Generative artificial intelligence (Gen AI) chats have begun to show up as evidence in both criminal and civil cases. As Gen AI features integrate into everyday interfaces for customers and employees, companies should consider preparing for at least four types of situations where they will need to disclose AI chat logs, Google’s former director of law enforcement and information security told the Anti-Corruption Report. This second article in a series on Gen AI chats as possible evidence examines the types of requests that companies might receive and provides steps they can take to manage responses to those requests. Part one reviewed the use of AI chat evidence in three criminal cases and the unsettled law around external requests for such evidence. See our two-part series on the FirstEnergy decision: “Reaffirming Upjohn’s Approach to Privilege in Internal Investigations” (Dec. 3, 2025), and “Best Practices for Preserving Privilege in Internal Investigations” (Dec. 17, 2025).