Three YouTubers accuse Apple of illegal scraping to train its AI models
apple openai
| Source: Mastodon | Original article
Three prominent YouTubers have filed a class‑action lawsuit accusing Apple of violating the Digital Millennium Copyright Act by scraping their videos without permission to train the language models behind Apple Intelligence. The complaint, lodged in a U.S. federal court on Tuesday, alleges that Apple’s data‑collection system harvested full‑length videos, transcripts and metadata from the creators’ channels, then used the material to improve the conversational abilities of its on‑device AI assistant. The plaintiffs claim the practice amounts to “systematic, large‑scale infringement” and seek statutory damages, an injunction against further scraping, and a court‑ordered audit of Apple’s training pipelines.
Apple responded through its press office, reiterating a statement to AppleInsider that the company “does not use YouTube video content without proper licensing” and that Apple Intelligence was built on publicly available data that respects creators’ rights. The firm has not disclosed the specific datasets feeding its models, a common opacity that has drawn scrutiny from regulators and competitors alike.
The lawsuit matters because it adds to a growing wave of copyright actions targeting AI developers. Recent cases against OpenAI and Google have forced courts to confront whether training on copyrighted works constitutes fair use, and whether existing DMCA exemptions apply to large‑scale machine‑learning. For Apple, the dispute could delay rollout of its AI features across iOS, macOS and the newly reopened Barcelona Store, and may pressure the company to negotiate licensing deals with content creators.
What to watch next: Apple’s formal answer, expected within 21 days, will reveal whether it will contest the claims or seek a settlement. Parallel proceedings in Europe under the Digital Services Act could amplify the issue, while the plaintiffs plan to request a preliminary injunction that could halt any further data harvesting pending trial. The outcome could set a precedent for how tech giants source training data in the Nordic market and beyond.
Sources
Back to AIPULSEN