A team of researchers from Columbia University has challenged the belief that each fingerprint on a person’s hand is completely unique. The team trained an AI tool to examine 60,000 fingerprints and found that it could accurately identify whether prints from different fingers came from the same person with 75-90% accuracy. However, the researchers are unsure of how the AI tool achieves this. They believe that the tool analyzes fingerprints in a different way than traditional methods, focusing on the orientation of ridges in the center of a finger rather than minutiae. The results of the study could have implications for biometrics and forensic science, allowing for the connection of fingerprints found at different crime scenes to the same person. However, more research is needed, and the tool is not currently suitable for use in court cases. Some experts have raised questions about the reliability of the AI tool’s markers and whether they remain consistent over time. The study has been peer-reviewed and will be published in the journal Science Advances.
Not for human consumption
