A growing wave of low-quality, AI-generated videos is taking over YouTube Shorts. A new study suggests that for new users, more than half of the platform’s short-video recommendations now fall into the category of AI “slop” or so-called brainrot content.
The findings come from research by video-editing company Kapwing, first reported by The Guardian.
According to the study, more than one in five videos recommended to new YouTube Shorts users are low-quality, AI-generated clips, commonly referred to as AI slop.
What data reveals about shorts recommendations
Kapwing analysed the first 500 videos shown to a brand-new, untouched YouTube Shorts account.
Of these, 104 videos (21%) were AI-generated, while 165 videos (33%) were labelled as brainrot, meaning nonsensical and low-quality content. Combined, this accounts for 54% of total recommendations.
What Is Brainrot Content?
Kapwing defines brainrot as “nonsensical, low-quality video content that creates the effect of corroding the viewer’s mental or intellectual state.”
The study notes that brainrot content is often AI-generated, blurring the line between human-made low-effort videos and automated production.
AI slop varies by country
The popularity of AI slop differs across regions.
Kapwing found that Spain-based AI slop channels have a combined 20.22 million subscribers, the highest total globally, despite having fewer such channels among the country’s top 100.
The United States ranks third, with 14.47 million subscribers, and has nine AI slop channels among its top 100 most popular channels.
YouTube is not alone in facing an influx of AI slop content.
The study aligns with broader observations that AI-generated, low-effort videos are increasingly flooding social media feeds, from fake animal surveillance clips to bizarre, clearly fabricated viral videos.
AI slop is not going away
Kapwing’s research suggests the trend is unlikely to reverse soon.
As reported earlier this month, AI slop continues to spread rapidly across platforms, driven by easy content generation and algorithms that reward frequent uploads and engagement over quality.
