A Digital Cleanup
Spotify removed 75 million bulk uploads, duplicate songs, and other "spammy" tracks last year, according to the company. LinkedIn said in July that "AI slop is a top priority for all of us" and introduced a button for users to report it. Google researchers described how a video service—widely understood to be YouTube—wiped 130,000 channels of low-quality content in six months. These actions, reported by Digital Trends and The Indian Express, are part of a broader industry effort to manage what many call an overabundance of low-grade, AI-generated material.
The scale of the problem is significant. Deezer, a music streaming service, said that at its June peak it received around 90,000 fully AI-generated tracks per day, accounting for more than half of all new uploads. Those tracks, however, represented only 1% to 3% of listening. Deezer excludes detected AI music from its recommendations and editorial playlists. A separate figure, reported by The Indian Express, suggests that AI generates more than a third of what is uploaded to Apple Music, though Apple has not confirmed this.
Aidan Walker, a content creator and researcher at the Carnegie Endowment for International Peace, called AI slop a "festering wound," noting that users have become increasingly upset. "The climate of opinion has now recognizably shifted, so these companies are finally confronting these problems," he said.
Platform Responses
YouTube's CEO, Neal Mohan, wrote in a blog post that "managing AI slop" was a top priority this year. Google researchers found and removed 50,000 coordinated clusters containing 130,000 channels for abusive behavior. YouTube responded to external studies by saying that academic research does not reflect its many layers of defense.
Pinterest updated its system to allow users to limit AI-generated posts, while TikTok tested a feature that lets users dial up or down AI content and automatically labeled more than 3 billion AI-generated videos. TikTok also removed more than 377,000 videos in the first three months of the year for violating its AI policies.
X (formerly Twitter) has taken steps as well. According to Nikita Bier, X banned apps that pay for content to encourage bots. The AI detection startup Pangram estimated that AI was used to create nearly half of all posts with more than 50 words on the platform. Substack introduced an AI detection tool powered by Pangram, though some writers complained about false positives.
Meta, Instagram's parent company, encountered its own backlash when it rolled out an AI image generator that automatically opted in public Instagram accounts; the feature was removed after three days. Adam Mosseri, head of Instagram, said it would be more practical to trace real media than fake media, a view that contrasts with the labeling approach adopted by TikTok.
The Ongoing AI Embrace
Despite the cleanup, tech companies are not abandoning AI. Spotify, for example, is working with major music companies on generative AI products, including licensed covers and remixes. Google has linked its Gemini model to email, maps, travel services, and search. This dual approach—removing slop while expanding AI—has led some observers to describe the effort as a "cleanup operation" rather than a rejection of AI.
According to a Kapwing experiment, more than 20% of initial short videos served to new YouTube users are AI-generated slop. One popular YouTube slop channel earned $4.25 million annually before being suspended from monetization. These examples illustrate the economic incentives that sustain slop production.
Camille François, a researcher, and nearly two dozen AI experts convened at a Slop Salon event in Menlo Park. François said the hodgepodge of tactics is a symptom of a systemic problem with the information environment. Merriam-Webster chose "slop" as its word of the year in 2025, reflecting its cultural prominence.
European researchers have warned that unchecked synthetic content could "crowd out human creativity, corrode cultural value and destabilize public discourse." As platforms continue to refine their moderation, the challenge remains balancing innovation with the need to preserve meaningful content.