By Colin Fernandez for The Chron, January 11, 2024, 4:05 p.m., updated January 11, 2024, 4:23 p.m
- AI shows that not every person's fingerprint is unique
- Thanks to the breakthrough, thousands of unsolved cases were solved
Thousands of cold cases could be solved thanks to a breakthrough in fingerprint analysis using artificial intelligence.
A computer that uses an artificial intelligence system has overturned decades of conventional wisdom that each person's fingerprint is unique.
So if a criminal left a thumbprint at one crime scene and an index finger print at another, there would be no way to connect the two.
The breakthrough came when a student at Columbia University tried to figure out whether artificial intelligence could find connections between seemingly very different fingerprints of the same person.
To test the idea, Gabe Guo, a graduate engineer with no background in forensic science, presented images of about 60,000 fingerprints in pairs to a computer.
Thousands of cold cases could be solved thanks to a breakthrough in fingerprint analysis using artificial intelligence. A computer using an artificial intelligence system has overturned decades of conventional wisdom that each person's fingerprint is unique (stock image)
READ MORE: Burglars beware! Scientists have discovered a way to collect human DNA from the AIR around us
If a criminal is suspected of passing through a crime scene, analyzing the air could reveal his DNA and therefore his identity
In some cases the fingerprints came from two different fingers of one person's hand, in other cases from different people.
Over time, the computer managed to detect telltale patterns that suggested two very different-looking fingerprints came from the same hand – something that had never been detected before.
In Science Advances, Guo and colleagues write: “Our most important discovery is that fingerprints from different fingers of the same person show strong similarities; “These results apply to all finger combinations, even to different hands of the same person.”
The results were initially rejected by the forensic community.
A well-known forensics journal rejected the investigation and an anonymous reviewer and editor concluded that “every fingerprint is known to be unique” and therefore it was not possible to detect similarities even if the fingerprints came from the same person.'
But Mr. Guo and his colleagues persisted.
Hod Lipson, a professor of engineering at Columbia University, said: “I don't normally argue about editorial decisions, but this observation was too important to ignore.”
“If this information is decisive, I can imagine that cold cases could be reopened and even innocent people could be acquitted.”
The saliency map highlights areas that contribute to the similarity between the two fingerprints of the same person
READ MORE: The misleading myths of crime drama: Experts reveal where CSI, Broadchurch and Line of Duty go wrong
Our favorite movies and TV shows are full of myths about the world of crime scene investigation
One of the sticking points that led to the research's rejection is that it was not clear what information the AI used to link seemingly unrelated fingerprints that had escaped decades of forensic analysis.
The team concluded that the AI had identified new patterns in the edges of the fingerprint centers that had not been seen before.
Professor Lipson said the research was an example of a novel insight from AI.
He said: “Many people think that AI can’t really make new discoveries – that it just regurgitates knowledge.”
“But this research is an example of how, given a fairly simple data set that has been sitting around the research community for years, even fairly simple AI can provide insights that have eluded experts for decades.”
He added: “Even more exciting is the fact that an undergraduate student with no background in forensics can use AI to successfully challenge a widely held belief across an entire field.”
“We are on the verge of seeing an explosion of AI-powered scientific discovery by non-experts, and the expert community, including academia, must prepare for it.”
The discovery has another potential use: for computer fingerprint identification. Some computer laptops and security systems identify people using a fingerprint. However, if the finger used to create the original print is damaged, for example bandaged, this means the user will be locked out.
With the new system, another finger could now be used instead.