If you’re a college or university student who’s looking up the AI rules at your school, you probably want answers to two questions: what’s allowed, and what actually happens if you get it wrong.
At the University of Virginia, neither answer is especially tidy.
The school has taken a flexible, instructor-driven approach to AI, and the Honor Code process has changed in recent years. That means the answers are somewhat murky, but one thing is for sure: the consequences of using AI in your coursework can stick around longer than you’d think.
If you are a UVA student facing an Honor Code allegation involving AI, contact us here or call (888) 535-3686 to speak with the Student Defense Team at the LLF National Law Firm today.
UVA’s Approach to Generative AI: Flexible, But With Real Boundaries
There isn’t a single, campus-wide AI rule at UVA. No blanket ban, but also no universal green light.
Instead, the university puts out general guidance through the Provost’s office and leaves the specifics to individual instructors, which sounds simple enough — until you realize what that means in practice.
In one class, AI might be encouraged as part of the process. In another, it might be limited to brainstorming or outlining. Somewhere else, it might be off the table entirely. All of those approaches are valid, and they all carry weight under the Honor Code.
The syllabus is shouldering a lot of the burden. If something isn’t clear, it’s on the students to get clarification.
Some students run into trouble by assuming the rules transfer wholesale from one course to another. What was fine according to one instructor could be verboten to another. Expectations can shift even within the same department.
What the Guidance Actually Requires
Even with that course-to-course flexibility, a few expectations show up consistently across UVA’s guidance.
- If AI played a role, you’re expected to disclose it …and not in a vague way. The expectation is that you identify the tool, explain how you used it, and make some effort to show how you checked or verified the output. UVA points students toward MLA-style practices here, which treat AI use as something that should be acknowledged, not quietly folded in.
- Using AI outside the rules — or not disclosing it — can trigger an honor code issue. This is where things stop being fuzzy. If your instructor doesn’t allow AI use at all, or you turn in AI-generated work as if it were entirely your own, that can be treated as cheating. Most cases hinge on that gap between what was permitted and what actually happened.
- A privacy angle that’s easy to miss. UVA specifically warns against putting sensitive or personal information into public AI tools. A lot of platforms store prompts, and some use them for training. To get around that, UVA offers a university-managed version of Copilot Chat where data stays inside that system. Moreover, students also aren’t required to sign up for outside tools that ask for personal information.
- AI detectors aren’t the safety net people assume. UVA has been pretty blunt about this. Checkers and detectors are not only unreliable, but they can also actually create additional problems — especially when student work gets uploaded into external systems, which conflicts with students’ intellectual property rights. This may provide meaningful protection for students, but it does not alter the obligation to follow course-level rules.
- Your work is your intellectual property. The UVA Provost’s guidance specifically notes that students’ original work is, in most cases, their intellectual property. In other words, instructors may not enter a student’s assignment into an AI system that will add it to a training dataset, nor can they employ AI-powered grading tools.
Different Schools, Different Lines
Additionally, some of UVA’s schools have added their own policies.
The School of Medicine, for example, allows AI as a support tool but draws a clearer line than most when it comes to independent reasoning. You can use it to assist, but not to stand in for your own analysis or reflection. If that line gets crossed, it can trigger both school-level discipline and the Honor Code.
Darden, the university’s graduate business school, has taken a fairly permissive approach in certain courses (including one core strategy course that actively encourages AI use) while still requiring full transparency about how it was used.
UVA’s overall picture is one of a university that is leaning into AI rather than shutting it down — but strictly insisting on honesty, attribution, and instructor-level clarity as the non-negotiable ground rules.
What it all boils down to is that AI use itself isn’t treated as misconduct. The issue is what part AI plays in a student’s coursework, and whether the student is upfront about it.
Worst Case Scenario
UVA’s Honor Code covers lying, cheating, and stealing. AI-related cases usually fall under cheating, especially when plagiarism is involved. The university explicitly states that plagiarism includes AI-generated content.
So while the rules on the front end can feel flexible, the back end isn’t.
Submitting AI-generated work without permission or without acknowledging it as such lands squarely within Honor Code territory.
How the Process Unfolds
Once a report is filed, things follow a set path, starting with the investigation. Two student investigators gather statements, collect materials, and build a record. That record then goes to a small panel, which in turn evaluates the likelihood that a violation occurred.
When that likelihood seems strong, the case moves forward.
From there, the student has a choice to make. You can contest the allegation and attend a hearing, where a larger panel applies the standard of “beyond a reasonable doubt.” Another option is to accept responsibility and move directly into the sanctioning stage — getting it over with, so to speak.
There’s also an option called an Informed Retraction; this is a structured way to admit the violation under defined conditions.
None of those paths is purely procedural. Each one comes with tradeoffs, making part of the process very difficult to navigate without experienced legal counsel.
Sanctions
Up until 2023, UVA defaulted to automatic expulsion for students found culpable of using AI. Today, there’s a range of possible outcomes:
- Permanent removal (expulsion)
- Temporary removal (suspension)
- Transcript notations
- Reflection letters
- Ethics seminars
- Mentorship requirements
- Letters of apology
Panels have a fair amount of leeway when it comes to their determination, however. What tends to matter most is context. Was the conduct intentional? Was it repeated? Did it involve planning? Panels also look at how a student responds once the issue is raised.
As a result, outcomes vary quite a bit.
One detail that sometimes catches people off guard: professors operate on a separate track. Regardless of what happens in the Honor Code process, an instructor can still assign a failing grade on the assignment, or even for the course as a whole.
For students in professional programs, the effects can be far more damaging than you might imagine. Findings of misconduct often rear their ugly heads down the line, in recommendations, licensing processes, or other career steps.
Protecting Your Future
An Honor Code allegation involving AI isn’t just a classroom issue. It’s a formal process with real stakes, and it doesn’t always unfold the way students expect.
The Student Defense Team at the LLF National Law Firm works with students navigating exactly these situations, from the initial investigation through hearings, sanctioning, and appeals. If you’re a student facing an Honor Code allegation at UVA, contact us immediately to learn how the LLF National Law Firm can help protect your academic future. Call (888) 535-3686 or fill out this form.