Artificial intelligence is transforming nearly every industry, and the legal profession is no exception.
At the University of Chicago Law School, faculty and administrators are weaving AI into the curriculum, preparing students for a profession in technological flux while ensuring they master the timeless skills of research, analysis and judgment.
“We have spent a lot of time thinking about AI in the Law School,” said William H. J. Hubbard, deputy dean and the Harry N. Wyatt Professor of Law. “Our aim is to find the right balance between encouraging students to explore these new tools that could be very useful, and not short-cutting in ways that would be pedagogically unhelpful.”
Hubbard, co-chair of the Law School’s AI Committee, observed that law firms, especially large ones, are embracing the technology, making it imperative for students and new graduates to arrive at their doorsteps equipped with a certain level of AI competency.
“We want to equip them with all tools of legal research and analysis—and that now includes AI,” he said.
Last year, the Law School added a new session to its fall orientation, “AI and the Legal Profession.”
Its purpose is to explain to incoming students the Law School’s policies around AI use and to provide a general overview of how AI can support and detract from learning as they begin their legal education.
Hubbard, who leads the session, said the main takeaways are: Generative AI is a useful and powerful tool that students will learn about as part of their legal education at the Law School; the technology has its limitations and risks; and AI does not fundamentally change the practice of law.
The Law School last year also added several new upper-level electives focused on AI to encourage students to think critically about how these tools can be used.
The courses also draw students’ attention to the pitfalls of relying on AI as a legal expert, says Mark Templeton, clinical professor and director of the Abrams Environmental Law Clinic.
“We can’t outsource expertise and knowledge to these AI models,” said Templeton, who cochairs the Law School’s AI Committee along with Hubbard. “These tools can generate what look to be beautiful pieces of writing, but when you look closely, there are so many errors because the tools don’t understand technical terms sufficiently. When you use AI, there is a duty to supervise it like you would a junior attorney or paralegal. And to fulfill that duty, you have to be the expert yourself.”
The Law School is also in the process of developing AI modules that will be required for all first-year law students to complete during their first quarter.
These modules, which will launch to students in early 2026, aim to bring everyone up to a minimum level of literacy with generative AI. They will be self-directed so that students who are more familiar with basic concepts can move quickly to more advanced sections.
“There are so many generative AI tools out there now,” said Hubbard. “With these modules we also want to steer students toward the right tools for legal work, ones that are both more tailored to the needs of lawyers and less likely to cause problems with confidentiality and privilege. These modules are meant to be building blocks for what we will continue to introduce to them during their time here.”
Weaving AI into coursework
The Law School’s Bigelow Program, which teaches the students legal research and writing over the course of the 1L year, has also been retooled to take AI into account.
Students are instructed not to use AI at all in the Autumn Quarter while they learn the basics of legal research and writing. In the winter, they are permitted to use AI tools, subject to certain guidelines and advice from faculty on how to use them appropriately and effectively.
This phased approach emphasizes the importance of developing foundational skills, then adding tools to support those skills. It also creates a supportive and low-risk environment for students to explore and practice using AI tools.
The Law School’s AI policy was designed to be flexible about how AI is used in the classroom. As such, professors have adopted many different approaches to using it—or not—in their teaching and coursework.
In his clinic, Templeton has fully permitted AI, allowing students to use and explore the tools in doing research, drafting discovery questions and even helping to write portions of briefs.
However, he requires that students disclose when they have used the technology as part of their process so that he and they can review and debate the quality of its output and improve their use of these technologies.
“There are so many tools out there designed to make a lawyer’s work easier,” he said. “We use Lexis and Westlaw to help us cross-check cases rather than Shepardizing. We use electronic word processors, whereas before we would use typewriters. So, the question to me is, how can these new tools augment our ability to work more efficiently while ensuring thoughtful, high-quality representation of our clients? I’m excited to be exploring this with students.”
Other faculty take a more restrictive approach. For example, Joan Neal, a professor from practice who teaches transactional skills courses, strictly prohibits using AI in her upper-level contract drafting course. The reason, she said, is that this is a foundational skills course.
“This is the first time they’re drafting transactional documents,” explained Neal, a member of the AI Committee. “They don’t have the base-level knowledge yet to judge the AI output to know if it’s any good or not. It’s extremely important—and ethically required—to focus on learning how to draft a contract the hard way first to develop this judgment.”
While the students themselves may not be using AI, Neal still addresses the topic in her transactional classes. She spends a lot of time critically analyzing and discussing with students the pros and cons of AI, the ethical considerations its use raises, and how they might use it in practice after they’ve acquired the basic skills.
“It’s beneficial for them to understand what AI is good at and what it’s not good at, and not to just completely stick our heads in the sand and ignore it,” she said.
Meanwhile, in her ethics class, Neal allows students to use AI in certain phases of the writing process for their final papers.
For example, they may use AI to help with brainstorming topics or for supplemental help with researching, but they are prohibited from using it to produce any of the text in the body of their papers. She requires students to reveal when they have used AI and briefly assess its usefulness, and she cautions them that they are ultimately responsible for everything they submit.
“I think students are finding that while AI can be helpful in some ways, it can fall short in many other ways,” she said. “Generative AI tends to give circular, generic answers to complex ethical issues. It’s not (yet) good at understanding the nuances of legal ethics rules.”
Neal hopes that her policies will help students conclude that they need to be masters of the material themselves, and that the tools are meant to supplement, not replace, their own critical thinking.
The launch of the AI Lab
One of the most innovative and hands-on AI learning opportunities the Law School has started to date is the AI Lab.
Launched this autumn, the lab focuses on creating AI tools—not just learning how to use them.
Legal tech entrepreneur Kimball Dean Parker, JD’13, teaches the class. A former student of Hubbard, Parker is the founder and CEO of SixFifty, a tech company that helps businesses stay compliant with employment law with a library of automated law documents and a proprietary AI database.
The AI Lab is following a similar model but instead focuses on renters’ rights, with students building a database of meticulously researched summaries of property rental laws nationwide. They’re learning to approach the project with an entrepreneurial mindset as they discover what goes into making a legal-tech product.
Part of the work involves determining scope of need. For that, students interviewed people to understand their questions on the topic and make sure the tool they create is useful to general users.
The final product will be an AI chatbot, similar to ChatGPT but much more accurate and reliable because of the specialized renters’ rights database it is drawing from. By the end of the course, this AI tool will be released to the public, giving it potential to have a real impact on people’s lives. It would serve users who need legal help but for financial reasons or otherwise do not have access to a lawyer.
“It’s unlike anything we have ever done,” said Hubbard. “There are only a few intensive workshops like the AI Lab anywhere in the country.”
“AI is like putty,” said Parker. “You have to play with it to understand it. The AI Lab is an opportunity for students to play in the sandbox of AI, to get hands-on experience with the defining technology of our age. Developing an understanding for the nature of the technology will make it easier for students to learn any similar tool they may encounter in the future, even ones that may not exist yet.”
Learning the law in the age of AI
As AI continues to evolve, the Law School will adapt alongside it. However, it will do so without losing sight of its core mission: preparing students to think critically, argue rigorously and own their work.
“Our graduates are hired for their judgment, for their mastery of the law,” said Adam Chilton, dean of the Law School and the Howard G. Krane Professor of Law. “That mastery is not something you can outsource to any technology or machine. A UChicago Law graduate’s good judgment, intellectual robustness and reputation in owning their own work is priceless.”
With that in mind, it is also important to recognize that AI is a part of how law is practiced today, and students must be prepared for it, Chilton continued.
“By incorporating AI thoughtfully across the curriculum, the Law School is ensuring its graduates not only adapt to a changing profession, but lead it,” he said.
—Adapted from an article originally published in the UChicago Law Magazine.
