The idea that every fingerprint on a person’s hand is entirely unique is currently being contested by Columbia University study.
An AI program was trained by a team at the US university to analyze 60,000 fingerprints in an attempt to determine which ones belonged to the same person.
According to the researchers, the system can determine if prints from many fingers belong to a single person with 75–90% accuracy.
However, they are unsure of its mechanism.
The study’s supervisor, Columbia University roboticist Prof. Hod Lipson, said, “We don’t know for sure how the AI does it.”
“It is clear that it isn’t using traditional markers that forensics have been using for decades,” Professor Lipson stated.
“It seems like it is using something like the curvature and the angle of the swirls in the centre.”
Professor Lipson noted that both he and undergraduate student Gabe Guo were taken aback by the result.
“We were very sceptical… we had to check and double check,” he stated.
Others in the field might not be surprised by that.
According to Hull University forensic science professor Graham Williams, the concept of distinct fingerprints has never been finalized.
“We don’t actually know that fingerprints are unique,” he stated. “As far as we know, all we can say is that
The findings of a study conducted at Columbia University may have implications for forensic science as well as biometrics, which is the use of a single finger to unlock a device or provide identity.
The AI tool might be able to determine that, for instance, an unidentified thumb print from crime scene A and an unidentified index finger print from crime scene B are not now forensically tied to the same person.
Despite the fact that none of the Columbia University team had experience in forensics, they acknowledged the need for additional study.
Large volumes of data are usually used to train AI technologies, therefore additional fingerprints would be needed to advance this technology.
Furthermore, every fingerprint utilized to create the model was entire and of high quality, in contrast to the reality where it is more common to find partial or subpar prints.
“Our tool is not good enough for deciding evidence in court cases but it is good for generating leads in forensics investigations,” said Guo.
However, Dr. Sarah Fieldhouse, Staffordshire University’s associate professor of forensic science, stated that, at this point, she did not believe the study would have a “significant impact” on criminal casework.
