Polygraphs fail honest people. George Maschke, a veteran with 11 years of security clearance, applied to the FBI in 1995 and told the truth under examination. The machine flagged him as deceptive anyway, specifically on keeping classified information secret and contacts with foreign intelligence.
This is not an edge case. The core problem with polygraphs is that they measure physiological stress responses, not deception itself. That distinction means innocent people fail and trained liars pass. The question the full article pursues is whether any successor technology actually closes that gap.
The piece is worth reading for what comes after Maschke's story: the scientific literature on polygraph validity rates, the specific alternatives being developed or already in use by federal agencies, and whether any of them solve the fundamental problem or just repackage it.
[READ ORIGINAL →]