Russia Accused of Using AI in New Bid to Meddle in US Midterms

U.S. officials have concluded that Russia is conducting a new covert online influence campaign aimed at disrupting the upcoming midterm elections, utilizing artificial intelligence to spread divisive content and erode public confidence. The operation, authorized by Moscow, is designed to amplify inflammatory political material on social media as American voters prepare to decide control of Congress.

This latest alleged interference follows a pattern Russia has employed in past U.S. elections, notably the 2016 presidential race. However, this time, artificial intelligence appears to be a key component, providing a powerful tool for rapidly generating convincing fake videos, images, and political content. Intelligence assessments suggest the campaign's primary objective is to sow chaos and deepen existing political divisions rather than to support a specific party, though recent examples have targeted Democratic Senate candidates.

Threat-analysis company Graphika has attributed a recent surge in activity to a Kremlin-linked operation, dubbed Operation Overload or Matryoshka. This network is known for impersonating legitimate media organizations and disseminating fabricated content across social media platforms. Researchers have detected multiple waves of activity over the past week, appearing on platforms like X and Bluesky, with expectations that the campaign will escalate.

Among the targeted candidates are Democrats Jon Ossoff of Georgia, Sherrod Brown of Ohio, and Roy Cooper of North Carolina, along with others in competitive states. The disinformation has focused on contentious issues such as corruption, antisemitism, and LGBTQ+ rights. One fabricated video falsely depicted actress Jamie Lee Curtis criticizing Democrats on transgender issues; her publicist confirmed the claim was untrue. Despite being false, the post reportedly garnered nearly 200,000 views on X.

While intelligence officials believe Russia traditionally prioritizes presidential elections and may be occupied with its ongoing war in Ukraine, the emergence of this new campaign demonstrates that its influence operations remain active. Russia has historically denied accusations of meddling in American elections.

Beyond Russia, U.S. intelligence assessments also indicate that China could potentially target individual congressional races or lawmakers perceived as hostile to its interests. However, officials suggest Beijing may be more cautious about detection due to potential blowback. Current and former officials describe China as generally more reserved in its approach compared to Russia or Iran.

Iran has also been identified as a potential actor, having previously spread propaganda aimed at undermining U.S. political figures. Meta, the parent company of Facebook and Instagram, recently identified a small network of accounts originating in Iran that focused on American politics, some of which used AI-generated content expressing "anti-Republican views."

The potential for foreign interference comes at a time when artificial intelligence is proving to be an increasingly volatile factor in international relations. In a separate incident, faulty AI-assisted intelligence earlier this year nearly led the U.S. military to intercept a Chinese vessel in the Middle East after its cargo was misidentified. A U.S. special operations analyst reportedly used an AI chatbot to analyze the ship's manifest, resulting in an incorrect assessment that escalated the situation.

Officials familiar with the election interference intelligence stress there is no evidence suggesting Russia is attempting to compromise voting machines or manipulate the casting or counting of ballots. Instead, the focus remains on influencing perceptions and beliefs online during a critical election period.