The Defense Counterintelligence and Security Agency is exploring how artificial intelligence could accelerate security clearance vetting, potentially slashing timelines from months to hours. But a law firm that represents clearance applicants warns that the technology may create more problems than it solves, particularly when human judgment is essential.
The push for speed
DCSA has led the government's background check process since 2019. The volume of applications is cyclical, often spiking during presidential administration changes. At any given time, tens of thousands of applicants sit in backlogs, said Anthony Kuhn, managing partner at Tully Rinckey PLLC. The agency sees AI as a way to identify objective issues that need further review, which could reduce those delays.
However, the vetting process is among the government's most sensitive, and bringing AI into it would mark a turning point for both the public and private sectors, the firm said. The law firm assists candidates whose clearances are in jeopardy over alleged violations of adjudicative guidelines - the 13 standardized rules the government uses to decide if someone is trustworthy enough to hold a clearance.
Where machines fall short
AI could flag certain responses during the application, but it cannot assess mitigation factors, Kuhn said. A candidate might admit to using marijuana, but only for medicinal purposes or because they lived where it was legal. "AI is not going to be able to catch mitigation or understand mitigation," he said.
That limitation carries real consequences. If the agency gives AI enough authority to move flagged applications toward a notice of intent to deny or revoke, the number of cases Tully Rinckey responds to and potentially litigates would increase significantly. A human operator, by contrast, can filter out those cases early when exceptions exist. "Artificial intelligence may help accelerate parts of the background investigation process, particularly by identifying objective issues that warrant further review," Kuhn said. "However, AI is only as reliable as the data behind it, and accuracy concerns remain. Security clearance decisions often require human judgment to evaluate context, mitigation and credibility in ways that a machine cannot."
Security and privacy risks
The highly sensitive vetting process is also susceptible to hackers, depending on how the agency contracts with AI providers. Kuhn said it will be interesting to see what preventative steps are enacted to restrict vulnerabilities. New government regulations would be needed to monitor the use and security of AI tools in the clearance process, he added.
Applicants themselves often misunderstand the technology's risks. "A misconception that people have is they think that AI is safe and they are plugging their information into an AI software platform and getting a response based on their prompts," Kuhn said. "What they are actually doing in a lot of cases is publishing what would otherwise be confidential or protected information."
Alternatives to AI
AI is not the only way to speed up clearances and cut backlogs, Kuhn said. The government could hire more adjudicators, remove questions that tend to provoke responses that get reversed, and apply a more consistent approach across the board. The only application where AI would clearly help, he noted, is a very clean one - a military member with no drug history, no arrests, and no "yes" answers to negative questions. Even then, AI cannot follow up with references like a prior employer to confirm the answers.
"There is a reason we have not implemented AI yet," Kuhn said. "It has been implemented in just about everything else."
Why this matters for Government professionals
The DCSA proposal highlights a tension that will appear across federal agencies: the desire to use AI for efficiency in AI for Government processes versus the need to protect due process and sensitive data. For clearance holders, applicants, and security officers, the outcome will shape how personal information is evaluated - and who or what makes the final trust decision. Understanding the limits of automation in adjudication is not theoretical. It could determine whether a clearance is granted, delayed, or wrongly denied.
Your membership also unlocks:
