That sports news story you clicked on could be an AI story

NBC Sportz did not respond to requests for comment. Neither NBCSport.co.uk nor BBCSportss.co.uk has an email address or other contact information publicly associated with them, so WIRED had no way to contact them. (All three sites were registered by domain management company Namecheap, as was the CBS News copycat site that DoubleVerify suspects is within the Artificial Echo network.)

Bad actors have tried to take advantage of successful media outlets by republishing their work without permission Many years. But now, AI tools allow variations of this scheme to proliferate at a newly accelerated pace. “This kind of low-quality content is not really new,” Saporta says. “But it’s much easier to replicate and scale these tools with these existing tools.”

The number of websites relying on AI has increased sharply year over year since generative AI tools became widespread in 2023. Last February, shortly after WIRED first began reporting on the emergence of AI content factories, a monitoring company announced Media NewsGuard has been determined 725 “news and media sites” filled with AI content. By January 2025, it had happened It has been identified At least 1,150 of these sites.

“The volume has gone up,” says Shovik Paul, chief operating officer at AI detection company Copyleaks. “A lot of these operations are run by foreign entities, and they are very suspicious, so how do you keep up with them?”

To make matters even more confusing for readers, a number of major media sites have done this I tried With the publication of news articles generated by artificial intelligence. (Sports Illustrated itself ran content allegedly generated by artificial intelligence, which the parent company said was provided by a third party.) In other cases, Domain name scammers I have purchased URLs for media properties that have fallen on hard times and He sent them Like AI content factories, they sometimes replace their previous podcasts with automated pablum.

Some of these sites actually create real-world confusion; In October, the SEO content mill Post an ad generated by artificial intelligence For a Halloween parade in Dublin, Ireland. Although no such event was planned, crowds of revelers showed up anticipating the festivities.

Copyleaks’ Paul described the way some of these sites hijack the brand identity of genuine outlets to spread spam as “a form of phishing.” In some cases, these sites appear to be making actual phishing efforts. One of the sites within the episode identified by DoubleVerify was designed to mimic a Fox news outlet based in Nigeria. It greets potential readers with a series of suspicious pop-up advertisements for software.

While the pop-ups appear inauthentic, the websites in this group appear to do an active business in programmatic advertising, which are ads placed through large-scale programmatic ad buys rather than a direct relationship between specific websites and advertisers. Many of them feature an abundance of banners run by popular automated ad servers like Criteo and Sharethrough. (Neither Criteo nor Sharethrough responded to requests for comment.) The DoubleVerify report notes that the operators of Artificial Echo chose sports as one of their main content categories specifically because it is considered more brand-safe than hard news.

Programmatic ads from a number of high-profile companies, including technology companies like Asana and Oracle, e-commerce major Net-A-Porter, makeup giant Sephora, and spa chain Kalahari Resorts, appeared as WIRED monitored these sites. None of these companies responded to requests for comment.

At a time when trust in media has declined and many media outlets have seen revenues decline, this kind of ring of the content mill is a double whammy. It pollutes the information ecosystem with spam and plagiarism, and drains programmatic advertising revenue from legitimate content producers.