Pentagon Wants $30M AI-Powered Lie Detector

The Pentagon is asking for $30.3 million over five years to develop a more advanced lie detector powered by artificial intelligence and machine learning. The program, referred to as Polygraph+ or Polygraph Next, would focus on scoring algorithms that analyze physiological and behavioral data to detect deception. The request appears in a Department of Defense budget document and signals continued government interest in automated interrogation tools. Supporters argue that AI could make polygraph-style testing more accurate, consistent, and efficient. Human examiners can be subjective, and traditional polygraphs have long been criticized for producing unreliable results. Machine learning might help spot subtle patterns across voice, pulse, breathing, skin response, and other signals. But critics see serious problems. Lie detection is technically difficult, and AI systems can inherit biases from their training data or produce confident false positives. A wrong result in a national security or criminal investigation could have severe consequences. Privacy advocates also worry about how such a system would be used, what data it would collect, and whether subjects could challenge its conclusions. The program's name suggests an upgrade rather than a replacement for existing polygraph methods, but even an AI-assisted scoring tool could shape interrogations and influence decisions about who is trusted. The broader debate is not only about whether AI can detect lies. It is about whether governments should rely on black-box systems to judge truthfulness in high-stakes settings. As the Pentagon pursues the technology, lawmakers and the public may demand transparency, independent testing, and clear limits on deployment. The outcome could set an important precedent for how AI is used in law enforcement, intelligence, and defense.