The internet’s AI revolution has come with an entirely predictable side effect — an explosion of ‘AI slop’. From uncanny videos and fake celebrity clips to AI-generated music and recycled social posts, internet users are getting frustrated by the cheap, repetitive or bizarrely synthetic content produced at enormous scale.
Now, some of the biggest technology companies are scrambling to clean up a mess they helped create.
Recently, music streamer Spotify announced that it removed 75 million bulk uploads, duplicate songs and other tracks last year, describing them as ‘spammy’ — a sizeable chunk of its total portfolio. In July, LinkedIn developed a button for users to report slop content on the platform. YouTube also wiped 130,000 channels of low-quality content off its platform over the past six months.
Since July last year, YouTube has explicitly said mass-produced, repetitive and generic videos are not eligible for monetisation, reiterated this warning in 2026 as well. Examples include templated videos, slideshows with minimal commentary and AI-generated material that visibly repeats a format without adding any fresh or original insight.
The challenge lies in the fact that AI slop can be easily monetised through various avenues. By making thousands of cheap, low-effort, formula-based videos, creators can chase advertising revenue without investing much time or creativity. YouTube is, therefore, trying to attack the economics of slop rather than banning AI itself.
The AI slop wave started off with viral memes like ‘shrimp Jesus’. Among the slop content now most commonly found are the AI-generated celebrity impersonations that are becoming harder to distinguish from authentic footage. Artists often talk about their likenesses being used in images and videos without their consent in public interviews. Another pertinent example is from March — Fruit Love Island, an AI-generated parody featuring anthropomorphic fruits in a reality-TV-style dating show, became a viral TikTok phenomenon, drawing millions of views and also sparking debate over whether such content should be monetised. Several other examples litter the internet space.
Short-form video making and sharing social media site TikTok, which has been banned in India, takes a different approach. The platform already requires creators to label realistic AI-generated images, audio and video. TikTok has also been testing improved detection systems aimed specifically at accounts posting AI-generated spam that crowds out original creators. It is strengthening provenance (content tracking) technology through its participation in the C2PA (Coalition for Content Provenance and Authenticity, an open technical standards group founded in February 2021 by Adobe, Arm, BBC, Intel, Microsoft, and Truepic) standards group, which works on establishing where digital content comes from.
Spotify is tackling the musical version of slop. Earlier this month, Spotify also announced the impending launch of an AI tag or label that will identify AI-generated music and allow such tracks to be excluded from recommendations, giving listeners more control over what reaches their feeds.
LinkedIn’s reporting option for suspected AI slop, reflects a broader reality on social media – automated detection alone is unlikely to catch everything. LinkedIn executives have described AI slop as a major priority as synthetic posts compete with human-written professional content.
Snapchat also decided in July that fully AI-generated videos would no longer qualify for rewards on its Spotlight platform, signalling a preference for human-made work even while Snapchat continues investing heavily in its own AI creation tools.
AI has become a boon and a bane for big tech companies — they cannot live with it, and they cannot exactly live without it either. The formula for treatment of AI slop lands on labelling it, detecting it, and removing it or blocking the monetisation avenue for it. The contention is less about eliminating AI content than deciding which synthetic content deserves to be seen — and which should simply disappear into the obscurity of the algorithms.