AI Can Identify When Different Fingerprints Belong to the Same Person, Study Finds

The groundbreaking research by Columbia University challenges the long-held belief in the uniqueness of each finger's prints and could revolutionize forensic investigations.

A new study by Columbia University has found that an AI can accurately identify when different fingerprints in a database belong to the same person, challenging the long-held belief that each finger has entirely distinct fingerprints.
The AI was trained on a publicly available database of 60,000 fingerprints and was able to identify fingerprints from the same person 77% of the time.
The AI's method of identifying fingerprints relies on the curvature of the swirls at the center of the fingerprint, in contrast to traditional forensics which analyze where ridges in the skin branch and end.
The researchers believe that the AI's accuracy can be improved with a larger training dataset of millions of prints.
This technology could potentially help solve cold cases, exonerate innocent people, and provide new leads in forensic investigations.