UC Berkeley law school bans students from using AI for coursework, exams

The University of California Berkeley School of Law has implemented a policy banning students from using artificial intelligence for coursework or exams beginning this summer. Students will not be allowed to use AI for conceptualizing, outlining, drafting, revising, or editing their work, including grammar corrections or translations. The policy permits AI use only for identifying legal sources such as cases, statutes, or secondary materials, though professors may authorize exceptions in writing with prior notice. The rule was proposed by Professor Chris Hoofnagle after observing suspicious legal reasoning in student submissions. Hoofnagle argued that AI undermines independent analytical judgment, which is critical for legal practice. Another professor, Jonathan Glater, noted that while the policy makes sense for introductory courses, he plans to teach proper AI use in advanced classes to prepare students for real-world applications. Faculty concerns about AI overreliance extend beyond Berkeley. A recent study by the American Association of Colleges and Universities and Elon University found that 95% of educators fear students will become dependent on generative AI, leading to weakened critical thinking and diminished academic integrity. The policy reflects broader anxieties in higher education about maintaining foundational skills in an AI-driven era. Berkeley’s approach balances restrictions with flexibility, allowing professors to adapt the rules for specific instructional needs. The school’s goal is to ensure students develop the cognitive and ethical skills necessary to strategically and responsibly use AI in legal practice.