Sign up to see the future, today
Can’t-miss innovations from the bleeding edge of science and tech
AI slop has become practically inescapable on social media, with low-rent and at-times terrifying AI-generated videos drowning out human creators across platforms.
Even the vertical video darling TikTok has fallen victim. According to a new report by San Francisco-based video editing company Kapwing, almost 60 percent of videos being served to new users on the app’s algorithmic “For You” page is now AI slop — three times as much as is served to new YouTube users.
It’s particularly worrisome for the platform’s youngest users: Kapwing found that the “category with the highest slop density by far was Kids.” The hashtag #cartoonkids was “almost entirely made up of slop, with only three of the 100 videos we checked being human-made,” according to the company.
Worse yet, once an account indicates to the algorithm it’s interested in AI, the feed quickly doubles down, serving even more slop.
The findings are a pertinent reminder of just how big of a problem AI slop has become. Young, impressionable minds are being exposed to a slurry of half-baked and brain-melting AI material that could endanger their brain development, experts warn. Meanwhile, photorealistic deepfakes are facilitating the spread of misinformation and political propaganda.
TikTok is far from the only platform being consumed. Meta’s Facebook and Instagram have also turned into a largely unrecognizable wasteland, with users — and likely bots — interacting with absurd images of impoverished children begging on the street without arms, or videos of humanoid cats unintentionally killing their offspring in a meat grinder.
In an effort to get a hold of the problem, TikTok announced in November that it would allow users to either dial up or down the amount of AI-generated content on their feeds.
“We know from our community that many people enjoy content made with AI tools, from digital art to science explainers, and we want to give people the power to see more or less of that, based on their own preferences,” TikTok’s European director of public policy for safety and privacy Jade Nester told The Guardian at the time.
YouTube also recently announced that it was making changes to how it would start labeling AI-generated content in a broader effort to tamp down on AI slop, but shopped short of changing “how a video is recommended or whether it’s eligible to earn money.”
It’s not clear what an easy fix would be. The tech has progressed to the point where discerning between reality and slop has become increasingly difficult; as the New York Times reported over the weekend, even Hany Farid, the “world’s leading deepfake expert,” has “stopped trusting his own eyes.”
More on AI slop: YouTube Announces Plans to Crack Down on AI Slop
