In a pivotal decision for the AI sector, a U.S. District Judge based in San Francisco, William Alsup, has affirmed that Anthropic, an artificial intelligence company, legally employed copyrighted books to cultivate its chatbot, Claude. However, although the AI training method got the green light, the company is not out of the woods yet, as it faces legal scrutiny over how it sourced these books, primarily drawn from online repositories of pirated copies, setting the stage for a trial in December to hash out the damages owed to the authors, according to Reuters.
While the court acknowledged that Anthropic’s AI system’s use of the texts was “quintessentially transformative”, making it fair under copyright law, the judge also pinpointed the fact that Anthropic had housed over 7 million pirated books in a central library which was deemed an infringement of authors’ rights, taking them to task over their methods of acquisition, as reported by Fortune. Anthropic’s backing comes from heavyweights like Amazon and Alphabet, highlighting the big tech interest in the development and application of generative AI technology, despite the contentious nature of how such systems are trained.
The lawsuit, spearheaded by authors including Andrea Bartz, Charles Graeber, and Kirk Wallace Johnson last year, brought to light Anthropic’s alleged use of pirated book copies to input into their AI, which the plaintiffs described as “large-scale theft.” In response to the legal fray, Anthropic argued that their use of the books was to “study Plaintiffs’ writing, extract uncopyrightable information from it, and use what it learned to create revolutionary technology,” a statement highlighted by Reuters.
Copyright for syndicated content belongs to the linked Source link
