Anthropic decided a collective lawsuit with a group of American authors who accused AI of copyright violation. IN Legal notification on TuesdayAnthropic claims that he negotiated the “proposed class settlement”, enabling the skipping process that would determine how much the company is guilty of alleged training of AI models in the field of pirate work.
The terms of the settlement are still not clear, but it results from the lawsuit of the copyright by the writers Andrea Bartz, Charles Graeber and Kirka Wallace Johnson last year, who claims that Anthropijka trained her Claude Ai models on Open Source data filled with pirate materials.
Anthropic won a sedate victory in June, when judge William Alsup ruled that AI models training in purchased books counts as allowed employ, but he left the door to further court disputes. In July, the ALSUP judge approved a collective lawsuit from American authors, who accused Anthropic of a violation of copyright “by taking millions of works in the style of Napster.” Anthropic was to go to a hearing regarding piracy claims in December, where he could face billions or over 1 trillions of penalties, According to Wired.
The settlement is to be finalized on September 3. “This historical settlement will bring benefits to all class members,” said Justin Nelson, authors’ lawyer, in a statement The Verge. “We are looking forward to the details of the settlement in the coming weeks.” Anthropic refused to comment.