It’s official: Nobody likes AI slop. Trust in the answers AI assistants give for news has sunk to just 20% worldwide, against 37% for news overall, according to the Reuters Institute’s 2026 Digital News Report, out last month. That number lines up with everything else showing up in the research this year: Influencer marketing agency Billion Dollar Boy reports that preference for AI-generated creator content has dropped 44% since 2023, with sentiment now split down the middle between people who see AI as a positive force and those who see it as a negative one. A Fractl survey this spring found the share of consumers who consider AI helpful fell from 82% to 54% in a single year. In news specifically, audiences consistently say that AI labels make them trust a story less.

Once the cost of content went to zero, the backlash was inevitable. Content-analytics firm Graphite reported last fall that the rate of production of AI-generated articles had pulled even with human-written ones, briefly edging ahead in late 2024. Human writing still dominates the places people actually look, though. Graphite found 86% of articles ranking in Google, and 82% of those cited by ChatGPT and Perplexity, were written by people.

The lazy solution is to just ban AI content, and a few outlets have done exactly that. Medium barred it from its paid Partner Program, and the sci-fi magazine Clarkesworld had to pause submissions after a flood of AI-generated spam. But bans are a sledgehammer, and they take down more than the target. People who use AI, paired with human judgment, to enhance and improve their content get swept out with the spam.

Bans also assume the filters can tell the difference, which they mostly can’t. AI detection is notoriously unreliable and prone to false positives. Layer in sophisticated prompting plus the generational jumps in AI models, arriving every few months, and the whole thing turns into an arms race the detectors keep losing. What works today may not work tomorrow.

The better approach is the scalpel, not the hammer: cut away the poor, valueless AI content, but leave intact the AI-enhanced work that audiences appreciate. The way to do that is to point the filters at outputs and outcomes, not the mere presence of AI.

Two moves against slop

In the last couple of weeks, two large platforms made moves that show the split between these approaches. YouTube introduced new controls on certain content types, including the all-too-common AI-narrated video padded out with stock B-roll or generative imagery. Those videos, along with a few other cases, are now harder to monetize, which removes the main incentive to make them.

In the world of text, Substack rolled out an AI detector powered by Pangram. It shows up two ways. First, writers get a button on the publish flow that scans the post and returns an estimate of how much was written by a human and how much by a machine. Most writers already know whether they used AI, but for larger publications with guest contributors, the feature could work as an extra vetting step. Second, readers can scan any article on the platform for AI writing as long as it was published after July 21, 2026.

Again, to be clear: no one likes slop, and it should be disincentivized. But whether a piece is synthetic tells you nothing about whether it’s any good. It’s ultimately up to each reader what to do with that score, but Substack’s scanner quietly pushes a simple equation into the room: AI equals bad.

Then there’s the false-positive problem that has dogged AI detection since day one. A Stanford study on GPT detectors found they misclassified 61% of essays by non-native English writers as AI-generated, and at least one detector flagged 97% of them, while essays by native writers drew just a 5% false-positive rate. The tools were mistaking unfamiliar rhythm for machine output.

At platform scale, even small error rates get ugly fast. As one analysis noted, even a 1% false-positive rate would wrongly flag thousands of pieces a year at a single mid-size operation. Extrapolate that across a Substack, or a Forbes.

Target content, not process

The more durable approach is to look downstream of the detection step, which is essentially what YouTube did. YouTube has an easier problem, to be sure, since at its scale distinct abuse patterns show up fast. But the principle travels: bad content gives itself away in the behavior it produces. Repetitive formulas, weak engagement, high bounce rates, negative comments, and the rest.

For Substack, the fix is to name the behavior specifically. The detector treats “made with AI” as the thing worth flagging, when the real target is content that’s valueless to readers. Those aren’t the same thing. Substack would do better to name the behavior it wants gone and target it directly: posts that use a lot of words to say nothing, auto-generated digests with no human judgment behind them, or even whole publications spun up to feed crawlers instead of readers. If a brand stands up a Substack and pipes in bots to flood the zone with narrative-shaped filler, downrank it or clear it out. Detection can help there, as a signal on the back end. But the label a reader sees should be about the quality of the work, not the tools behind it.

In the interest of full disclosure: The Chatbox, the news digest inside my Substack newsletter, is built with heavy AI assistance. It’s also prompted against a knowledge base tuned to what media people need, edited by a human, and published because a person decided it was worth your time. The AI is already disclosed to readers. A provenance scan adds nothing to that and introduces a signal readers may use as a blunt filter.

Filters are easy, calibrating is hard

Zoom out and all of this is progress. Both YouTube and Substack’s moves point to platforms getting smarter about filtering. That’s good news for anyone in the business of making or reading things online. The market never cleaned up slop on its own, so having filters in place is welcome.

The open question is calibration. Get it right, and the internet a couple of years from now looks better than it does today: creators using AI to sharpen what they make, audiences getting more of what they actually value, and the slop filtered out before it clogs the feed and degrades everything. Get it wrong, and we’ve just built a more expensive way to distrust each other.

The filters are here, and mostly that’s welcome. The harder part is pointing them at the right targets.

A version of this column appears in Fast Company.

The post What YouTube and Substack got right and wrong about AI slop appeared first on The Media Copilot.

YouTube trust and safety chief Matt Halprin sat down for a Creator Insider video last week to make more explicit what the platform means by “inauthentic content.” The result, rolled out July 16, is a set of clarifications that spell out three specific buckets of video that can no longer earn money through the YouTube Partner Program.

The update isn’t a brand-new rule. As TechCrunch reported, YouTube already moved last year to stop creators from making revenue off mass-produced, repetitive videos that AI tools make cheap and fast to churn out. This latest change adds detail to those existing guidelines rather than replacing them.

The first category is generic, repetitive or template-based content. Halprin described channels stuffed with cookie-cutter clips made through AI, CGI or templates that barely change from one video to the next.

The second targets what YouTube calls off-putting content, meaning videos built to distress or emotionally manipulate viewers into clicking. Halprin’s example: an animal shown in distress before someone conveniently arrives to rescue it.

“We’ve heard from our viewers that that’s not something that they like,” he said. Channels dedicated to this content lose Partner Program access whether or not AI made the videos.

The third bucket goes after AI personas, which are AI-made representations of real people.

Halprin was careful not to frame AI as the villain. “AI can actually allow people to make a lot of videos,” he said. “Sometimes those videos are great, and it really enhances creativity.” The same tools, he added, also let people spit out large volumes of near-identical clips with no narrative arc, which is the content farming YouTube wants out of its monetization program.

The Partner Program, which pays creators through ads and subscriptions, is central to YouTube’s business, and the platform now pulls in more ad revenue than Disney, Paramount and Warner Bros. Discovery. Letting the feed fill with low-quality AI output risks the viewer trust that keeps those ad dollars flowing.

For publishers and newsrooms experimenting with AI-assisted video, the takeaway is about intent, not tools. YouTube isn’t penalizing AI use itself. It’s penalizing volume without originality, manipulation for clicks and synthetic voices on topics where accuracy carries real stakes.

The biggest unanswered question is where YouTube will draw the line. Halprin said channels with too much repetitive, low-effort or manipulative content will lose monetization, but he did not define the threshold. The clarified policy applies immediately to all YouTube Partner Program members.

The post YouTube tightens monetization rules around AI slop appeared first on The Media Copilot.

In 2023, a company called Alabama Community News LLC supposedly spent $3.2 million to buy 47 weekly newspapers across the state. The corporate owners fired the local staff, replaced them with an artificial intelligence system that scraped high school sports scores, and promptly drove the entire network into bankruptcy. The story even named a specific 26-year-old campaign staffer who generated 70 percent of the copy.

None of it actually happened. The entire 1,900-word saga was a fabrication published by a site called The Editorial, according to an investigation by Nieman Lab. The targeted newspapers, including the Shelby County Reporter and the Centreville Press, are still printing. The angry local advertisers quoted in the piece do not exist. The story falsely claimed the roll-up was funded by 1819 News, a real conservative outlet in the state, adding a layer of plausibility to the hoax.

The fake story gained traction among journalists on social media platforms like Bluesky before the operators pulled it down. They replaced the page with a sterile retraction notice citing “fact-verification concerns.” But the Alabama hoax was not an isolated incident.

The Editorial has built a bizarre subgenre of AI-generated obituaries for real American newspapers. The site previously published fabricated stories detailing the collapse of the Chattanooga Times Free Press, the Kenosha News, and the Macon Telegraph. The nonexistent reporters credited with these stories sport fake resumes claiming past stints at ProPublica and Reuters.

The motive behind the site remains murky. Domain registration and payment records point to a Finnish technology company called Nordiso Group, which develops AI study apps. Yet the site’s political sections suggest a different angle. The Editorial publishes a high volume of geopolitical content focused on Taiwan and the South China Sea, heavily pushing narratives that highlight Chinese military dominance.

These geopolitical stories share obvious synthetic fingerprints. Nearly every piece opens with a variation of the exact same scene: a nondescript, windowless conference room where a secret document slides across a table. This repetitive structure aligns with tactics tracked by groups like the Stanford Internet Observatory, which monitors state-sponsored disinformation campaigns. It also highlights how cheap synthetic media allows operators to flood niche topics, a trend we track closely at The Media Copilot.

For publishers, this represents a strange new vector of reputational risk. Newsrooms are used to fighting disinformation about elections or public health, but now they must monitor for synthetic hoaxes about their own business operations. A fake story about a newspaper shutting down or firing its staff can spook actual advertisers and confuse real subscribers before the publisher even realizes the rumor exists.

The barrier to generating convincing local news copy is gone. Operators no longer need to understand the nuances of a community to write a plausible story about it. They only need a prompt and a target, leaving local editors to clean up the mess when the synthetic fallout hits their own backyards.

The post AI fake news network invents the collapse of 47 local Alabama newspapers appeared first on The Media Copilot.

A former startup CEO who pleaded guilty to defrauding investors has been linked to a network of AI-generated fake local news outlets — built, by his own account, to manipulate how both people and search engines perceive people like him.

Drew Chapin, who in 2021 pleaded guilty to investor fraud in connection with a failed tech startup, founded The Discoverability Company, a Philadelphia-based online reputation management firm. According to an investigation by The Florida Trib in partnership with KCRW’s Question Everything podcast, that firm is behind the South Florida Standard, a website that presented itself as a local news outlet but was, in fact, entirely AI-generated.

The South Florida Standard featured fake reporters with AI-created headshots and fabricated biographies. Stories were lifted from legitimate news organizations, run through AI, and republished as original content. The site published three stories a day, seven days a week—including Easter Sunday—under bylines that had no professional history or digital footprint outside the site.

The problem of AI-generated content disguised as legitimate journalism is one our coverage has tracked closely, and it’s getting harder to solve. Even as platforms roll out content credentials and provenance tools, Wikipedia has moved to ban AI-generated text entirely from its 7.1 million articles, citing hallucinations and fabricated citations. The South Florida Standard is a case study in why those bans exist.

Frechette’s analysis traced the South Florida Standard to at least two sister sites: the Charleston Sentinel in South Carolina and the San Francisco Download in California. All three were built from the same source code and controlled by the same entity. Across the network, The Florida Trib identified at least nine “reporters” who share names with people accused of or convicted of fraud or conspiracy.

Chapin acknowledged responsibility for the network, describing it as a six-month experiment to build what he called “geographic topical authority,” testing whether AI-generated news sites could rank alongside legitimate outlets in search results. He said he stood up 17 similar sites across the country, producing more than 3,500 URLs drawing more than 44,000 visitors.

“I don’t know whose job it is to make sure that people are represented fairly and wholly online,” Chapin told the outlet.

The experiment, by Chapin’s own account, didn’t work particularly well. Search engines, he said, could tell the difference between The New York Times and a fake outlet. His sites weren’t breaking through.

But the broader damage is harder to quantify. Known in academic research as “pink slime” journalism—named for the cheap meat by-product used as a filler—these fake news sites now outnumber local daily newspapers in the United States, according to data analysis firm NewsGuard. As of June 2024, NewsGuard identified 1,265 such outlets nationwide, surpassing the 1,213 daily newspapers still operating.

Florida, with the lowest number of news outlets per capita in the continental US, may be especially vulnerable. Researchers have found the state is already home to dozens of pink slime outlets, part of a national network of more than 1,000 sites backed by conservative think tanks, donors, and political operatives.

“Stuff like this has zero value to the public,” said Kelly McBride, a senior vice president at The Poynter Institute. “And in fact it has a negative impact on the news ecosystem, because it clutters the environment.”

The findings arrive as traditional local news continues its historic collapse. Real newsrooms are already experimenting with AI-generated content, and their own workers are pushing back. Since 2005, the country has lost almost 2,900 newspapers and roughly two-thirds of its newspaper journalists—43,000 positions—according to Northwestern University’s Medill Local News Initiative. As real local news disappears, the vacuum doesn’t stay empty for long.

Edited by Pete Pachal

The post A fraudster built a network of fake AI news sites to manipulate search results appeared first on The Media Copilot.

YouTube has removed several of the most popular AI-generated content channels on its platform, The Verge reports. The purge includes CuentosFacianantes, which had 5.9 million subscribers and 1.2 billion views churning out low-quality Dragon Ball content, and Imperio de Jesus, a 5.8-million-subscriber channel featuring AI-generated religious quizzes.

Both channels were flagged in Kapwing’s November 2025 report on the rise of AI “slop” — the spam of the video-first age. At least 16 other channels from that report have since been deleted or emptied of content.

The contradiction

The removals came weeks after YouTube CEO Neal Mohan announced plans to “reduce the spread of low quality AI content” in his 2026 priorities letter. But here’s where it gets complicated: YouTube is simultaneously encouraging creators to use AI tools for Shorts and plans to let them generate AI likenesses of themselves.

Mohan has called generative AI “the biggest game-changer for YouTube since the original revelation that ordinary folk wanted to watch each other’s videos.” The company is betting on AI as a creative tool while cracking down on AI as a content farm.

Why this matters for publishers

YouTube’s crackdown signals that platforms are starting to draw quality lines around AI content. The distinction isn’t AI versus human — it’s whether AI is being used as a creative tool or a content mill.

For media companies producing video:

• AI-assisted production (editing, thumbnails, research) remains encouraged
• AI-generated content farms are now explicitly targeted
• Authenticity signals may become more important in recommendation algorithms

The channels YouTube removed weren’t small experiments. Bandar Apna Dost, an Indian AI slop channel featuring a “realistic monkey in hilarious human-style situations,” had 2 billion views and estimated annual earnings of $4.25 million, according to Kapwing’s analysis.

That’s real money YouTube is walking away from. It suggests the platform sees AI slop as a long-term threat to advertiser confidence — the same calculus that eventually forced action on misinformation and clickbait.

Publishers who’ve worried about competing with AI-generated content farms may find the playing field tilting back in their favor. But the rules remain fuzzy: YouTube hasn’t published clear guidelines on what separates acceptable AI-assisted content from removable AI slop.

The post YouTube is purging AI slop channels while pushing creators to use AI tools appeared first on The Media Copilot.

YouTube will crack down on low-quality AI-generated content while simultaneously rolling out more AI tools for creators, CEO Neal Mohan wrote in his annual letter this week.

“The rise of AI has raised concerns about low-quality content, aka ‘AI slop,'” Mohan wrote. “As an open platform, we allow for a broad range of free expression while ensuring YouTube remains a place where people feel good spending their time.”

The platform plans to build on existing spam and clickbait detection systems to reduce the spread of “low quality, repetitive content.” But the same letter teases AI-powered Shorts that let creators generate video using their own likeness and new tools for experimenting with AI-generated music.

Mohan framed the tension as manageable: “Just as the synthesizer, Photoshop and CGI revolutionized sound and visuals, AI will be a boon to the creatives who are ready to lean in.”

For newsrooms producing video, the letter signals where YouTube’s algorithmic priorities are heading. Quality signals will matter more as the platform tries to distinguish human-crafted content from AI filler. At the same time, YouTube’s own AI tools could help smaller teams produce more polished output — or flood the platform with more of what Mohan calls slop. The 2026 Reuters Institute predictions flagged exactly this tension between AI-enabled efficiency and content quality.

The company also promised stronger deepfake detection, expanded AI content labels and tools to let creators protect their likenesses from unauthorized AI use.

“It’s becoming harder to detect what’s real and what’s AI-generated,” Mohan acknowledged. “This is particularly critical when it comes to deepfakes.”

YouTube’s balancing act reflects a broader industry pattern: platforms racing to offer AI features while managing the quality collapse those same features enable. Whether the spam filters can keep pace with the creation tools remains an open question.

The post YouTube pledges to fight AI slop while doubling down on AI creator tools appeared first on The Media Copilot.

Someone paid $26,000 for the CU Independent’s old web domain. Now it’s churning out AI garbage while the real student journalists scramble to afford a legal challenge.

The University of Colorado Boulder’s student newspaper migrated from cuindependent.com to cuindependent.org in 2024. Staff had lost track of who owned the old domain, which dated back to the 2000s. When they stopped paying WordPress to host it, the site went dark.

Then it came back. In July, the domain resurfaced with the Independent’s name and logo, publishing articles like “How Many Albums Does Drake Have?” and “Professional Movers in North Carolina for a Smooth and Secure Move.”

“I looked at it, and obviously was shocked and horrified,” editor in chief Greta Kerkhoff told the Washington Post’s Daniel Wu.

The copycat’s About page claims it honors “what CU Independent stood for: strong voices, independent thought, and stories that matter.” It’s a brazen identity theft that 21-year-old Kerkhoff has spent her senior year fighting instead of just running a newspaper.

“It really feels so weirdly malicious,” she said.

This isn’t an isolated case. NewsGuard has identified more than 2,000 AI-generated news sites as of October, according to the Post. These content farms use generic newspaper names to appear legitimate. But taking over a real publication’s recently active domain is rare.

“This is like their dream,” NewsGuard senior editor Sofia Rubinson told the Post.

Kerkhoff reported the site to the Colorado Bureau of Investigation’s business fraud unit. They told her the state couldn’t investigate unless she proved the copycat was making money. The domain’s owners are hidden behind a proxy service.

Her only path forward is filing a complaint with ICANN, the nonprofit that manages internet addresses. That costs $1,500, forcing the university-unaffiliated paper to launch a fundraiser.

Attorney Alexandra Bass, representing the Independent, said impersonating news sites has become more common since generative AI went mainstream.

“Generative AI can allow bad actors to produce content at a rapid pace — potentially flooding the web with misinformation, and at times directly regurgitating the works of dedicated journalists,” Bass told the Post.

Student Press Law Center attorney Jonathan Gaston-Falk said this could become standard for student and local newsrooms that can’t match big publishers’ legal resources.

“It’s frustrating because I think that a lot of these acts … are premised on the idea that student journalism somehow isn’t protected in the same way as professional media happens to be,” he said.

The copycat site has since changed its logo and removed links to the real Independent’s social media but it has kept the name and continues publishing.

The post AI slop site hijacks student newspaper’s domain, and the fix costs $1,500 appeared first on The Media Copilot.

Merriam-Webster has named “slop” its 2025 Word of the Year, officially enshrining a term that media professionals have been using all year to describe the avalanche of AI-generated junk flooding the internet.

The dictionary defines slop as “digital content of low quality that is produced usually in quantity by means of artificial intelligence.” That covers everything from bizarre AI-generated videos and off-kilter advertising images to fake news, junky AI-written books, and what Merriam-Webster calls “workslop” reports that waste coworkers’ time.

The choice reflects a year when AI output became impossible to ignore. Publications noticed.

“AI Slop is Everywhere,” warned The Wall Street Journal. “AI Slop Has Turned Social Media Into an Antisocial Wasteland,” reported CNET.

For newsrooms, the rise of slop creates a dual challenge. Outlets must now compete for attention against a tide of AI-generated content while also guarding against publishing slop themselves. The word’s selection signals how central this issue has become to public discourse about AI.

The etymology is fitting. “Slop” originally meant “soft mud” in the 1700s, evolved to mean “food waste” in the 1800s, and eventually became shorthand for rubbish.

Merriam-Webster’s editors noted the word carries a defiant tone. “In 2025, amid all the talk about AI threats, slop set a tone that’s less fearful, more mocking,” the dictionary wrote. “The word sends a little message to AI: when it comes to replacing human creativity, sometimes you don’t seem too superintelligent.”

Other words making Merriam-Webster’s 2025 list included “touch grass,” meaning to participate in real-world activities instead of online interactions, and “performative,” describing actions done for show rather than substance.

The selection of slop as the year’s defining word suggests public sentiment around AI may be shifting from wonder to wariness.

The post Merriam-Webster’s 2025 Word of the Year takes direct aim at AI-generated content appeared first on The Media Copilot.