This qualitative study rates the level of understandability, actionability, and procedure-specific content in postoperative instructions generated from ChatGPT, Google Search, and Stanford University.