Jump to a brief
Grouped by topic first; AM/PM are editions inside the topic.
Here's your report:
---
🗑️ AI Slop Report — 2026-04-17 06:45
Trending Slop Formats
The slop machine never sleeps, and it's gotten more creative in 2026:
• AI Kids Content — The most disturbing trend. A NYT investigation found ~40% of videos recommended to children on YouTube and YouTube Kids appear to be AI slop — plotless, hypnotic, Cocomelon-style visual sludge. Top channels are pulling in $4.25M/year from this. Congrats to everyone involved.
• "AutoBait" Content Farms — Cybersecurity firm DoubleVerify uncovered a network of 200+ websites churning out templated AI articles and images purely for ad revenue. The operation uses cookie-cutter LLM prompts at industrial scale.
• AI Phishing Tsunami — Phishing attacks are up 204%, with enterprises now receiving a malicious email every 19 seconds. AI has eliminated the typos and broken English that used to be red flags. The scammers finally learned grammar.
• "Slopaganda" — Both the US and Iran are flooding social media with AI-generated political noise, mixing real footage with game clips and AI imagery. State-sponsored slop is now a genre.
• Academic Slop — Over 11,000 academic papers were retracted in 2025 alone, many AI-generated and rubber-stamped through peer review. Science itself is getting slopped.
Platform Watch
• YouTube — Ground zero. Research estimates 21-33% of YouTube's feed is AI slop, generating ~$117M/year in ad revenue. YouTube terminated major offenders like Screen Culture and KH Studio, but the hydra has many heads.
• Google Search — AI-generated articles now make up over half of English-language web content. Google rolled out both a February 2026 Discover Core Update and a March 2026 Spam Update specifically targeting slop farms.
• The Open Web — Some estimates put AI-generated content at up to 90% of all new web content. The internet is now mostly talking to itself.
Notable Incidents
• 200+ Organizations vs YouTube — In April, over 200 child advocacy groups sent a letter demanding YouTube ban AI slop from its kids platform. The 40% stat from the NYT investigation was the breaking point.
• Svedka's Super Bowl Disaster — During Super Bowl LX, Svedka vodka aired an AI-generated ad featuring robots dancing at a club. Viewers hated it. Turns out people don't want their annual cultural event slopped.
• Satya Nadella's "Please Stop" Tour — In a move that absolutely reeks of self-awareness, Microsoft's CEO wants you to stop using the word "slop" to describe AI content. Bold strategy from the company that Bing-ified the problem.
• Orwell Called It — Open Culture published a piece connecting AI slop to Orwell's "prolefeed" — mass-produced entertainment designed to pacify. The comparison is uncomfortably accurate.
The Grift Corner
The "make money with AI" content ecosystem is now a self-sustaining slop ouroboros:
• Every major platform — Coursera, Shopify, Printful, random SEO blogs — is publishing "How to Make Money with AI in 2026" guides. Many of these guides are themselves AI-generated. It's slop all the way down.
• AI course generators (Coursebox, Course AI, Mini Course Generator) let you create entire courses about AI — using AI — to sell to people who want to use AI to make money. The grift has achieved recursion.
• The "Grift Economy" is now a recognized term: AI-generated fake testimonials, fake reviews, and fake case studies — all for the cost of a monthly subscription. The tools of extraction have been democratized.
Silver Lining
Not all is lost — the immune response is kicking in:
• Kagi's SlopStop — Kagi Search now lets users flag AI slop, downranks slop domains, and is building the largest community-driven dataset of AI-slop sites. Finally, a search engine that treats slop like the disease it is.
• Google's Ad Firewall — Google blocked 8.3 billion policy-violating ads and suspended 24.9 million advertiser accounts in the past year, 4 million for scam-related activity. Gemini catches 99%+ before they reach users.
• "Your AI Slop Bores Me" — A viral interactive web game that went huge on Hacker News in March 2026. The cultural antibodies are forming.
• SLOP SCAN Hackathon — A 72-hour hackathon (May 29–Jun 1) offering prizes to build AI slop detection tools. Fighting fire with code.
• AntiSlop.online — Publishing annual "State of Slop" reports tracking detection accuracy and egregious cases. Someone's keeping score.
The word "slop" was named Word of the Year 2025 by both Merriam-Webster and the Australian National Dictionary. The problem now has a name everyone agrees on. That's step one.