You’ve probably noticed the internet’s getting slower. Not your Wi‑Fi — the web itself. The pages that used to load in a blink now take five seconds. The obscure music service you love just turned off its radio feature. And the reason? Two armies you never expected to fight: BTS superfans and greedy AI scrapers.
This isn’t a Netflix drama. It’s happening right now to ListenBrainz, a tiny open‑source project run by volunteers who just wanted to build a better music database. In the past month, ListenBrainz went from 50 listening submissions per second to over 500 — because the entire BTS stan army migrated from Last.fm in a single, coordinated exodus. Then, just as the servers started to scream, the AI scrapers showed up. Companies hungry for training data began hammering the same endpoints. The result? ListenBrainz had to kill its radio feature and popularity endpoints just to stay alive.
Open‑source projects are the unpaid janitors of the internet. And we just handed them a mop and asked them to clean up after a riot and a machine‑gun fight.
Let’s be clear: the BTS fans didn’t do anything wrong. They’re passionate, they love their music, and they needed a new home after Last.fm’s decline. But when a community of that size moves overnight, it doesn’t just knock on the door — it breaks the hinges. And the AI scrapers? They don’t even bother knocking. They treat every public API as an all‑you‑can‑eat buffet, with no regard for the humans keeping the buffet stocked.
This is the hidden cost of the AI gold rush. Every time a startup brags about “training on the open web,” they’re really saying: we extracted value from someone else’s unpaid labor and didn’t pay a dime. Combine that with a sudden, organic migration of a massive fanbase, and you get a perfect storm that no donation‑funded server can survive.
When the BTS army moved from Last.fm to ListenBrainz, they brought passion. When the AI scrapers followed, they brought destruction. The platform never stood a chance.
And here’s the twist: most people will never hear about this. ListenBrainz is not Wikipedia or Reddit. It’s a niche service for music obsessives. But the pattern is universal. Small, independent projects are the scaffolding of the open web — the forums, the databases, the tools that big companies rely on but never fund. Every time a fandom migrates or an AI crawler gets aggressive, another piece of that scaffolding cracks.
I’m not saying we should ban BTS fans or halt AI research. I’m saying we need to acknowledge that community‑funded open‑source projects are acting as free, unconsenting infrastructure for both massive fandom migrations and the AI data gold rush — receiving all the costs and none of the support.
Meta and OpenAI would never dream of collapsing Wikipedia — they’d be sued. But scraping a small open‑source project? Free game. The legal system hasn’t caught up, and the tech giants are more than happy to exploit the gap.
So what do we do? First, if you run a project, put rate limits on everything. Second, if you’re an AI company, stop acting like public APIs are a charity buffet. And third, if you care about the independent web, donate to the projects you love. Not with a tweet — with actual money.
Otherwise, the open web will not die with a bang. It will be silently suffocated by the very users and machines that claim to love it.
FAQ
Q: Isn't this just a scalability problem? Can't they just add more servers?
A: Scaling costs money. ListenBrainz is run on donations and volunteer time. Adding servers isn't free — and even if they could afford it, the root cause is that two massive, uncoordinated demand spikes hit a system never designed for either. Throwing hardware at a structural abuse problem just kicks the can down the road.
Q: What's the practical implication for me as a regular internet user?
A: The indie services you rely on — niche forums, personal blogs, open-source alternatives to big tech — are dying under the same pressure. If you want them to survive, you need to support them financially or advocate for legal protections against unlimited scraping. Otherwise, you'll end up with only the corporate platforms left.
Q: Aren't the BTS fans the real problem? Shouldn't they be blamed for overwhelming the service?
A: Blame the fans is a convenient scapegoat, but it misses the point. The fans did what any community does: they moved to a better platform. The real culprits are the AI scrapers that added a second, automated load on top of the organic one — and the systemic failure to fund and protect open-source infrastructure. Punishing users won't fix a broken model.