NSA, Partners Issue Advisory on Combating Deepfake Technologies

The National Security Agency and its federal agency partners have released a joint cybersecurity information sheet on how deepfake technology could present cybersecurity risks for the Department of Defense, defense industrial base organizations and national security systems.

According to NSA applied research mathematician Candice Rockell Gerstner, cyber actors use deepfake to manipulate authentic media easily, generating new challenges to national security. She said organizations and employees must know how to recognize deepfakes and have a plan in place to deter and respond to deepfake-enabled attacks.

Deepfake technology comes in various forms, including impersonating people, using fraudulent communications to gain access to networks and mimicking organizations’ identities. Technological advancements such as enhanced computational power and deep learning have made deepfake generation easier, less expensive and more effective.

The information sheet includes several recommendations to detect and deter deepfakes, including real-time verification, passive detection and the protection of high-priority officers and their communications, NSA said.

The NSA’s warning comes amid a significant jump in deepfake-enabled fraud across North America. According to the market intelligence provider Payments Cards & Mobile, deepfake deployment has more than doubled from 2022 to the first quarter of 2023, jumping from 0.2 percent to 2.6 percent in the United States and from 0.1 percent to 4.6 percent in Canada.