Sign In

Communications of the ACM

ACM TechNews

Privacy Threat Combines Device Identification with Biometric Information


View as: Print Mobile App Share: Send by email Share on reddit Share on StumbleUpon Share on Hacker News Share on Tweeter Share on Facebook
Using a smart assistant.

Researchers have uncovered a privacy threat from electronic devices that allows cyber attackers to access and combine device identification with biometric information.

Credit: Getty Images

Computer scientists at the University of Liverpool in the U.K., New York University, the University at Buffalo SUNY, and the Chinese University of Hong Kong have discovered a new privacy threat from electronic devices that allows cyber attackers to access and combine device identification with biometric information.

Over a one-month period, the team collected and analyzed more than 30,000 biometric samples from over 50 users and more than 100,000 different device IDs. Using these samples, the researchers were able to de-anonymize more than 70% of device IDs and collect biometric data (such as facial images or voices) of device users with 94% accuracy.

Said Liverpool's Chris Xiaoxuan Lu, "Technically, we present a data-driven attack vector that robustly associates physical biometrics with device IDs under substantial sensing noise and observation disturbances."

From University of Liverpool
View Full Article

 

Abstracts Copyright © 2020 SmithBucklin, Washington, DC, USA


 

No entries found