The good old days for AI content creators might be coming to an end.

Two pieces of news that dropped almost simultaneously point in the same direction. First: YouTube updated its monetization policies, clarifying which AI-generated low-quality content (AI slop) can't earn ad revenue — videos mass-produced with AI that have no substantive content, created purely to game traffic, can no longer make money through the YouTube Partner Program. Second: Sony Music Entertainment filed a lawsuit against AI music tool Udio, alleging it infringed the copyright of over 30,000 songs — from Elvis Presley's "Hound Dog" to Beyoncé's "Say My Name" to Harry Styles' "As It Was," spanning decades.

Platforms cleaning up garbage content on one side, copyright holders collecting royalties on the other. The wild growth era of AI content creation is ending at a pace visible to the naked eye.

Why YouTube Is Cracking Down on AI Slop

YouTube's AI slop policy isn't exactly new, but this update is more specific and more actionable.

Previous policies were vague — "we don't encourage low-quality content" — but what counts as low quality, how it's determined, what the consequences are, all remained unclear. This update spells it out: if you use AI to mass-produce videos with no substantive original value, just gaming keywords and traffic to earn ad revenue, your channel will lose monetization eligibility, and in severe cases get terminated.

Why does YouTube care? Because AI slop is threatening its core assets: user experience and advertiser trust. Open YouTube, search for a question, and if the top 10 results are all AI-generated fluff, users will search elsewhere next time. Advertisers finding their ads running in front of AI garbage videos will see brand damage and move their budgets elsewhere. YouTube's business model is built on the flywheel of "high-quality content attracts users, users attract advertisers" — AI slop is sand in the gears.

This policy won't much affect creators who genuinely use AI for creativity — if you use AI as a tool but your content has your own perspective and value, you're fine. The ones really affected are the "content farms" churning out AI garbage in bulk — the ones generating hundreds of videos a day with AI, flooding long-tail keywords, making money through ad networks. That model used to be a gray area. Now it's explicitly against the rules.

Why Sony Has Udio in Its Sights

Sony suing Udio also isn't an isolated incident. It's another battlefield in the broader AI copyright war.

What does 30,000 songs mean? Sony's music catalog is one of the most comprehensive in the world, spanning classic oldies to the latest hits. It chose 30,000 as representative works for the suit — not that only 30,000 were infringed, but that 30,000 were chosen as samples. The actual number used in training could be hundreds of thousands or even millions. If the court finds infringement, damages could be astronomical.

Sony's strategy is clear: hit the most prominent one first, kill the chicken to scare the monkeys. Udio and Suno are currently the two hottest AI music generation tools, with large user bases and high visibility. Sue one first, win, then go after the rest. It's the same playbook record labels used back in the day — hit Napster first, then take on the whole P2P industry.

But unlike the Napster era, the legal status of AI music tools isn't settled yet. The Anthropic copyright case, while establishing a trend toward "training might be fair use," was about text and in the Ninth Circuit. Music has its特殊性 — melody, rhythm, lyrics — more dimensions of copyright protection. Does an AI-generated song that "sounds similar" to an original count as infringement? Judges are still figuring that out.

Putting YouTube's policy and Sony's lawsuit together, a clear trend emerges: the AI content industry is shifting from 'more is better' to 'better is better.'

Over the past year, the narrative around AI content was "100x productivity gains," "one person is a whole team" — as if more content was always better content. But now platforms are doing quality control, and copyright holders are collecting fees. The quantity dividend is eaten up. Next comes competition on quality, creativity, differentiation.

For real creators, this is actually good news. AI lowered the barriers to creation, but it also made mediocrity cheap. When garbage content gets cleaned up and copyright costs get internalized, content with real ideas, style, and depth will actually be worth more. AI won't replace good creators — it will make good creators even more powerful, and make mediocre creators nowhere to hide.

See you tomorrow.