21% of YouTube’s Content Is Now AI Slop as Brainrot Rises
A new study says more than one in five videos shown to brand-new YouTube users are not made by people at all, but by AI. Even worse, many of them are low-quality, repetitive clips designed mainly to grab views and make money.
What the study found
Video-editing company Kapwing created a brand-new YouTube account and tracked what appeared in the Shorts feed. Out of the first 500 videos:
- 104 were AI-generated “slop”
- 165 were labeled “brainrot,” meaning low-effort, attention-farming content
- A full 21% of the feed was AI slop
To understand how big the problem really is, Kapwing also looked at 15,000 of the world’s top YouTube channels. The findings were worse than expected.
- 278 top channels post only AI slop
- 63+ billion views across these channels
- 221 million total subscribers
- About $117 million in yearly earnings
This is not a niche corner of the internet. It is mainstream digital entertainment.
Where people watch it the most
AI slop is spreading globally. Some countries see especially high engagement:
- Spain: ~20 million subscribers
- Egypt: ~18 million subscribers
- United States: ~14–14.5 million subscribers
When it comes to pure views, the ranking shifts:
- South Korea: ~8.45 billion views
- Pakistan: ~5.3 billion views
- United States: ~3.4 billion views
Money plays a huge role in why this content keeps growing. India’s Bandar Apna Dost has more than 2 billion views and may be earning over $4 million a year. A Spanish-language U.S.-based channel, Cuentos Fascinantes, has nearly 6 million subscribers and more than a billion views.
Why AI slop works
These videos are built to hook you quickly. They often use:
- Fast-moving visuals
- Simple or absurd plots
- Repetitive characters and sounds
- Content that requires zero thinking or context
Many target kids, who are easy to keep locked onto screens with bright colors and constant movement. Others use shock, drama, or emotional baiting.
Researchers say the real force behind all of this is not creativity. It is the algorithm. Platforms reward content that keeps people watching longer, not necessarily content that is meaningful or informative. AI makes it cheap and easy to flood YouTube with endless clips, so creators chase volume and profit.
What YouTube says
YouTube says AI is simply another creative tool and insists that all videos must follow platform rules. The company claims it focuses on connecting people with quality content “no matter how it is made.”
Back in July, YouTube tried to get ahead of this problem. On July 15, the company updated its YouTube Partner Program (YPP) monetization policies to more clearly define what counts as “inauthentic” or mass-produced content. The goal was simple: stop creators from making money off repetitive, spam-style videos that AI tools have made incredibly easy to produce. At the time, YouTube framed the change as nothing new, just a clarification of its long-standing rules about originality. Creator Liaison Rene Ritchie even reassured people that it was only meant to better identify mass-produced or recycled content, not to punish reaction videos or genuine commentary. YouTube also stressed that low-effort, repetitive content was technically already ineligible for monetization.
Yet despite that policy shift, AI slop has only continued to spread. Instead of shrinking, it has grown into something almost unavoidable, making the earlier policy update feel less like a solution and more like a quiet acknowledgment of a problem the platform no longer seems able to control.
For more articles like this, check out our Tech News page!