To feed its artificial intelligence, does Apple have the right to plunder YouTubers’ videos for free? That is the entire issue at the centre of an explosive legal fight.
Artificial intelligence has an insatiable appetite and, to grow, it must ingest vast quantities of data. But where do Big Tech companies get this raw material? For many content creators, the answer is straightforward: they take it. In April 2026, a group of influential YouTube channels decided to hit back in court.
Why YouTube data has become a battleground for AI
The claimants say the generative AI industry-now worth trillions of dollars-has been fattening itself at their expense, without paying them a penny. They describe it as an unacceptable assault on their community.
They also argue the problem extends far beyond one company. The same YouTubers have filed similar actions against other major players in the sector: Meta, NVIDIA, Snap and ByteDance, TikTok’s parent company.
The class action targeting Apple’s AI training
The proposed class action is being led by Ted Entertainment, owner of the well-known channel h3h3Productions and its podcast with millions of subscribers. The lawsuit takes direct aim at Apple.
According to the filing, Apple deliberately bypassed YouTube’s protective barriers in order to siphon off millions of videos without any permission, so it could train its own AI models.
Apple’s defence
Apple moved quickly to respond. Three months after the case was filed, the company formally asked a US court to throw out the claims, arguing that the creators have no legal basis to sue. Apple’s position is blunt: anything that can be accessed freely online can be used.
The company points to how Google’s platform works. “The plaintiffs claim to have uploaded audiovisual works to YouTube, and that any member of the public can watch them on that platform,” Apple counters.
“No password. No payment. No lock. No key. YouTube would, according to the allegations, use technical measures to prevent unauthorised downloading. But because YouTube provides public access to these videos, the alleged technical measures do not control access to the works, as required by Section 1201(a),” it continues.
In practical terms, because the content remains public by nature, Apple argues these protections are not enough to make extraction illegal: in its view, it merely accessed data that had been left openly available.
A ruling that could shape the web’s future
The dispute lands in the middle of a huge legal grey area shaking the tech world. Copyright law was never designed with AI systems in mind-systems capable of swallowing millions of videos in seconds to build new commercial services.
The decision from the federal court in California will therefore be watched extremely closely across the industry. If the judge sides with Apple and dismisses the case, it would send a strong message to Silicon Valley: any content available on the web without a paywall could be legally siphoned off to train AI.
If, on the other hand, the creators win, Big Tech companies could be forced to negotiate and pay licensing fees to authors, as some already do.
Comments
No comments yet. Be the first to comment!
Leave a Comment