Tens of millions of voters face AI-driven election lies

Tens of millions of Americans are preparing to vote in the November midterms while facing a surge in artificial intelligence generated disinformation. Experts state that these fabricated posts are slipping through platform defenses at an accelerating rate.

Social media companies have scaled back their trust and safety operations in recent years. This shift followed a backlash against earlier efforts to restrict speech, a movement that intensified after Elon Musk acquired X and Donald Trump won the presidency in 2024. The result has been substantial layoffs from teams dedicated to content moderation.

Frances Haugen, a former Facebook data engineer, noted that the cost of launching a high quality information operation has dropped dramatically. She explained that operatives can now tweak posts slightly to bypass detection filters, allowing false narratives to propagate with minimal resistance.

James Talarico, the Democratic candidate for US Senate in Texas, illustrated the speed of these campaigns. While he recovered from the flu earlier this month, false posts claimed he was being treated at a mental health facility in Ohio. The claim originated from a deceptively edited phone call and persisted for several days despite Talarico releasing security camera footage from his home.

The Digital Democracy Institute of the Americas tracked how the lie mutated as it spread. The initial claim about hospitalization evolved into a false assertion in Spanish media that Talarico had attempted suicide. Roberta Braga, the institute’s executive director, said that platforms enforce rules more strictly in English than in other languages. She added that the move toward AI driven moderation means even more content is likely slipping through the cracks.

In Pennsylvania, Democrat Janelle Stelson faced a different type of fabrication. Viral posts claimed an undocumented immigrant arrested for indecent exposure was a paid canvasser for her campaign. Charles Wall, the deputy director of Immigration and Customs Enforcement, amplified the claim in an official statement. Stelson’s team released records proving the man, Chimaobim Adedayo Igbojekwe, held no paid position, yet her Republican opponent Scott Perry continued to run ads using the false narrative.

Platform policies define election misinformation narrowly. Meta requires advertisers to disclose synthetic images and enforces automated labeling on AI generated content. YouTube states that all users are subject to the same community guidelines, banning content that misleads voters about voting times or locations. X has replaced internal safety reviews with a crowdsourced community notes program.

Nora Benavidez, an attorney with Free Press, observed that federal agencies are increasingly involved in amplifying political claims. She stated that the administration has leveraged agencies across the government to influence the election cycle. Stelson’s campaign has sent cease and desist letters to media outlets and launched a complaint accusing the Department of Homeland Security of violating the Hatch Act.