Bloomberg Law- Law Clerk vs. AI? Courthouse Test Highlights Judicial Curiosity [paywalled] [unpaywalled] “Law clerks and interns for federal Judge Xavier Rodriguez recently spent weeks poring over evidence from a high-profile trial on challenges to Texas’ voting and election laws, and then summarized key testimony for the court’s official findings of fact and conclusions of law. This summer, an AI tool is doing the same thing. While only the human-powered work will become part of the court record, Rodriguez plans to publish results on how well and how quickly an artificial intelligence tool performed the summarization and analysis compared to trained young lawyers and law students, he told Bloomberg Law. “I am not aware of other judges doing similar experiments,” Rodriguez said in an email, “but I would be surprised if they’re not.” A member of the Texas AI legal task force, Rodriguez is among the federal and state judges dipping their toes into how the technology can be used in and around the courtroom. The questions they’re posing were highlighted on May 28 when Judge Kevin Newsom of the US Court of Appeals for the Eleventh Circuit wrote a concurring opinion in an otherwise mainstream insurance case floating the idea that generative AI large language models could be used to help determine the “ordinary meaning” of legal text. “Here’s the proposal, which I suspect many will reflexively condemn as heresy, but which I promise to unpack if given the chance,” Newsom wrote in a concurrence that at 29 pages was longer than the opinion. “Those, like me, who believe that ‘ordinary meaning’ is the foundational rule for the evaluation of legal texts should consider—consider—whether and how AI-powered large language models like OpenAI’s ChatGPT, Google’s Gemini, and Anthropic’s Claude might—might—inform the interpretive analysis. There, having thought the unthinkable, I’ve said the unsayable.” Newsom’s concurrence in a case about whether installation of an in-ground trampoline and retaining wall fit the common understanding of the term “landscaping” largely focused on the possibilities and pitfalls of AI. And it marked a “starting gun” for other judges who want to experiment with those parameters in the courtroom, said Dazza Greenwood, executive director of MIT’s Computational Law Report. “This technology exists,” said Greenwood. “It can and should be used in some ways for this.”