Researchers have created artificial fingerprints that they say can be used to hack into smartphones and other devices.
These “DeepMasterPrints” are machine-learning methods that act as a master key, according to the researchers from New York University and Michigan State University, CNBC reports.
Many brands offer fingerprint recognition as a means to access a device.
The resulting fingerprints can be “used by an adversary to launch an attack…that can compromise the security of a fingerprint-based recognition system,” the report continues.
The news comes as what is being called the first hack of the year occurred in Australia. Data on roughly 30,000 civil servants was downloaded after a government employee’s email addresses was phished in the state of Victoria, according to CBR.
The fingerprint researchers — Philip Bontrager, Aditi Roy, Julian Togelius, Nasir Memon and Arun Ross — told CNBC that “phones and many more devices don’t capture your entire fingerprint.”
They add: “There’s not enough space on the device, so they capture a partial fingerprint — which is not as secure as the full image. (People assume) the device stitches images of their fingerprint together, but that’s not really what happens — it keeps sets of partial fingerprints.”
Firms can defend themselves by moving fingerprint sensors from buttons to screens, allowing higher-resolution images, CNBC says.