US Judge Rules in Favor of AI Company Anthropic in Copyright Dispute

Anthropic, a company funded by Amazon and Alphabet, could face potential damages of up to $150,000 for each copyrighted work due to holding over seven million pirated books in a central library, as ruled by a judge. This case addresses ongoing legal disputes regarding the use of copyrighted material in training Large Language Models (LLMs). Judge Alsup noted that Anthropic’s LLMs are designed to create something new rather than replicate existing works, emphasizing the transformative nature of their training process. He pointed out that the authors did not claim the training resulted in infringing replicas of their works through the Claude tool. The outcome suggests that, if such claims had been made, the situation would have been different, highlighting the legal complexities surrounding the use of copyrighted material in AI development.

Source link