In brief
AI‑generated political ads, largely undisclosed, are flooding U.S. social media ahead of the Nov. 3 midterm elections, Xinhua reports on November 2, pinpointing a sharp rise in 220 ads valued at $137 million. Experts warn the content can mislead voters and threaten electoral integrity, while a growing industry tailors AI narratives for candidates.
AI‑driven disinformation is flooding social‑media feeds in the run‑up to the United States' November 3 midterm elections, Xinhua reports on November 2, pinpointing a sharp rise in undisclosed, AI‑generated political ads.
The Wesleyan Media Project, which tracks AI‑generated ads, said it identified 220 such ads in October alone, representing more than $137 million—vastly outpacing the roughly $50,000 spent on AI ads during the 2024 cycle, and it cautions that the figure is likely a conservative estimate.
Some of the ads are politically incendiary: one depicts a Republican gubernatorial candidate clutching former President Donald Trump‑s former rival Hillary Clinton cheek‑to‑cheek, with the slogan "Clinton Lover. Trump Hater," while another shows Democratic Senate hopeful Abdul El‑Sayed laughing at cards that read "Death to America" and "Celebrate 9/11".
Frances Haugen, former Facebook engineer turned whistleblower, said:
You can run an information operation now with quality that was impossible in 2016 for super, super cheap, and you can just keep tweaking the post a little bit each time you create them, and some of them are going to get through.
A survey by the American Association of Political Consultants Foundation found that only 1 % of respondents viewed AI as the most credible source of election information, yet 20 % of voters plan to use AI to research a race, and 41 % have already received election details through an AI‑generated summary.
Co‑operating with the advertising surge, a cottage industry has emerged that advises candidates on shaping AI‑generated narratives; the left‑wing group Run for Something has launched a platform called CampSight that tracks AI‑enabled content.
With the midterm elections just days away, regulators and voters alike are grappling with the implications of undisclosed AI use, as this surge of political content points to a growing challenge in maintaining electoral integrity.
