Fighting Deep Fakes: Think Like the Attacker

The report describes how deepfake-enabled impersonation has become a fast-growing social engineering technique, with attackers combining voice, likeness, and open-source intelligence to convincingly mimic trusted personas and pressure targets into breaking workflows. Brian Long of Adaptive Security argues that organizations should prioritize process-based verification triggered by urgency, secrecy, and exception requests rather than relying on detecting media flaws. He emphasizes that the attack surface now includes voice and SMS—especially in BYOD environments—and details how attacker-modeled simulations can expose gaps to drive targeted training and control enhancements.