Researchers at New York University have discovered that Artificial Intelligence can be used to create fake fingerprints that could fool biometric scanners and could be used as a hacker’s tool.
According to the report by The Guardian, the team led by Philip Bontrager of the New York University engineering school developed a new Neural Network called“DeepMasterPrints” that can artificially create fake fingerprints with an error rate just one out of 1000. The research was presented at a biometrics conference in Los Angeles last month.
The research presents a detailed insight that how this technique could be used to create replicated fingerprints that could be used in hacking and authentication of biometric ID scanners. The research revolves around the point that many fingerprint scanners only read a partial print, and some different fingertip areas have more resemblance than others.
So when researchers replicated new prints by giving a set of real fingerprints into a generative adversarial network, they were only required to develop that prints that complement with certain portions of other fingerprints—the areas that have a resemblance.
However, Bontrager doesn’t think that this could be used to your phone. “A similar setup to ours could be used for nefarious purposes, but it would likely not have the success rate we reported unless they optimized it for a smartphone system,” he said. “This would take a lot of work to try and reverse engineer a system like that.”
However, if a hacker tried to access a system that had multiple fingerprint-accessible accounts, then it might be possible they would have a good chance of being successful at cracking into a few of them.
Bontrager and his team want the latest research to push the companies and enterprises to beef up fingerprint-security efforts. “Without verifying that a biometric comes from a real person, a lot of these adversarial attacks become possible,” Bontrager said. “The real hope of work like this is to push toward liveness detection in the biometric sensor.”