In a high-stakes election year, experts continue to track the vast expansion of AI-fueled state-sponsored disinformation campaigns. Agencies have noted that foreign actors are utilizing AI to improve and accelerate efforts to meddle in our democratic process as Election Day approaches.

Deploying audio, video, image, and text deepfakes, these malicious actors strive to embolden conspiracy theories and deepen divisions between U.S. voters on sensitive domestic issues. To distribute their content, disinformation groups leverage AI-powered automation, a wealth of inauthentic social media profiles and websites made to look like legitimate news sources, and label deepfakes as “leaks” to gain more engagement.

Concerns about deepfake disinformation campaigns trying to sway voters toward one candidate or another have shifted to fears of AI deception disrupting the electoral process on the ground. In the fast-paced environment of vote tallying, AI-generated videos of supposed election fraud could lead to voter manipulation, incite unrest, disrupt operations, and endanger election workers.

One Successful Deepfake, Months of Chaos

These concerns extend far beyond election day. With states expecting a long vote count that could lead to uncertainty about the outcome for days or weeks, the environment of ideological division and low trust in media sources is ripe for exploitation by AI-generated chaos. Deepfakes allow malicious actors to cast doubt on specific parts of the vote count, create inflammatory fake videos of officials, and attempt to manipulate the certification process.

Long after the next president is inaugurated, regardless of the winning party, these groups will continue using deepfakes to sow doubt about the election’s legitimacy, exploiting the ideological rifts in American society. Disinformation groups, particularly those with state-sponsored funding, are improving their tactics in step with advances in AI, becoming more effective and dangerous every day. No matter the damage deepfakes cause during this election and in the ensuing months, 2024 merely a rehearsal for more sophisticated deepfake campaigns of the future.

To protect our democracy against this ongoing threat, we must combat the proliferation of AI-powered disinformation by identifying and debunking deepfakes before they can advance dangerous narratives. Robust tools capable of detecting deepfakes across platforms and at every point of content dissemination ensure that users encountering deepfakes are informed. By spotting deepfakes before they take on a life of their own, we can disrupt the narratives of malicious actors attempting to meddle in the most essential aspects of our democracy.

As a research-forward cybersecurity company, we are constantly looking ahead to anticipate the future of deepfake disinformation. While the fight against AI-fueled chaos begins with reliable deepfake detection, we constantly update our detection models to ensure we are not only keeping up with, but staying ahead of the newest techniques engineers of disinformation use to attack our institutions. Regardless of the methods malicious groups adopt next, Reality Defender will continue to track and combat the deepfake threat today, tomorrow, and into the future.