Over one in five videos shown to new YouTube users are 'AI slop': Researchers
The study found that more than 20% of videos suggested to first-time users were low-quality, AI-generated clips designed primarily to attract views
A growing share of content recommended to new YouTube users consists of what researchers describe as "AI slop," according to a recent study examining the platform's recommendation systems.
The study found that more than 20% of videos suggested to first-time users were low-quality, AI-generated clips designed primarily to attract views. Researchers estimate the fast-expanding sector generates about $117 million in annual revenue across major social media platforms, says the Guardian.
The analysis identified 278 of the world's most popular YouTube channels as being entirely composed of this type of content. Together, those channels have accumulated more than 63 billion views and 221 million subscribers.
Keep updated, follow The Business Standard's Google news channel
Researchers said the channels have large audiences across multiple regions, including Spain, Egypt, the United States and Brazil. Many of the videos appear to target children through bright animations, while others rely on surreal or plotless scenes featuring anthropomorphic animals or superheroes.
Some creators were found to use artificial intelligence to produce what the study described as "uncanny" depictions of natural disasters, including large-scale flooding, often paired with calming background music. The researchers said such videos are designed to maximize watch time and advertising revenue.
The study noted that much of the industry is driven by creators based in middle-income countries such as India, Nigeria and Kenya, where earnings from YouTube can exceed local median wages.
YouTube has said its recommendation systems aim to connect users with high-quality content, regardless of whether it is created by humans or generated using AI tools. Critics, however, argue that large platforms operate as "A/B testing machines," prioritizing content that performs well with algorithms rather than rewarding originality or craftsmanship.
The researchers compared the phenomenon to "digital junk food," describing it as mass-produced content created using inexpensive, automated methods to trigger algorithmic engagement while offering limited artistic value.
