How to prove you didn't use AI: version history in Word and Google Docs

If Turnitin flagged something you wrote in Word or Google Docs, the version history built into that file is the strongest timeline evidence you can bring. It shows the document being edited across multiple sessions over days or weeks, not pasted in one go. It does not prove no AI was ever involved. It proves the file has a writing history, and the AI score cannot see that history because the score only looks at the final text. The rest of this article is how to export it, what to label, and what to bring alongside it.

There is a second thing to know before you do any of this. The company that built the tool has said, in writing, in three separate help documents, that an AI score should not be used as the sole basis for action against a student. That is not your argument. It is theirs. We will come back to it because it changes what the version history is for. You are not bringing it to prove a negative. You are bringing it so the meeting is about evidence and not about a single number.

Version history is a timestamped log of changes to a document. Word keeps one through OneDrive autosave and Track Changes. Google Docs keeps one natively. Overleaf keeps a Git-style page history. All three record when edits happened and roughly how large each revision was.

What it can show:

• The document was edited over multiple sessions on different dates.
• Some revisions were small (a paragraph, a sentence) and some were larger.
• The file existed in earlier states, sometimes weeks before submission.
• Edits cluster around feedback events, supervisor comments, or your own revision passes.

What it cannot show:

• That every sentence was written by a person.
• That no AI tool was used at any stage.
• That the final text matches the intermediate drafts line for line.

This is the part most people get wrong. You are not using version history to prove a negative. You are using it to show that the document has a history consistent with being written over time, and that history is something the score does not see. The score looks at the final text. The version history looks at how the final text got there.

The distinction matters because of what Turnitin itself has said about its own score. The AI writing report guide says:

"Our AI writing detection model may not always be accurate (it may misidentify human-written, AI-generated, and AI-paraphrased text), so it should not be used as the sole basis for adverse actions against a student."

The sentence immediately after it is the one that gets dropped every time this quote is used:

"It takes further scrutiny and human judgment in conjunction with an organization's application of its specific academic policies to determine whether academic misconduct has occurred."

Read both together. The first says do not use this number alone. The second says what should happen instead: further scrutiny, human judgment, the institution's own policies. That is the company that built the tool telling the person holding the number to do more work before drawing a conclusion. Your version history is what makes that further scrutiny possible.

There is also a separate page titled How should I review the AI Writing report? which puts it differently:

"It is not meant to provide definitive answers in isolation. More important than any tool is the educator who sees the score and makes decisions balancing this information with their personal knowledge of their students, their work, and institutional policy."

The next sentence completes it:

"When educators look at the AI writing score and utilize it as a single data point rather than a definitive response, then it is being used as intended."

That is the manufacturer calling its own score a single data point. Not a verdict. Not a definitive response. If the meeting starts and ends with a number, the person running it is using the tool in a way the manufacturer says it should not be used. We went through what the score does and does not tell you in how to read a Turnitin AI writing report.

Word keeps two separate records and they are not the same thing.

Track Changes is an in-document feature that marks insertions, deletions, and formatting changes inline. If you had it on while writing, the document itself carries a visible edit log. If you had it off, that log does not exist, and no setting will bring it back after the fact.

Version history is a file-level record kept by OneDrive when autosave is on. It stores snapshots of the document at intervals and lets you restore or view earlier versions. To access it: open the document in Word for the web or the desktop client with OneDrive synced, go to File > Info > Version History (desktop) or File > Version history (web). A pane opens on the right listing saved versions by date and time.

What to do with it:

1. Open the version history pane and scroll to the earliest version. Note the date. If the file is months old and has dozens of versions, that already tells a story.
2. Export or screenshot the list of versions. You want the dates and the author labels. Word for the web lets you open each version in a read-only view.
3. Pick three or four versions along the timeline and note what changed. Not every version. A first draft, a version after a round of revisions, a version after supervisor feedback, and the final. One sentence per version is enough.
4. If you used Track Changes, keep a copy with the marks still visible. Do not accept all changes before the meeting. The marked-up version is the one that shows the editing process inline.

If autosave was off and there is no version history, you still have something. File modification dates on earlier drafts, email attachments sent to your supervisor, and any cloud-synced backup folders all carry timestamps. They are not as clean as a version history pane, but a folder of dated drafts tells the same story.

Google Docs keeps version history natively and it is the easiest of the three to export.

To access it: open the document, go to File > Version history > See version history. A pane opens on the right with a vertical list of revisions, each with a timestamp and the editor's name. Google also lets you name versions, which is useful if you never did it during writing. You can name them now.

What to do with it:

1. Open the version history pane and identify the earliest revision. Google compresses minor edits into larger chunks, so the list is usually manageable.
2. Name the key versions. "First full draft", "After supervisor feedback", "Methods revision", "Final polish before submission." Names give the timeline a narrative a reviewer can follow in thirty seconds.
3. Screenshot or export the list. Google Docs does not have a one-click export for the version list itself, so take a screenshot of the pane with the dates visible. If you need a document, paste the screenshots into a single page and label them.
4. Open two or three versions and compare. Google shows what changed in each revision by highlighting deletions and insertions. Note the scale of each revision. A version that added three paragraphs to the discussion section looks different from a version that fixed a typo.

The strength of Google Docs here is that the history is on by default and cannot be turned off by the user. If you wrote the paper in Google Docs, the history exists. The only question is whether you have looked at it.

Version history is the centre of this but it is not the whole of it. Bring the surrounding material.

• The earliest draft you still have, even if it was rough. A rough draft is evidence of a writing process, and its date is evidence the document existed before the final version.
• Notes, outlines, or annotated PDFs you produced while working. These are the residue of research, not of generation. Their dates trace a path that does not look like a single text output.
• Email threads with your supervisor or anyone who gave feedback. If you sent a draft on a specific date and got comments back, that exchange anchors a point on the timeline.
• The AI writing report as a PDF, if you have it. You are entitled to see it. Turnitin's FAQ states: "Please note, only instructors and administrators are able to see the indicator." The same page also says: "The AI writing detection indicator and report are not visible to students." That sounds absolute, but the next sentence changes it: "However, with the PDF download feature, instructors can download and share the AI report with students." So if nobody has given you the report, ask for the PDF. Not a screenshot of the percentage. The full report, because the full report shows you where the highlights sit, how many words were analysed, and whether the score is a real number or an asterisk.

All of this together does one thing. It moves the conversation from a claim about a number to a discussion of a document and its history. We walked the general version of this in flagged but you wrote it.

You do not need a slide deck. You need one page with a timeline on it.

1. Draw a horizontal line. Mark the date of the earliest draft on the left, the submission date on the right.
2. Put the version history dates on the line. Not every version. Five or six. The ones where something changed.
3. Put the supervisor emails and feedback dates on the same line. These anchor the version history to events that happened outside the file.
4. Put the Turnitin report generation date on the line. This is the date the score was produced, not the date you submitted. They are sometimes different.
5. Write one sentence under the line. Something like: "The file was edited across [N] sessions between [date] and [date]. Supervisor feedback was incorporated on [date]. The AI report was generated on [date]."

That page does two things. It gives the reviewer something to look at besides the score, and it shows you came to participate in a review, not to defend against an accusation.

One more thing before you walk in. Turnitin's FAQ lists characteristics of text that tends to produce false positives:

"Sometimes false positives (incorrectly flagging human-written text as AI-generated), can include content without a lot of structural variation, text that literally repeats itself, or text that has been paraphrased without developing new ideas."

The next sentence is the one that matters in your meeting:

"If our indicator shows a higher amount of AI writing in such text, we advise you to take that into consideration when looking at the percentage indicated."

Read what that covers. Content without a lot of structural variation. Text that literally repeats itself. Text that has been paraphrased without developing new ideas. Those describe a large share of academic writing, especially literature reviews and methodology sections. If your flagged sections are in that territory, the company that made the tool is telling the person reading the score to discount it for exactly that reason. Not you. The company.

Three things will undercut the evidence you have.

Do not edit the version history before showing it. Google Docs revision names can be changed, but the timestamps and the content of each revision cannot. If you rename a version "First draft" and the timestamp shows it was created the night before submission, the rename makes things worse. Let the history speak.

Do not claim version history disproves AI use. It does not. It shows the document was edited over time in a pattern consistent with human writing. That is what it shows. Claiming more than that hands the other side an easy rebuttal.

Do not bring the old "first and last sentences" argument as if it were a current flaw. Turnitin published a release note in May 2023 acknowledging that false positives had been concentrated in the first and last few sentences:

"Since launch, we have observed a higher incidence of false positive detection in the first few or last few sentences of a document. Many times these sentences consist of introduction or conclusion content written in a generic way. As a result, we have changed our detection logic to help reduce these false positives."

The key phrase is the last one: "we have changed our detection logic to help reduce these false positives." This was a fix announced in 2023. It is not a current weakness you can point to. If you bring it as evidence that the tool is currently broken, the person across from you can look it up, find the fix, and your credibility on everything else drops. We went through this in more detail in is 20% AI too high.

HumanPen is a document rewriter. We are not going to tell you that running your paper through a rewriter will make an AI score disappear, because we do not make claims about future scores. What we can say is what the tool does: you upload the document with its AI writing report, and only the passages the report flagged are rewritten. The rest of the document stays verbatim. You review the output before it becomes anything final.

If you want to go into the meeting with the original document and let the version history speak, that is one path. If you want to revise the flagged sections beforehand, that is another. Both are legitimate. The meeting comes first.

Does version history prove I did not use AI? No. It shows that the document was edited over time in a pattern consistent with human writing. It cannot prove a negative. What it does is give the reviewer something to look at besides the score, and that is its value.

What if I did not have autosave on in Word? You lose the version history pane, but you do not lose everything. File modification dates on earlier drafts, email attachments sent to your supervisor, and cloud-synced backup folders all carry timestamps. They are not as clean, but they tell the same story.

My professor only sent me a screenshot of the percentage. What do I ask for? The full AI writing report as a PDF. Turnitin's FAQ says students cannot see the indicator, but also says instructors can download and share the report through the PDF download feature. Ask for that. Not the screenshot. The report.

What if the report was generated months ago? That matters. Turnitin's AI writing detection model page carries dated release notes. The 9 August 2023 note fixed a bug that highlighted references. The May 2023 note changed detection logic for first and last sentences. A report generated before those dates was produced under different logic than one generated today. The generation date is on the report. Find it before you discuss the score.

Will running my paper through a rewriter fix the score? We do not make claims about future scores. What HumanPen does is rewrite the flagged passages you select. Whether the next report comes back lower is something we will not promise, because the model changes and the report is produced by Turnitin, not by us.