A survey on proactive deepfake defense: Disruption and watermarking

The rapid proliferation of generative AI has led to led to unprecedented capabilities in synthesizing realisticdeepfakes (DFs) across multiple modalities. This raises significant concerns regarding privacy, security, andcopyright protection. Unlike passive detection approaches that operate after DFs have been created anddistributed, proactive defense mechanisms aim at preventing the generation of malicious synthetic content atits source. This article provides a comprehensive survey of current proactive DF defense strategies, including Disruption and Watermarking. Disruption approaches protect individuals’ data by introducing imperceptibleperturbations that prevent unauthorized exploitation by generative models, while watermarking approachesembed verifiable messages into data or models to enable content authentication and attribution. We alsoanalyze proactive approaches across various evaluation metrics (imperceptibility, protectability/detectability,transferability, traceability, and robustness), and examine their effectiveness in real-world settings. Furthermore,we review the evolution of DF generation techniques, highlighting their rapid developments. Finally, weidentify key challenges and promising future research directions to enhance proactive defense mechanisms.