Pentagon Requested $30.3 Million for AI Polygraph Program
The DCSA seeks to modernize credibility assessments for a workforce of 2.8 million employees through 2031.
Updated on Sept. 25, 2026 in Artificial Intelligence
Live Poll
Do you trust AI-based tools to accurately monitor employee behavior or verify their honesty?
The Defense Counterintelligence and Security Agency has requested $30.3 million to develop an automated AI-driven polygraph program named Polygraph+ or Polygraph Next. This proposal represents a research-stage initiative intended to assist in vetting staff and identifying potential insider threats.
Why it matters
The program aims to replace traditional human-administered assessments with standoff sensing technology to scale security clearance efforts. It arrives against the backdrop of long-standing academic skepticism regarding the accuracy of physiological credibility testing.
The system utilizes AI to analyze physical readings without physical attachments, contrasting with standard polygraphs that claim 80% to 94% accuracy rates. Future performance benchmarks remain unknown as the program transitions from 2023 prototype development to potential funding.
The players
Defense Counterintelligence and Security Agency
The primary federal organization responsible for personnel security and background investigations.
Defense Innovation Unit
The Pentagon organization tasked with accelerating the adoption of commercial technology for military use.
Presage Technologies
A developer of non-invasive sensor technology for measuring physiological vital signs.
Altec Research
A research entity focused on computer vision and physiological data tracking.
The details
The proposed system uses standoff sensing, a method of gathering data from a distance without attaching physical sensors to a subject, to monitor heart and breathing rates. Companies selected for prototypes by the Defense Innovation Unit in 2023 utilize specific modalities: Presage Technologies focuses on cardiovascular metrics, while Altec Research tracks physiological indicators including facial skin temperature, pore activity, and head movement. These measurements are processed via machine learning models designed for automated scoring.
Timeline
2003: National Research Council labeled polygraph evidence as weak.
2023: Defense Innovation Unit picked two companies for prototypes.
2027: First year of proposed DCSA program funding.
2031: Final year of proposed DCSA program funding.
The Tech Race
The push for automated Polygraph+ tools follows a pattern set by the 2003 National Research Council findings, which cast doubt on existing reliability benchmarks. This project marks a shift toward AI-based sensing to overcome the limitations of human-led assessments.
If approved by Congress, this technology would be implemented to vet the 2.8 million individuals currently employed by the Pentagon. It is not currently available for civilian or private sector use, and its deployment timeline depends entirely on the approval of the 2027 budget.
The takeaway
The development of non-contact physiological sensing signals a transition toward fully automated security vetting in high-stakes environments. Readers should watch for upcoming congressional budget hearings between 2027 and 2031 to see if the proposed funding is authorized.
Further reading
For more on how government agencies are integrating automated systems, visit Artificial Intelligence.
Live Poll
Do you trust AI-based tools to accurately monitor employee behavior or verify their honesty?
