The current midterm election cycle is being significantly influenced by what experts call slopaganda. This term describes campaign materials that combine generative artificial intelligence output with propaganda and misinformation. The concept was coined in early 2025 by a group of AI specialists in a research paper. It now represents a major operational challenge for major internet platforms such as Instagram, YouTube, and X as they strive to maintain transparency and monitor user content.
There is currently no comprehensive federal law that specifically governs the use or disclosure of AI in political advertisements. While approximately 30 state laws exist with varying requirements regarding AI disclosure and usage, candidates do not always comply with these regulations. Consequently, users on platforms operated by Meta, Google, and Elon Musk’s SpaceX frequently encounter AI generated ads and posts from politicians mixed with authentic media.
A recent analysis by the Wesleyan Media Project highlights the scale of this trend. The organization reported that more than $187 million has been spent on nearly 300 ads that were either generated or enhanced by AI. The report notes that this figure is likely an undercount because many advertisements lack proper disclaimers. The data indicates that over 90% of this spending came from Republican candidates, party committees, or Republican leaning groups, with most of these ads failing to include disclosures.
CNBC identified several AI generated ads running online without adequate disclosure. One notable example involves New York Democratic Governor Kathy Hochul, whose opponent, Republican challenger Bruce Blakeman, posted a fake video of her. In Texas, Republican senatorial candidate Ken Paxton released an AI generated video showing Democratic opponent James Talarico making controversial statements. The video included a three second disclosure that flashed in the middle, and the campaign shares it as organic content on YouTube and X. Neither the Blakeman nor the Paxton campaigns immediately responded to requests for comment.
Platform responses vary in their approach to synthetic media. YouTube stated that it applies prominent labels to realistic AI content. For organic content, political candidates can request the removal of videos where their face appears to be altered or generated by AI. For advertisers, the company requires disclosure when an election ad contains synthetic or digitally altered people or events. Google, YouTube’s parent company, also mandates that political ads complete a verification process and include paid for by disclosures.
X places the responsibility on advertisers to comply with applicable laws, rather than the platform itself. The company did not respond to CNBC’s request for comment. Meta, meanwhile, pointed to its existing guidelines for political ads, which include restrictions on usage in the week prior to an election. The company mandates paid for by labels and maintains an ad library. In June, Meta enhanced its labeling efforts by introducing a new about this ad tool for ads detected as being created with third party AI tools.
The role of tech platforms in election integrity has been scrutinized in the courts as well. Last month, a New Mexico court ordered Meta to pay $25 billion in penalties related to a data privacy case stemming from the Cambridge Analytica scandal. That political consulting firm allegedly accessed the information of 87 million Facebook users in an improper fashion nearly a decade ago.
Researchers are also testing how large language models handle requests for deceptive content. The Brennan Center for Justice tested several models, including ChatGPT, Gemini, Grok, Claude, Perplexity, and DeepSeek. An August study found that while models might avoid familiar conspiracies, they can often be manipulated into creating misleading material. A retest in October showed that ChatGPT had made progress, as it began redirecting prompts for harmful election images into comic style illustrations or refusing to generate them entirely.
Legislators are attempting to address the issue through new legislation. The Protect Elections from Deceptive AI Act, supported by senators from both parties, would prohibit the knowing distribution of materially deceptive AI generated video or audio about federal candidates. Other bills, such as the AI Ads Act, are led by Democrats in Congress and seek to expand current bans on campaign misrepresentation to include generative AI content. However, none of these proposed bills have yet advanced out of committee.
