The University of Chicago Law School has released a strategy statement, Rethinking Legal Education in the AI Era, and its most concrete commitment is a change to the first-year classroom. Beginning this fall, and running through the 2026–2027 academic year as a pilot, the school will prohibit laptops, tablets, and phones across every section of its nine core 1L courses, from Civil Procedure to Transactional Lawyering. Examinations in those courses will be given in class “without access to the internet, electronic files, or apps.” Instructors may designate a rotating “scribe” to take notes electronically for the class, may enable devices for a specific activity such as polling, and the school says it will accommodate disabilities “in conformity with applicable law.”
This is a more careful document than a headline about banning laptops suggests. Chicago Law has not banned AI. Its strategy rests on three themes—AI-resilient pedagogy, elevating the human skills that distinguish good lawyers, and teaching responsible use—and on a motto: students should learn to think with, without, and about AI. The first-year legal research and writing curriculum makes the point plainly, treating “writing without AI as the foundation” and layering “writing with AI onto it,” so that students build their own judgment before they are asked to supervise a model’s. That is the same sequencing I have defended before, and it is more discerning than Berkeley Law’s blunter default, which I discussed recently. Where Berkeley Law forbids the tool by default and defers instruction to an upper-level elective, Chicago Law builds supervised use into the required first-year course. On the question the two schools share, when to introduce AI and how, Chicago Law has the better answer.
The device ban is a different instrument from the AI strategy that houses it: screens in a classroom are not strictly an AI problem. The ban is really an attention-and-engagement policy, justified on the ground that “reliance on devices to take notes or assist with answering questions tends to inhibit reflection and reasoning,” and that “active, in-person engagement is conducive to learning.” The goal is one I share, but the policy rests on two assumptions the strategy statement does not examine, and each has an effect on students that the document leaves out. This is the same shape of problem I found at Berkeley: a defensible aim paired with an instrument whose costs fall on the wrong people.
The exception that exposes the students it exempts
Every general prohibition on classroom devices has to make room for students whose disabilities require one, and Chicago Law’s does. The difficulty is that a disability accommodation is not supposed to be visible, and a blanket ban makes it so. In a lecture hall where no one else may open a laptop, the student who does has announced something. As Ruth Colker, a disability-law scholar at Ohio State, argued in a study built on a ban in her own law school classroom, a no-laptop rule with a disability exception “serves to stigmatize students with invisible disabilities who will need to request an exception.” Their classmates can see who is typing, and some students, rather than be marked, go without the accommodation to which they are entitled.
This inverts the confidentiality the accommodation process is built to protect. Federal disability law lets a student keep the fact and nature of a disability private; the point of the interactive accommodation process is that a professor, and certainly a room of peers, need not know who holds one. A device ban quietly defeats that, and it compounds a burden that is already substantial. To obtain the accommodation in the first place, as Katherine Macfarlane has written, students “must disclose intimate details about their health” through a process many find arduous enough to forgo. The ban then asks them to disclose again, informally and to everyone, every time they open the laptop the process granted them. This is not hypothetical: the ABA Journal has reported on an autistic law student with a laptop accommodation who was seated apart and drew comments about his device.
The scribe exception, which lets a rotating student take notes electronically for the class, does not close this gap, because it solves a different problem. A shared set of notes helps everyone who wants the record, but a student who needs to type her own notes, because she cannot write by hand at the speed of a first-year class or because handwriting itself is the barrier, still has to seek an individual accommodation and still becomes the identifiable exception when it is granted. The remedy for the exposure problem is not a scribe; it is a policy that does not require the accommodated student to stand out in the first place. A ban that permitted quiet, non-networked note-taking on a locked-down device for anyone, or that presumed a mix of laptop and longhand across the room, would preserve most of what Chicago wants from the classroom without turning an accommodation into a public fact.
What a handwriting requirement measures
The ban’s second assumption is empirical: Chicago Law grounds it in “an emerging scholarly consensus” that devices inhibit learning. The intuition that writing by hand forces the reframing that typing lets students skip has support. Mueller and Oppenheimer’s 2014 study reported that laptop note-takers, prone to verbatim transcription, did worse on conceptual questions than students who wrote longhand. That paper is the one most often cited to justify laptop bans. What the consensus framing leaves out is that its central result has not held up. A preregistered direct replication led by Heather Urry in 2021 reproduced the difference in what students wrote but not the difference in what they learned, and a mini meta-analysis across similar studies pointed the same way; the authors concluded the results “do not support the idea that longhand note taking improves immediate learning.” A 2019 replication and extension by Morehead, Dunlosky, and Rawson likewise found no reliable longhand advantage. So the evidence is contested, and that makes forced handwriting a thin foundation for a uniform mandate imposed on every 1L in the building.
The exams raise a sharper concern, because the policy leaves their format unspecified. First-year examinations will be given in class “without access to the internet, electronic files, or apps,” but the statement does not say whether students will type on locked-down machines or write in bluebooks. If the exams are handwritten, the school has introduced into its measuring instrument a variable that has nothing to do with legal analysis. Handwriting speed is not evenly distributed, and it constrains performance under time pressure: Connelly, Dockrell, and Barnett found that undergraduates’ handwriting fluency measurably depressed the marks they earned on timed exam essays, independent of the quality of their thought. A handwritten exam, then, measures reasoning and penmanship at once and cannot fully separate them. In instructional-design terms this is a constructive-alignment problem, the same one I raised about Berkeley and Texas: an assessment should measure the outcome it targets, and a first-year exam is meant to test whether a student can analyze a legal problem, not how fast she forms letters. The variance handwriting introduces is construct-irrelevant, and it falls hardest on exactly the students the first section was about: those with dysgraphia, motor impairments, or simply slow hands. That is where the two effects meet.
None of this reaches the core of a good 1L course: the Socratic exchange, the sustained back-and-forth I have argued is what scarce class time should protect. A student is no less present in that exchange for holding a paper notebook rather than a laptop. If the case for the ban is engagement, the Socratic method already secures it, and the school can have that benefit without resting on a note-taking result that has not replicated or an exam format that may not measure what it grades.
Weighing the second-order effects
I am not opposed to what Chicago Law is trying to protect. The first year is when the foundations of legal reasoning are laid, when a student’s ability to judge a machine’s output is at its weakest, and when the case for keeping the room’s attention undivided is strongest. Under ABA Standard 314, every school must assess whether its students reach the outcomes it claims for them, and a classroom built for sustained attention serves that obligation.
But a policy this good deserves to be judged against its own second-order effects, not only its aims, and two of them are troubling. A device ban exposes the student who with an accommodation. The assumption that longhand improves learning has not been reproducible, and, worse yet, for handwritten exams, it risks unintentionally grading a mere incidental skill. Chicago Law needs to calculate what the ban costs the students it should most want to protect.
This post draws on the University of Chicago Law School’s AI strategy statement, Rethinking Legal Education in the AI Era, and reporting on the 1L device policy from LawSites. On accommodations and classroom device bans, it draws on Ruth Colker, Universal Design: Stop Banning Laptops!, 39 Cardozo L. Rev. 483 (2017); Katherine A. Macfarlane, Accommodation Discrimination, 72 Am. U. L. Rev. 1971 (2023); and the ABA Journal’s report on laptop prohibitions. On note-taking and assessment, it draws on Pam Mueller & Daniel Oppenheimer, The Pen Is Mightier Than the Keyboard, 25 Psychol. Sci. 1159 (2014); Heather Urry et al., Don’t Ditch the Laptop Just Yet, 32 Psychol. Sci. 326 (2021); Kayla Morehead, John Dunlosky & Katherine Rawson, How Much Mightier Is the Pen than the Keyboard for Note-Taking?, 31 Educ. Psychol. Rev. 753 (2019); and Vince Connelly, Julie Dockrell & Jane Barnett, The Slow Handwriting of Undergraduate Students Constrains Overall Performance in Exam Essays, 25 Educ. Psychol. 99 (2005). It builds on earlier posts on the delegation framework, answer quality versus learning, revised Standard 314, and Berkeley, Texas, and Boston College.