“We don’t know for sure how it does it” – AI baffles scientists by correctly assigning individual fingerprints to one person, but they are unclear how it actually works

Key Takeaways:

– A groundbreaking study at Columbia University challenges the belief that each fingerprint is entirely unique.
– The research team developed an AI tool that can analyze fingerprints to determine if they belong to the same person.
– The technology has an accuracy of 75% to 90% in identifying whether prints from different fingers originated from one person.
– The exact workings of how the AI achieves this are still unknown.
– The tool focuses more on the orientation of ridges in the center of a finger rather than the pattern of individual ridges.
– Further research is needed, including a larger dataset and lower-quality prints, to refine the technology.
– The findings could have a significant impact on biometrics and forensic science.
– The AI tool could potentially link unidentified fingerprints from different crime scenes to the same individual.
– The revelation that fingerprints may not be entirely unique is not surprising to some experts.
– The study has undergone peer review and will be published in the journal Science Advances.

TechRadar:

A groundbreaking study at Columbia University is challenging the long-held belief that each fingerprint on a person’s hand is entirely unique. 

According to the BBC, the research team has developed an AI tool capable of analyzing human fingerprints to determine if they belong to the same individual.

Source link

AI Eclipse TLDR:

A groundbreaking study conducted at Columbia University challenges the widely held belief that each fingerprint on a person’s hand is entirely unique. The research team has developed an AI tool capable of analyzing human fingerprints to determine if they belong to the same individual. The technology was able to identify, with an accuracy of 75% to 90%, whether prints from different fingers originated from one person. However, the exact workings of how the AI achieved this remain unknown. The researchers theorize that the tool analyzes the orientation of ridges at the center of a finger, rather than the pattern of individual ridges. While further research is needed, the findings could significantly impact biometrics and forensic science. If an unidentified thumbprint is discovered at one crime scene and an unidentified index fingerprint at another, the AI tool could potentially link the two prints to the same individual, assisting police investigations. It is worth noting that fingerprints may not be entirely unique, as no two people have yet demonstrated the same fingerprints.