Photo by RDNE Stock project on Pexels

A federal judge in the Southern District of New York has delivered a blow to OpenAI in a consolidated copyright case involving its ChatGPT model. Judge Stein denied a motion to dismiss, focusing on the issue of ‘output-side’ copyright infringement. The ruling hinges on whether ChatGPT’s generated outputs are substantially similar to copyrighted works used in its training. The plaintiffs, representing fourteen consolidated AI copyright cases, successfully argued that ChatGPT’s outputs could indeed infringe on existing copyrights, specifically citing examples referencing George R.R. Martin’s *A Game of Thrones*. While the judge’s decision doesn’t delve into ‘fair use’ arguments, it establishes a crucial precedent for future legal battles concerning AI-generated content and copyright law. This marks a significant step in defining the boundaries of AI-generated content within existing copyright frameworks. Further details on the case and other AI copyright rulings can be found on Reddit: [https://www.reddit.com/r/ArtificialInteligence/s/B0V5Ny2p0j]. The original Reddit discussion can be found here: [https://old.reddit.com/r/artificial/comments/1ohsqmx/important_new_ai_copyright_outputside_ruling_in/]