-: FOLLOW US :- @theinsaneapp
Anthropic, a generative AI startup, responded to accusations of copyright infringement by music publishers Concord, Universal, and ABKCO in a court filing.
-: FOLLOW US :- @theinsaneapp
The publishers filed a lawsuit alleging that Anthropic's chatbot Claude, now replaced by Claude 2, unlawfully scraped song lyrics from the internet to train its AI models, reproducing copyrighted lyrics in chatbot responses.
-: FOLLOW US :- @theinsaneapp
Anthropic argues that its use of lyrics for training is transformative, adding a different purpose to the original works and having no substantially adverse impact on the market for copyrighted works.
-: FOLLOW US :- @theinsaneapp
The company claims licensing the vast amount of text needed for training is technically and financially unfeasible, joining OpenAI in this argument.
-: FOLLOW US :- @theinsaneapp
Anthropic introduces a novel argument suggesting that the plaintiffs, not Anthropic, engaged in the "volitional conduct" required for direct infringement liability, stating that the label plaintiffs caused its AI model to produce the infringing content.
-: FOLLOW US :- @theinsaneapp
Anthropic contests irreparable harm, pointing to a lack of evidence that licensing revenues decreased since Claude's launch and arguing that monetary damages could make the publishers whole.
-: FOLLOW US :- @theinsaneapp
The company implemented safeguards in Claude to prevent further display of copyrighted lyrics and argues that the request for an injunction is unjustified.
-: FOLLOW US :- @theinsaneapp
Anthropic challenges the lawsuit venue, stating it has no relevant business connections to Tennessee where the lawsuit was filed, and users agreed to litigate disputes in California courts.
-: FOLLOW US :- @theinsaneapp
The filing highlights the broader copyright battles in the generative AI industry, with more artists joining lawsuits and the emergence of a nonprofit group advocating for a "licensed model" certification for data used to train AI models.
-: FOLLOW US :- @theinsaneapp
Generative AI companies seem to be coalescing around fair use and harm-based defenses, setting the stage for ongoing legal battles with copyright plaintiffs.