Author(s): Raghu Gopal
Fingerprint readers have become common in smartphones. During the past half-decade, fingerprint sensors have been incorporated into a rising number of top-end devices as an added convenience, but are increasingly a standard feature in many cases. One swipe of the finger unlocks a phone, giving access to apps and services. But this handy feature may be leaving security holes that can be exploited for nefarious purposes.
Last week, researchers at New York University and Michigan State University published findings that suggest smartphones can easily be fooled by fake fingerprints digitally composed of many common features found in human prints. In computer simulations, researchers were able to develop a set of artificial "master prints" that could match real prints similar to those used by smartphones.
Although no two individual fingerprints are identical and full prints are difficult to counterfeit, fingerprint scanners on smartphones are so small that they actually only read partial fingerprints. When a user sets up fingerprint security on a phone, the device typically takes eight to 10 images of a finger to make it easier to make a match. Many users record more than one finger — usually the thumb and forefinger of each hand. A finger swipe has to match only one stored image to unlock the phone, and the system is vulnerable to false matches.
The researchers didn't test their results in a wider real-world study. Nonetheless, the team's conclusion that the use of partial fingerprints for verification can be spoofed so easily is worrisome. Phone makers have acknowledged that fingerprint sensors aren't perfect, but these findings raise questions about the effectiveness of fingerprint security on smartphones.
Current smartphones don't support military-grade biometrics. These fingerprint readers aren't bulletproof, despite being increasingly used for financial transactions, which require a high level of security. Device makers will increasingly explore multifactor authentication to approach foolproof methods of verification, and we have already seen other technologies such as iris recognition being implemented. The novelty stage of biometrics in mobile devices has passed and suppliers of biometric sensors will need to continue exploring ever higher grade security. This includes larger or higher-resolution fingerprint sensors to reduce the risk of biometric hacking, and, even better, combining more than one security measure to protect users further.