Fingerprint analysis has been a dependable tool in crime-solving for more than a century. Investigators lean on fingerprint evidence to identify suspects or connect them to specific crime scenes, believing that every print offers a distinctive code.

Yet, a team of researchers has found that prints from different fingers of the same person can sometimes appear more alike.

This insight came from an artificial intelligence model that revealed surprising connections between prints.

Hod Lipson, from Columbia Engineering, stands out in this effort to question widely accepted forensic norms, in collaboration with Wenyao Xu from the University at Buffalo.