Exploiting Voice Cloning in Adversarial Simulation
Demonstrates how deep learning voice cloning bypasses enterprise voice verification systems in realistic adversarial simulations.
Summary
Rise of Voice Verification Systems
Financial institutions and customer service help desks increasingly rely on voice biometric authentication to verify caller identities during sensitive account operations.
Deep Learning Voice Cloning Capabilities
Advancements in neural acoustic modeling allow researchers to generate highly convincing synthetic voice clones from only brief public audio samples.
Bypassing Enterprise Voice Biometrics
Recorded adversary simulations demonstrate how cloned voice samples successfully bypass automated voice verification systems across several major commercial institutions.
Defensive Countermeasures for Red and Blue Teams
Red teams require structured frameworks to emulate synthetic voice attacks, while defensive teams must implement multi-factor verification checks to mitigate AI-driven pretexting.
Related talks
Creator Stage·19 min
Katherine Rackliffe
Presents empirical results from a university research study evaluating user vulnerability and defensive safeguards against AI voice-cloned vishing attacks.
Main Stage·21 min
Gal Zror
Demonstrates real-time executive impersonation using open-source deepfake video and audio models running on commodity GPU hardware.
War Stories·47 min
Jayson E. Street
Demonstrates how generative AI tools enhance phishing and vishing operations and outlines strategies to build organizational situational awareness.