Head of the Class

For more than two years, publications including National Jurist and The National Law Review have highlighted USF’s leadership in preparing students for the future of legal practice. When Johanna Kalb became dean in 2024, conversations with alumni made it clear that artificial intelligence was an immediate priority.

Among those voices was Aaron Crews ’04, who told her something she couldn’t stop thinking about: Her current students were going to graduate into a profession their legal education hadn’t prepared them for. Crews said, “Students need to learn how AI can operate as a thinking partner for lawyers, and they need hands-on practice with it.”

She started looking for ways to move forward and discovered that several of her faculty members were already getting started. Legal Writing professors Megan Hutchinson and Nicole Phillips, both practicing employment lawyers, had independently started experimenting with AI in their classes back in 2022. Neither considers themselves a technology enthusiast. What brought them to AI was their own experience in practice, where they could already see AI reshaping legal work. As Hutchinson says, “It was like we both created the wheel separately.” By fall 2025, AI instruction had become part of the required USF Law curriculum. Kalb says she provided resources and enthusiasm, and credits Hutchinson, Phillips, and other faculty members for laying the groundwork for everything that followed.

While other law schools were building certificate programs and optional AI tracks, USF chose a different path: Weave genAI instruction into the required first-year curriculum and frame it as one professional tool among many, rather than a separate subject requiring its own special designation. It was the first law school in the country to take this approach.

USF’s response was to build verification into the curriculum at its most foundational level. In the redesigned Legal Writing final exam, students scrutinize every legal proposition in an opposing motion for accuracy. Phillips and Hutchinson give their students three questions to apply whenever they are considering using AI in their work: Does it make you more efficient? Can you verify it? Would you be comfortable putting your name on it? All three need to be a “yes” before they proceed, and the standard applies to any document a lawyer puts their name on, regardless of who or what produced it. Those lessons don’t stop after the first year.

The technology may be new, but the underlying lesson isn’t. Good lawyering has always depended on sound judgment. “AI is like a 14-year-old who’s also a genius,” Hutchinson says. “You never know if you’re getting garbage or something brilliant, and the only way to find out is to check.” The lawyers making headlines for submitting AI-generated briefs full of invented citations are not, Hutchinson would argue, suffering from a technology problem so much as a verification problem; AI has simply made the consequences arrive faster and in more public ways.

For Kalb, the context matters. “If AI instruction lives off to the side,” she says, “it won’t connect with what students are actually learning in their doctrinal and skills courses.”

Initially, that context was Legal Research and Writing. In the fall semester at USF, students learn the fundamentals without AI. In the spring semester, after they’ve mastered the basics, the AI tools come out alongside more difficult questions: When should you use AI? When shouldn’t you? How do you verify what it produces? And who is ultimately responsible for the work? Hutchinson describes the sequencing with this analogy: You learn math by hand before you learn it on a calculator.

AI instruction now appears throughout the curriculum, from required courses in Professional Responsibility and Evidence to electives in Contracts Drafting, Community Property, and International Human Rights Law. In Professional Responsibility, students grapple with how the rules of the profession structure the responsible and ethical use of AI. In Evidence, students learn to use AI to improve their workflow. In a project designed by Crews, students work through an incomplete case file with Claude, mapping claims and defenses, surfacing evidentiary gaps, and building a litigation strategy. As students’ knowledge of legal practice and doctrine deepens, so does the sophistication of the work they’re asked to do with AI.

If AI instruction lives off to the side, it won’t connect with what students are actually learning in their doctrinal and skills courses.”

Kalb and the faculty also view emerging technologies as a tool for expanding opportunity within legal education. USF draws students from varied backgrounds, and not all of them arrive on campus having worked in a professional office, navigated a client-facing job, or had the chance to absorb the unspoken norms of legal culture simply by being around it.

Nicole Phillips built her bot ecosystem to address these gaps. The cold-call practice bot she created lets students choose their professor type, demanding or encouraging, theory-focused or practically focused, and rehearse before class. The client intake bot gives them their first simulated client meeting before real consequences are attached to the outcome. The negotiation bot lets them practice something most lawyers don’t attempt until they’re already billing for it. The goal, as Phillips frames it, is straightforward: Students who arrive without a professional head start should leave on equal footing with those who did. Kalb is working with the faculty to scale and adapt these simulations.

USF’s Jesuit mission shapes AI-related efforts in other ways. Through the law school’s Center for Law, Tech, and Social Good, students are encouraged to engage with AI as a transformative force in society. Michele Neitz, the center’s director, says the focus is on responsibility. “We want to help students build the habit of asking what a given technology will do to vulnerable populations before asking what it will do for anyone’s bottom line.”

Hutchinson and Phillips haven’t stopped questioning their own model, either. They’re already rethinking the math-before-calculator model, because each new class of students will have grown up with these tools in ways the prior year’s students didn’t, which means the curriculum that made sense at the beginning may become out of date as quickly as it’s completed. They discuss it roughly once a day. Hutchinson says, “It’s a constant evaluation of what we’re doing, why we’re doing it, and whether it’s still the best way to get students where they need to go.”

That kind of relentless self-examination is not what most people picture when they think about institutional change in legal education, which tends to move at the speed of committee consensus and accreditation cycles. What USF has built is something nimbler than that, and more honest: a program that treats its own assumptions as revisable, its tools as temporary, and its students and their growth as the priority.

Students Help Write the Rules

As the law school expanded its use of AI, Kalb wanted students to have a voice in shaping how the technology would be used. Working with Liza Levitt ’00, Intuit’s VP and deputy general counsel for platforms, responsible AI, and emerging technology, Kalb created a student advisory process to help develop guiding principles for AI use across the institution. Student representatives conducted interviews and surveys throughout the law school community, gathering perspectives from classmates, faculty, and staff before drafting a set of recommendations that were ultimately adopted through the school’s faculty governance process. Their surveys found a striking degree of consensus across different constituencies.

The concern that rose to the top, cited equally by students and faculty, was not job displacement or academic integrity or any of the fears that tend to dominate public conversation about AI. It was the loss of critical thinking: the worry that somewhere in the accumulating efficiency, something essential to the practice of law was being quietly handed off.

The principles they ultimately created emphasize responsible innovation, professional ethics, transparency, equity, and the development of core lawyering skills such as critical judgment and creativity. They now help inform decisions about curriculum, policy, and AI governance at the law school.

Henry Cunningham, a member of the student advisory team, said it was important for “students’ concerns about the use of technology to be addressed. One strength of our principles is that they are flexible and can be adapted as this landscape continues to change.”

AI in Action

For upper-division students interested in civil litigation, the Applied AI course offers a glimpse of their future workplace. Taught by alumnus and civil litigator Shounak Dharap ’16, the course explores how attorneys are using AI to research legal issues, analyze evidence, develop case strategy, conduct discovery, and prepare motions.

“Students come in wanting to know what the tools can do and the best prompts to get them there,” Dharap says. “What surprises them is how much of using AI well is about practice and restraint. You try everything—and then you figure out which tasks you still have to do the slow way, because that’s the only way you build the judgment necessary to turn an output into a real piece of lawyering.”