A group of researchers has found that the rich data that is gathered from driving is enough to identify people with accuracy.The scientist team from the University of Washington and the University of California at San Diego were able to identify drivers based only on data from internal computer networks of the vehicle their test subjects were driving.Known as the CAN bus, the computer records data on everything from braking to steering.In terms of results the team was able to “fingerprint” the correct driver out of 15 individuals about nine times out of ten, after just 15 minutes of driving. Furthermore, it was possible to reach high identification rates using less than 8 minutes of training data.”With very limited amounts of driving data we can enable very powerful and accurate inferences about the driver's identity,” Miro Enev, a former University of Washington researcher who worked on the study told Wired.In terms of real-world identity applications, the tech could have implications for everything from insurance to law enforcement and vehicle personalisation.However, the study also notes that such technology could raise privacy concerns.”Our results suggest that vehicle sensor data has signifcant potential for enabling powerful inferences-some of which may be undesirable from the perspective of the driver whose actions were captured in the data streams.”While we do not expect to diminish consumer appetite for interactive, personalized, and connected experiences in their vehicles we believe that our work sheds new light on the potential privacy risks with computerized automobiles.”

Also in the news:

Default ThumbnailPandemic has ‘pummelled’ biometrics market: ABI Default ThumbnailUIDAI introduces 'Virtual ID’ Default ThumbnailNeurotechnology unveils face recognition algorithm