AI-driven disinformation floods social media a week before the Nov. 3 midterms

Social media feeds are filled with what experts call an unprecedented amount of AI-driven disinformation just days before the Nov. 3 midterm elections. They warn that many false posts are likely "slipping through the cracks." ¹

Companies including Meta, which owns Facebook and Instagram, have scaled back efforts to protect voters from election lies. The Guardian reports that a backlash from the right over fears of speech restrictions led to significant layoffs on relevant teams. That backlash grew after Elon Musk bought X, formerly Twitter, and after Donald Trump's 2024 election. ¹

Platforms also lean more on AI, not people, to respond. Frances Haugen, a former Facebook data engineer turned whistleblower, said bad actors can keep tweaking a post until one gets past the rules. ¹

You can run an information operation now with quality that was impossible in 2016 for super, super cheap

Fake stories in Texas and Pennsylvania

In Texas, Democratic Senate candidate James Talarico took time off earlier this month to recover from the flu. Online posts falsely claimed he was in a mental health facility in Ohio. His campaign released security-camera video showing him at home in Texas. ¹

From Channel GOPWe are going to shock the world and beat billionairesJames Talarico · 136 plays · captioned, with the source named

The Digital Democracy Institute of the Americas tracked how that lie changed as it spread. In conservative Spanish-language media, it became a false claim that he had attempted suicide. The group's executive director, Roberta Braga, said platforms enforce their rules more consistently in English than in Spanish. ¹

Talarico's Republican opponent, Ken Paxton, ran an ad earlier this year using AI-generated footage of Talarico supposedly rewriting the Bible. Talarico told NBC News in September that such ads do not help voters understand their choice. ¹

In Pennsylvania, viral posts claimed an immigrant arrested in the central part of the state worked for Democrat Janelle Stelson's congressional campaign. A Department of Homeland Security statement, from Immigration and Customs Enforcement deputy director Charles Wall, repeated the claim. Stelson's campaign released records showing the man had no paid position with it. ¹

The campaign of Stelson's opponent, Republican Rep. Scott Perry, continues to run ads with the claim. Citizens for Responsibility and Ethics in Washington filed a complaint accusing DHS of violating the Hatch Act, which limits political activity by government employees. ¹

A more cautious view

A Foreign Policy report from Oct. 22 offers a less alarming view of foreign efforts. Matt Masterson of Microsoft said attempts to use AI to deceive "haven't been particularly effective." ²

Other experts said AI helps bad actors work faster and at larger scale, though not necessarily more persuasively. The report also found that election officials, companies and voters are better prepared than in 2024. ²

What happens next

Meta and X did not respond to requests for comment. YouTube said all users are subject to the same guidelines. ¹ YouTube bans content that misleads voters about the time, place or eligibility rules for voting. It does not ban every false claim about a candidate. ¹

The Stelson campaign says it "continues to consider every legal avenue to fight the false claims in Perry's ads." ¹

That matters because many Americans are already voting, and a lie can travel faster than a correction. The pattern in these cases is that a claim shifts form as it spreads, so one correction rarely catches every version. ¹

The Record

1. Sep 1, 2026Talarico told NBC News in September that AI deepfake ads do not add to civic discourse. ¹
2. Oct 22, 2026Foreign Policy reported that experts see AI-aided deception attempts but little proof they have been effective. ²
3. Oct 27, 2026The Guardian reported experts warning of an unprecedented flood of AI-driven election disinformation. ¹

Sources

1. The Guardian, “AI-driven disinformation running rampant on social media before US midterms, experts warn” (2026-10-27). theguardian.com/us-news/2026/oct/27/ai-disinformation-social
2. Foreign Policy, “Midterm election AI disinformation: OpenAI, Anthropic, Microsoft” (2026-10-22). foreignpolicy.com/2026/10/22/midterm-election-ai-disinformat

Found an error? Tell the newsroom. Corrections are noted at the top of the story.