WASHINGTON (Azat TV) – A coordinated network of 37 websites masquerading as local news outlets has been identified by digital security firm VisionHeight as heavily boosting Democratic candidates across competitive U.S. races ahead of the November 2026 midterm elections, according to an NPR report published on October 30, 2026. These platforms, often described as pink slime journalism, have surfaced directly in search results and citations generated by OpenAI’s ChatGPT and other AI chatbots, creating a new avenue for stealthy political advertising and deceiving voters seeking non-partisan information less than two weeks before Election Day.
The Anatomy of the Network and AI Integration
According to findings gathered by VisionHeight and independent forensics experts, the network comprises interconnected groups of websites set up simultaneously. Registry and digital certificate data reveal that all 37 websites were launched on September 15, with all but one registered within an eight-minute window. Hundreds of articles appeared in batches throughout September and October, suggesting automated content generation using artificial intelligence.
The operation is structured to maximize visibility among AI crawlers. Every website in the network explicitly permits scraping by major AI crawlers and contains a specialized file known as llms.txt, which indexes articles in plain text for retrieval by automated systems. In testing conducted by NPR across popular AI tools including ChatGPT and Gemini, chatbots frequently cited articles from these partisan sites when prompted about competitive congressional races and candidate platforms.
Impact on Competitive Races and Lack of Accountability
The sites target high-stakes contests that could determine control of the U.S. House of Representatives. For instance, in Michigan’s 4th Congressional District, a site called the Kalamazoo Beacon published 16 stories exclusively promoting Democratic candidate Sean McCann, framing him as the most qualified candidate on affordability. When queried about the district, ChatGPT cited the Beacon at the top of its response while linking to it as a credible local reporting source, despite the site functioning solely as a promotional vehicle.
Despite the scale of the operation, the operators and funding sources behind the 37 websites remain unknown. JoAnna Mendoza, a promoted candidate in Arizona, stated she had no involvement and did not know who ran the sites. Major Democratic campaign arms, including the Democratic Congressional Campaign Committee, denied involvement during interviews with NPR, while other allied super PACs did not respond to inquiries.
Institutional Stakes and Regulatory Challenges
Security experts emphasize that masquerading as independent journalism crosses a critical line into deceptive manipulation. Renée DiResta, an associate research professor at Georgetown University, noted that while campaigns naturally optimize for AI visibility, concealing political sponsorship behind fake news mastheads undermines public trust and skews automated search aggregation. Alisha Ukani, a postdoctoral researcher at the University of Washington’s Center for an Informed Public, pointed out that the inclusion of plain-text indexing files demonstrates deliberate technical optimization designed to capture search traffic ahead of the national vote.
As election authorities and frontier AI companies grapple with automated curation flaws, the phenomenon underscores growing vulnerabilities in how voters access verified political information through algorithmic intermediaries.