In brief
Across the United States, AI‑generated political advertisements are saturated on social feeds, with a spike from $50,000 in 2024 to over $137 million in 2026, many lacking disclosure of AI use. Experts warn the content could mislead voters and erode trust, while tech firms lobby against tighter AI regulation. A recent survey shows a small fraction of voters view AI as credible, yet many plan to use it for election research.
In cities from Sacramento to Atlanta, scrolling phones are now filled with artificial‑intelligence–generated political ads that appear identical to human‑created content, according to a report by the Xinhua news agency. The surge comes a day before the U.S. Senate and House elections scheduled for November 3, and it has raised alarm among election‑watching experts.
The scale of the wave is staggering. Wesleyan Media Project data, released this month, identified 220 AI‑produced advertisements worth more than $137 million—an almost certain underestimate because detecting AI use is difficult. That figure dwarfs the $50,000 spent on AI‑generated ads during the 2024 election cycle.
Many of the newly created ads lack any disclosure that AI was involved. Analysts warning that this omission could mislead voters and undermine election integrity. One controversial example shows a Republican candidate for governor embracing Hillary Clinton cheek‑to‑cheek, with the slogan “Clinton Lover. Trump Hater.” Another depicts Democratic Senate hopeful Abdul El‑Sayed laughing while looking at cards that read “Death to America” and “Celebrate 9/11.” A third shows Jame Fishback, a Republican primary candidate for Florida governor, hosting a party with Iranian leaders who attack the Star of David with scimitars, a symbol of Jewish identity.
“You can run an information operation now with quality that was impossible in 2016 for super, super cheap,” said Frances Haugen, a former Facebook data engineer turned whistleblower and advocate for transparency and accountability in big tech.
Haugen added that “repositories of deception in AI content,” enable revising a post “just a little bit each time you create them,” crawling through moderation filters. The ability to tweak posts continuously creates a “cottage industry” that advises candidates on how to craft AI‑generated narratives.
An American Association of Political Consultants Foundation survey highlights how voters interact with the technology. While only 1 % of respondents said they considered AI their most credible source of election information, one in five voters plans to use AI to look into a race on their ballot. Twice that proportion already receives election details through an AI search summary, the survey found.
The emerging flood of AI misinformation has caught the eye of policymakers and tech lobbyists alike. The report noted that large tech firms are increasingly lobbying against stricter AI regulation, arguing that such rules could stifle innovation and “undermine the rapid growth of the AI sector.” Meanwhile, experts argue that the unchecked spread of unverified AI ads could distort public debate and challenge the integrity of the democratic process.
