Whether large language models are "memorizing," "regurgitating" or "learning" information is at the center of many debates around generative artificial intelligence's legal standing, including in the most recent New York Times' complaint against OpenAI and Microsoft alleging copyright infringement.