Overview

• A groundbreaking study led by a team of researchers at Columbia University has challenged the long-held belief that every human fingerprint is unique.
• Using a neural network, the researchers were able to identify similarities between different fingerprints belonging to the same person with a success rate of up to 77 percent.
• The AI focused on the angles and curvatures of the swirls and loops in the center of the fingerprint, rather than the traditional 'minutiae' used in fingerprint comparison.
• While the AI's ability to match different prints to the same person is not yet good enough for real identification purposes, the researchers are confident that the system can be developed further for a higher success rate.
• The findings have received pushback from the forensics community, with some experts arguing that the study hasn't disproven anything radical and that similarities between fingerprints belonging to the same person have been known since the start of fingerprinting.