A US federal judge has approved a class action lawsuit against AI firm Anthropic, brought by authors alleging copyright infringement. The lawsuit centres on claims that Anthropic used copyrighted books, obtained from illicit sources, to train its AI models. The plaintiffs, a group of authors, can now represent a nationwide class of writers who claim their works were used without permission. The case alleges that Anthropic knowingly downloaded copyrighted material to develop its Claude model, impacting authors' earnings and rights. This ruling allows a potentially large group of authors to collectively pursue damages and injunctive relief against Anthropic for the alleged copyright violations.
The lawsuit highlights concerns about AI companies using copyrighted material for training purposes without proper licensing or compensation to creators. The authors seek damages for copyright infringement and an injunction to prevent further unauthorised use of their works. This case is part of a growing trend of copyright litigation against AI developers, testing the boundaries of fair use and the rights of content creators in the age of artificial intelligence.
Related Articles
Amazon Deepens AI Alliance
Read more about Amazon Deepens AI Alliance →AI Models' Reasoning Transparency
Read more about AI Models' Reasoning Transparency →AI Faces Copyright Scrutiny
Read more about AI Faces Copyright Scrutiny →AI Chatbots' Sycophancy Problem
Read more about AI Chatbots' Sycophancy Problem →