(New York, N.Y.) — Deep fakes appear to be playing a role in the Russia-Ukraine war, demonstrating its potent potential. Recently, Berlin Mayor Franziska Giffey spoke on a video call with an individual posing as Kyiv Mayor Vitali Klitschko. Giffey abruptly ended the call when the imposter began asking her series of questions that were “unusual” within the context of the war, causing her to question the identity of the individual on the other end of the line.
Counter Extremism Project (CEP) Senior Advisor and University of California, Berkeley professor Dr. Hany Farid has long emphasized the threat that deep fakes pose to national security and the world-altering impacts of this technology—especially in war time.
“Deep fakes pose a serious threat to national security, particularly as the technology advances and especially in the fog of war when false news about foreign policy, the military, and terrorism can easily spread,” said Dr. Farid. “Following Russia’s invasion of Ukraine, we have already seen deep fakes used more frequently. Although the recent incident involving the mayor of Kyiv appears to have avoided any catastrophic consequences, it remains only one of many instances. For example, in March, there was a deep fake of President Volodymyr Zelenskyy calling on troops to surrender that circulated across social media. As these kinds of videos become more believable, it is clear the consequences could be dire. Governments and the tech industry must be ready to combat deep fakes as another front within modern warfare.”
Following earlier deep-fake videos of Zelenskyy, Dr. Farid and student researcher Matyáš Boháček developed a “facial and gestural behavioral model that captures distinctive characteristics of Zelenskyy’s speaking style.” This behavioral model can differentiate between a real video of Zelenskyy and a deep fake, creating an important safeguard against imposters especially at war time.
With the growing threat for real time deep fakes and the use of them in video conferences, Dr. Farid along with Dr. Candice Gerstner of the National Security Agency, also recently wrote about a technique, active illumination, for detecting real time deep fakes during live video calls: “This technique leverages the fact that a video call typically places a user in front of a light source (the computer display) which can be manipulated to induce a controlled change in the appearance of the user’s face. Deviations of the expected change in appearance over time can be measured in real time and used to verify the authenticity of a video-call participant.”