AI feedback on a thesis can count as cheating or academic misconduct when your rules don't permit it. At the University of New Hampshire, for example, unapproved generative AI in a thesis is academic misconduct. A permitted tool, reported the way your program asks, is a different case. Ask your advisor first and keep the answer.
No one sees their own blind spots, with or without AI help. Esmond points them out in your draft from within your field, along with factual errors and inconsistencies.
The words differ per school, and the consequences with them. These passages are quoted as published. Open the source before you rely on them, because policies change.
| School | What it says |
|---|---|
| University of New Hampshire, Graduate School | “Unapproved use of generative AI in exams, theses, or dissertations constitutes academic misconduct and is grounds for failure of the relevant exam or thesis based on the judgement of the evaluating committee.” |
| University of Georgia, Graduate School | “the use of generative AI in theses and dissertations is considered unauthorized assistance per the Academic Code of Honesty” |
| Wake Forest University, Graduate School of Arts & Sciences | “unauthorized use of generative AI platforms will be considered academic misconduct.” |
| Brigham Young University, Chemistry and Biochemistry Graduate Program | “Unauthorized usage of AI tools for scholarly work may be considered an offense of research misconduct” |
| Illinois State University, Graduate School | “Unauthorized use of generative AI may be considered a violation of Policy 1.8 Integrity in Research, Scholarly, and Creative Activities” |
| University of Waterloo, Academic Integrity | “A student who does not comply with the instructor's rules about the use of such tools will be subject to Policy 71 and an investigation into academic misconduct.” |
A hypothetical example of a commentYour AI statement says you used a tool for spelling only, and its next sentence says the tool coded your interviews. Which is correct?
A first comment on one part of your own draft, right away and without an account.
Get a free previewEach of the six passages turns on whether a use was approved, authorized or within the rules. Wake Forest's guidance is issued for courses and Waterloo's passage concerns an instructor's rules, so read those two as examples of the wording. Georgia's policy goes on to allow what the advisory committee specifically authorizes, and New Hampshire says AI tools may be used for approved and clearly acknowledged purposes. So the same tool can be permitted in one program and misconduct in another, depending on what each program decided.
New Hampshire's Graduate School has no universal policy and leaves the choice to programs. Your school may work the same way, so the page you need can be your own program's handbook.
The passages above don't set feedback apart. They speak of generative AI, and a tool that only comments can fall under that. Don't assume you are outside the rule because every sentence stays yours.
Definitions differ in reach. Northwestern's Department of Chemistry repeats the university's definition of cheating, which includes “allowing another person or resource (including, but not limited to, generative artificial intelligence) to do one's work and submitting that work under one's own name without proper attribution”. That definition doesn't say how comments are treated, so ask your program. Georgia's policy reproduces broader guidance from its academic honesty rules, where during an assignment “any help (such as books, notes, calculators, technology, internet resources, or conversations with others) is considered unauthorized unless the instructor explicitly allows it.”
Some guidance compares AI with a person. Stanford's default policy for courses, adopted in October 2026, says: “If the use of a human tutor is permitted for a given task, then it is likely that the use of generative AI as described in the examples above is similarly permitted.” Its examples are practice and comprehension, and students “may not use AI tools to create deliverables”. That guidance covers coursework, and a thesis can have its own rules, so read it as one way to think about the question.
Your school sets the consequences. New Hampshire's passage above names failure of the exam or thesis. Brigham Young's chemistry program says unauthorized use “could result in an unsatisfactory rating and possible dismissal from the program.” At Waterloo, a case leads to an investigation under Policy 71. Look up your own school's procedure before you need it.
Esmond evaluates every assignment before you submit. It never writes it.
Esmond is an AI tool that doesn't write for you. If your course doesn't allow any AI tools, don't use Esmond for that work. If you're unsure, ask your instructor or advisor.
Esmond returns comments and adds no text to your thesis. It works with language models, the same kind of technology as generative AI tools, so the rules above can apply to it. No college or university has recognized or approved it.
Can I use Esmond? lists what to check and how to name it, and the sample report shows what comes back, so you can show your advisor what you are asking about.
It can, if your rules don't permit it. At the University of New Hampshire, unapproved generative AI in a thesis is academic misconduct, while approved and clearly acknowledged uses are allowed. Check your own school's wording.
It can be. The passages quoted here turn on approval and disclosure, and none of them exempts a tool because it adds no text.
At some schools. Stanford's default course policy permits surface-level editing such as correcting spelling and grammar. Your thesis rules may differ, so ask.
A missing policy doesn't settle the question. Ask your advisor by email, and keep the answer.
Read on November 5, 2026.
Get your paper evaluated before you submit