquisition and the speed of innovation. Early pilots within the Defense Innovation Unit have already shown promise, with AI-assisted compliance reviews reducing processing time from six months to just three weeks.
The bottlenecks of traditional acquisition
One major reason acquisition processes are slow is the amount of data and in-depth research that this workforce contends with.
Take compliance, for example. Before the U.S. Air Force can acquire a new capability, the request must navigate guidance from the DoD, the Chief Digital and AI Office (CDAO), and the Department of the Air Force, each with its own rules, processes and timelines. Compliance documentation alone can exceed 1,000 pages, so this results in acquisition professionals spending months, if not years, parsing through regulatory documents just to determine whether a proposed tool meets requirements.
Then there’s the need for exhaustive market research. Before making a purchase, teams must investigate whether a similar solution already exists in the federal inventory or if better alternatives are available. This requires searching through multiple systems and various service-specific databases.
Both compliance review and market research involve processing vast amounts of data spread across numerous systems, databases and agencies. These tasks are prime examples where AI can serve as a powerful asset.
How AI can unlock faster, more informed decision-making
The recent AI Action Plan recognizes that “AI is far too important to smother in bureaucracy.” This principle is especially critical for defense acquisition, where AI can serve as a tool to speed up decisions. By using small language models (SLMs) — customized, compact versions of large language models — the DoD and U.S. military can train AI systems on their own acquisition policies, regulations and past procurement data. Unlike general-purpose large language models, SLMs offer superior security, can run on classified networks, and can be fine-tuned specifically for defense acquisition terminology and requirements.
These models excel at specific capabilities critical to acquisition:
Document summarization: Condensing 1,000 page compliance documents into actionable insights
Pattern recognition: Identifying similar purchases across services to prevent duplication
Anomaly detection: Flagging unusual contract terms or pricing that merit further review
AI can also revolutionize market research efficiency. By analyzing contract records from the Federal Procurement Data System (FPDS) and previous purchases in the System for Award Management, AI can highlight duplicate efforts, recommend existing tools already in use, or point to better alternatives. For instance, the Navy’s recent pilot program used AI to identify $12 million in potential savings by flagging redundant software purchases across different commands.
In addition, AI can support long-term planning. By scanning large amounts of defense-related research and data, AI can help spot future capability gaps, giving teams early warnings before those gaps impact operations. The Air Force Research Laboratory is already exploring predictive models that analyze technology trends to forecast capability needs 5-10 years in advance.
Addressing security and accountability concerns
For AI to be trusted in defense acquisition, security and accountability must be paramount. All AI systems handling acquisition data must operate within existing classification frameworks, with SLMs deployed on air-gapped networks for classified procurement activities, ensuring sensitive acquisition data never leaves DoD-controlled infrastructure.
Equally important is maintaining human accountability. AI serves as a decision support tool, not a decision maker. Acquisition professionals retain full authority over all procurement decisions, with AI providing analysis and recommendations that humans validate and approve. This human-in-the-loop approach ensures that the nuanced judgment required for complex acquisitions remains with experienced professionals while leveraging AI’s ability to process vast amounts of information quickly.
Integration with existing systems like contract writing systems and FPDS requires careful planning but is technically feasible through secure APIs and data exchange protocols already in use across the defense enterprise.
Best practices for implementing AI in acquisition
While the benefits of AI are clear, it’s how it’s ultimately implemented that will determine its success. Based on early adoption experiences, here are guidelines for federal agencies and defense organizations to responsibly and effectively integrate AI:
Start with early adopters
Begin with a select group of acquisition professionals who are already open to innovation. These early users can then become trusted advocates, helping to build confidence and support throughout the organization.
Training plays a key role in this process. Teams need to understand how to use AI tools effectively: how to interpret their results, recognize limitations, and spot potential errors.
Build gradually and intentionally
Start with simple and low-risk applications, such as compliance checks or initial market scans. Success metrics for this phase might include:
50% reduction in time for initial compliance reviews
90% accuracy in identifying duplicate purchases
30% decrease in market research timelines
As trust in the system grows and the technology matures, AI’s use can expand to higher-level functions such as predictive acquisition planning, automated generation of full acquisition strategies, and real-time risk assessment across portfolios. These advanced applications would allow defense leaders to simulate procurement scenarios, forecast capability gaps years in advance, and dynamically allocate resources based on mission-critical needs and geopolitical developments.
Prepare for cultural change and growing pains
Technology adoption is as much about people as it is about tools. Introducing AI into the acquisition process will raise concerns: fears of job replacement, mistrust in model accuracy, or resistance to new workflows. Agencies should anticipate these reactions and invest in change management strategies. Identify champions from early pilots who can demystify the tools, share success stories, and encourage broader buy-in.
Address concerns through town halls and working groups, emphasizing that AI augments rather than replaces human expertise. Share concrete examples of how AI frees acquisition professionals to focus on strategic decision-making rather than manual data processing.
Establish governance structures
Create AI governance boards that include acquisition professionals, technologists, legal advisors and mission owners. As the AI Action Plan notes, the federal government should establish frameworks to “ensure that the government only contracts with frontier LLM developers who ensure that their systems are objective and free from top-down ideological bias.” These boards should oversee:
Model performance monitoring and continuous improvement
Ethical use guidelines and bias prevention
Resource allocation (expect initial investments of $2-5M per major command)
Cross-service coordination to maximize learning and prevent duplication
Building an acquisition system ready for the future
Defense acquisition must become faster, smarter and more agile if we are to maintain our competitive edge. Every month saved in acquisition translates to enhanced warfighter capability. AI presents a long-awaited solution to a decades-old problem. But this transformation won’t come from technology alone. It requires a deliberate strategy.
The goal is not to automate away human judgment but to amplify it — giving acquisition professionals the tools they need to make faster, more informed decisions. AI can help us build an acquisition system that delivers relevant capabilities not in a decade, but in real time. The question is not whether to adopt AI in acquisition, but how quickly we can do so responsibly and effectively.
Rick Hubbard is a chief scientist at Core4ce and lead of the Autonomy, Artificial Intelligence and Machine Learning (AAIM) Lab.
The post How AI can accelerate defense acquisition and help U.S. forces keep pace with technology first appeared on Federal News Network.