Academic Integrity Statement

1. Why this page exists

Tools like this one are usually vague about what they actually do. We would rather be plain. ProfessorAI does two things that matter for your integrity as a researcher: it finds and cites real literature, and it writes text.

Both have consequences. This page sets out what we guarantee, what we refuse to claim, and what remains yours to decide.

2. The one thing we enforce mechanically

The most common way AI tools damage academic work is by inventing references. ProfessorAI is built so that this cannot happen quietly:

• Sources come from a live search of open scholarly databases. Nothing is invented at that stage, because nothing is generated at that stage.
• When the professor writes with citations, it may cite only records from your own search results.
• Before a draft is saved, an automatic check confirms that every cited author and year exists in your source list. A draft that cites outside the list is rejected and written again.
• Where no source fits a claim, you are shown a visible marker instead of an invented reference.
• Reference lists are cross-checked against Crossref, OpenAlex, PubMed, and other databases. Entries that cannot be matched are marked unverified rather than hidden.

This is a strong guard and it is deterministic, not a matter of the model behaving well. But it proves that a record exists. It does not prove that the record supports the sentence it is attached to. Read what you cite.

3. It writes, and we are not going to pretend otherwise

ProfessorAI produces academic prose: outlines, sections, paragraphs, alternative sentences with citations, summaries, and cover letters. It also edits text you wrote. That is the product, and hiding it would be the first dishonest thing here.

Treat what comes out as a draft handed to you by an assistant who reads quickly, writes fluently, and is sometimes wrong. Argue with it. Cut it. Rewrite it in your own voice and against your own reading of the sources. Whatever you keep, you are answerable for, exactly as if you had typed it yourself.

4. Disclosure is yours to make

Institutions, supervisors, funders, journals, and publishers differ on what AI assistance is permitted and how it must be declared, and those rules are still changing. We do not decide for you, we do not speak for your institution, and we do not claim that using this service makes disclosure unnecessary. Find the rules that apply to your work and follow them. If a journal asks how a manuscript was prepared, answer accurately.

5. What we do not claim

• We do not certify that a text was written by a human. No tool can do this honestly, including ours.
• We do not promise that any text will pass an AI detector. Those detectors are unreliable in both directions, and we will not build a promise on them.
• We do not check your work for plagiarism or textual similarity.
• We do not guarantee acceptance by any journal, reviewer, examiner, or committee.
• We do not decide what your institution requires you to disclose.

6. If we fall short

If you find a generated citation pointing to something that does not exist, a reference the checker passed wrongly, or anything else short of this statement, write to [contact email]. We treat reports like these as defects rather than as opinions, and we would rather hear about one than have it reach a reviewer. The Terms of Use set out the rest of the agreement.