"Deepfakes Will Hurt 30% of Organizations' Trust in Biometrics by 2026"

Gartner analysts predict that deepfakes, which are Artificial Intelligence (AI)-generated replicas of a person's likeness, will lower confidence in face biometric authentication solutions for 30 percent of companies by 2026. According to Akif Khan, VP analyst at Gartner, face-based identity verification and authentication systems will struggle to catch up with AI imitations as they become more realistic and easy to generate. Currently, most face biometric solutions rely on Presentation Attack Detection (PAD) to determine the "liveness" of a person trying to authenticate using their face. Presentation attacks involve placing a mask or video of the actual user in front of a camera or scanner. PAD aims to distinguish between a real human face and these imitations. However, attackers are increasingly performing higher-complexity digital injection attacks with deepfakes. These attacks bypass a physical camera by inputting imagery directly into the system's data stream through tools like virtual cameras. This article continues to discuss the expected degradation of trust in biometrics due to the growing sophistication of deepfakes.

SC Media reports "Deepfakes Will Hurt 30% of Organizations' Trust in Biometrics by 2026"

Submitted by grigby1