Deepfake CAPTCHA: A Method for Preventing Fake Calls

TL;DR

This paper introduces D-CAPTCHA, an active defense mechanism that challenges real-time deepfake models to expose their limitations, thereby improving detection accuracy and highlighting the threat of sophisticated deepfake attacks.

Contribution

The paper proposes D-CAPTCHA, a novel active challenge-based method that enhances real-time deepfake detection by exploiting the AI's content generation capabilities.

Findings

D-CAPTCHA achieves 91-100% detection accuracy, outperforming existing detectors.

Most volunteers could not distinguish real from fake audio, indicating high realism of deepfakes.

Preliminary results show the effectiveness of challenge-based detection in real-time scenarios.

Abstract

Deep learning technology has made it possible to generate realistic content of specific individuals. These `deepfakes' can now be generated in real-time which enables attackers to impersonate people over audio and video calls. Moreover, some methods only need a few images or seconds of audio to steal an identity. Existing defenses perform passive analysis to detect fake content. However, with the rapid progress of deepfake quality, this may be a losing game. In this paper, we propose D-CAPTCHA: an active defense against real-time deepfakes. The approach is to force the adversary into the spotlight by challenging the deepfake model to generate content which exceeds its capabilities. By doing so, passive detection becomes easier since the content will be distorted. In contrast to existing CAPTCHAs, we challenge the AI's ability to create content as opposed to its ability to classify content. In this work we focus on real-time audio deepfakes and present preliminary results on video. In our evaluation we found that D-CAPTCHA outperforms state-of-the-art audio deepfake detectors with an accuracy of 91-100% depending on the challenge (compared to 71% without challenges). We also performed a study on 41 volunteers to understand how threatening current real-time deepfake attacks are. We found that the majority of the volunteers could not tell the difference between real and fake audio.