Deepfake caller challenges improve detection in voice and video calls
Deep-Fake CAPTCHA: Mitigating Next-Generation Social Engineering Attacks
Cryptography and SecurityArtificial Intelligence
Summary
It can be really hard for people to tell if someone on a live video or voice call is using a real-time deepfake impersonation. The authors propose a system called DF-CAPTCHA that actively asks the caller to do simple actions that are easy for real people but tricky for deepfake technology to fake quickly and accurately. The system checks if the responses look real, match the person’s identity, complete the task, and happen fast enough. Their experiments show that this method is better at spotting deepfakes than just looking for signs in the video or audio.
What this means in practice
- •For call center security teams: Use challenge-response tasks during live calls to better detect and block deepfake impersonators trying to socially engineer agents.
- •For online meeting platform developers: Integrate active challenges to verify participants’ identity during live video or voice meetings to reduce risks from deepfake-based attacks.
Authors
Guy Frankovits, Lior Yasur, Fred M. Grabovski, Yisroel Mirsky
Abstract
This paper presents DF-CAPTCHA, an active defense against real-time deepfake impersonation in voice and video calls. Instead of passively searching for artifacts, DF-CAPTCHA prompts the caller to perform simple challenge-response tasks that are easy for humans but difficult for current real-time deepfake systems to generate convincingly. The framework verifies the response using four criteria: realism, identity consistency, task completion, and response time. We evaluate the approach across both audio and video modalities using user studies and experiments with real-time deepfake models. Results show that people often struggle to distinguish real-time deepfakes from authentic media, while DF-CAPTCHA substantially improves detection performance over passive methods, reaching high accuracy in both modalities. These findings suggest that active challenge-based verification is a practical and robust defense against next-generation social engineering attacks based on real-time deepfakes.