A growing body of AI-generated children's content -- researchers are calling it "AI slop" -- has flooded platforms like YouTube, and some of what's showing up isn't just low-quality. It's actively dangerous.
What's Actually In These Videos
Reporting on the phenomenon documented specific, concrete examples. One video, "Vroom Vroom! Car Ride Song," had racked up nearly 50,000 views while teaching incorrect traffic safety -- including the claim that "green means right" -- and depicting children unbelted in moving vehicles, floating beside cars, and walking in roadways. Another video showed an infant eating whole grapes and honey, both genuine choking and health hazards for babies. Educational videos meant to teach vowels displayed consonants instead; a "50 States Song" mangled state names into nonsense like "Louggisslia" and "Oklolodia"; and other videos have shown children frightened by AI-generated dinosaurs chasing them.
This isn't a handful of bad actors. One channel, "Jo Jo Funland," posted more than 10,000 videos in seven months -- an average of 50 videos a day. For comparison, Sesame Street produced roughly 3,900 episodes across 20 years.
How Much of This Is Out There
A November 2025 Kapwing report found that roughly 21% of YouTube's overall feed now consists of low-quality AI-generated video. For children's content specifically, that means a meaningful share of what autoplay serves up to a toddler could be unvetted, AI-generated, and factually wrong -- or worse, actively modeling unsafe behavior.
What Researchers Are Saying
Dr. Kathy Hirsh-Pasek, a professor of psychology and neuroscience at Temple University, put it bluntly: "We're at the beginning of a monster problem, and we have to get hold of it quickly." Dr. Dana Suskind, a professor of surgery and pediatrics at the University of Chicago, went further, describing what's happening as "toddler AI misinformation at industrial scale" and coining a specific term for the risk to very young viewers: not "brainrot," but "brain stunt" -- because a young child's brain is still actively building neural connections, and this kind of content can disrupt that development in a way that's different from screen-time effects on an older child or adult.
What Happened After the Report
Following the initial reporting, YouTube terminated six channels for violating the platform's terms of service, and a seventh for violating its spam policy. It's a start, but the scale of the problem -- one channel alone posting 50 videos a day -- suggests platform moderation is playing catch-up against content that can be generated far faster than it can be reviewed.
Why This Matters for Families
Autoplay and algorithmic recommendation don't distinguish between a well-produced, age-appropriate video and an AI-generated one with wrong facts or unsafe modeling baked in -- both can look equally polished at a glance. For parents of young children especially, this is a reminder that "made for kids" and "reviewed by a human who understands child development" are no longer the same guarantee they used to be. Checking who actually made a channel's content, not just whether it looks bright and colorful, matters more than ever.