Copyright class actions could financially ruin AI industry, trade groups say.

Copyright class actions could financially ruin AI industry, trade groups say.

AI industry groups are urging an appeals court to block what they say is the largest copyright class action ever certified. They’ve warned that a single lawsuit raised by three authors over Anthropic’s AI training now threatens to “financially ruin” the entire AI industry if up to 7 million claimants end up joining the litigation and forcing a settlement.

Last week, Anthropic petitioned to appeal the class certification, urging the court to weigh questions that the district court judge, William Alsup, seemingly did not. Alsup allegedly failed to conduct a “rigorous analysis” of the potential class and instead based his judgment on his “50 years” of experience, Anthropic said.

  • @[email protected]
    15•1 day ago

    I’ve read this entire thread and could not find a single person who seems to have actually read anything about this case.

    The article is a huge pile of bullshit.

    Here is what happened: An industry group filed an amicus brief during the appeal of a ruling where the judge certified the 3 plaintiffs as a class. Boring legal minutae in a case that doesn’t matter, see below.

    The author is either incompetent at understanding legal filings or deliberately being misleading to write clickbait trash. Human slop, if you prefer.

    This is not noteworthy, at all. The issue being argued about is if the 3 people can represent the class of “everyone Anthropic downloaded books from”. This is a non-story, unless you’re a legal nerd and care about exactly how courts define classes and the legal steps required for the analysis.

    But, more importantly for the frothing anti-AI masses:

    In the order certifying the defendants as a class, the judge dismissed the plaintiff’s claims of copyright violation related to the training of LLMs. The judge said that training LLMs was transformative and thus fair use under copyright law and since this is so obvious that that argument could be summarily dismissed.

    Don’t believe me, go click on the links in the article to the summary judgement yourself. The information is not hard to find if you read farther than the headline.

    The only remaining issue in the lawsuit is if Anthropic is civilly liable for downloading the books on bittorrent.

    This case isn’t even about AI anymore, it’s the same kind of lawsuit that we’ve seen since Napster was popular. Uploading copyrighted material, like when you use BitTorrent, is a copyright violation and you could be sued.

    That’s all this case is now, the argument that everyone is fighting over in the comments: “Is training an LLM on copyrighted material a violation of copyright?” is already answered by the judge:

    No, using copyrighted material to train a LLM is so obviously fair use that the argument was summarily dismissed.

    Here’s the relevant quote from the judge, in summary judgement:

    To summarize the analysis that now follows, the use of the books at issue to train Claude and its precursors was exceedingly transformative and was a fair use under Section 107 of the Copyright Act. The digitization of the books purchased in print form by Anthropic was also a fair use, but not for the same reason as applies to the training copies. Instead, it was a fair use because all Anthropic did was replace the print copies it had purchased for its central library with more convenient, space-saving, and searchable digital copies without adding new copies, creating new works, or redistributing existing copies. However, Anthropic had no entitlement to use pirated copies for its central library, and creating a permanent, general-purpose library was not itself a fair use excusing Anthropic’s piracy.