UC Berkeley Law Bans AI Use

- UC Berkeley School of Law released a new artificial intelligence policy on May 22, 2026, barring students from using AI on assignments and exams. - The rule bans AI for “conceptualizing, outlining, drafting, revising, translating, or editing” graded work, while allowing limited source-finding research, Berkeley Law says. - The policy takes effect in Summer 2026, with instructors allowed to set different rules in courses designed to teach AI fluency.

UC Berkeley School of Law released a new artificial intelligence policy this week that sharply limits how students can use generative AI in coursework. The policy, effective Summer 2026, bars AI use for assignments submitted for credit and for any purpose in exams, according to Berkeley Law’s registrar page and the full policy document. The school said students may still use AI in a narrow way to identify research sources, such as cases, statutes or secondary materials. The move places one of the country’s best-known law schools among the stricter institutions in higher education on student AI use. ### What exactly did Berkeley Law ban? Berkeley Law’s written rule says “the use of AI is prohibited” for conceptualizing, outlining, drafting, revising, translating or editing any work submitted for credit. The same policy says AI use is prohibited “for any use for any purpose in any exam situation.” The policy also bars students from uploading course materials — including assignments, readings, slides, class recordings and other class content — into generative AI systems. (law.berkeley.edu) Berkeley Law said AI can be used on papers only for the limited purpose of identifying sources. ### Which uses count as violations under the new rule? The Berkeley Law policy lists examples that go beyond asking a chatbot to write a paper. (law.berkeley.edu) The school says prohibited conduct includes asking AI to brainstorm a paper topic or thesis, propose an organizational structure, compose a paragraph summarizing a legal rule, identify repetitive passages to cut, correct grammar, generate an exam outline or translate a paper into English. Amanda Robert of the ABA Journal reported on May 22 that the school announced the policy this week and described those same categories as barred uses for class assignments and exams. The ABA Journal article said the rule was released Thursday. ### Why did Berkeley Law say it made the change? Berkeley Law said in the policy that “thinking remains the sine qua non of good lawyering” and that the rule is meant to keep courses focused on students’ own cognitive skills. (law.berkeley.edu) The school said the purposes are to provide “the best legal education possible” and to promote “fairness and administrability.” (abajournal.com) Chris Hoofnagle, a Berkeley law professor, told the San Francisco Chronicle, as quoted by the ABA Journal, that he proposed the new rules after seeing questionable legal reasoning in an increasing number of student assignments. Hoofnagle said, “If you don’t have your own analytical judgment, AI will do it for you, and then it’s no longer your judgment.” ### Does the policy ban AI everywhere in the law school? (law.berkeley.edu) The policy is a default rule, not an absolute campus-wide prohibition. Berkeley Law says instructors may depart from it in courses “designed intentionally to teach AI fluency” or in other classes where a professor decides a different rule is appropriate. The ABA Journal also reported that professors who teach courses involving AI fluency, or who want students to use AI in certain situations, can seek exemptions from the policy. (abajournal.com) Berkeley Law’s own academic pages continue to promote AI-focused courses, degree offerings and research programs, underscoring that the school is restricting student use in graded work while still teaching AI as a legal subject. (law.berkeley.edu) ### How does this compare with Berkeley’s earlier approach? A 2023 Berkeley Law rule, cited in a Berkeley Law article and archived policy document, had allowed some uses of generative AI while barring uses that would amount to plagiarism if a human had produced the material. That earlier framework was narrower than the Summer 2026 rule. The Summer 2026 policy now sets a broader baseline across the law school. (abajournal.com) Berkeley Law’s registrar page says the rule is already posted as an academic policy, and the PDF version identifies it as the “post-faculty meeting” policy for implementation in Summer 2026. Summer 2026 is the next concrete milestone in the rollout. Berkeley Law said faculty members can set course-specific exceptions, and students can find the full rule on the school’s registrar and policy pages. (law.berkeley.edu) (law.berkeley.edu)

Get your own daily briefing

Scout delivers personalized news, insights, and conversations tailored to your role and industry.

Download on the App Store

Shared from Scout - Be the smartest in the room.