What we do, and what we do not

Drafting with AI is now ordinary, and that is not where the argument is. The argument is over two things: whether the result is represented honestly, and whether the author still answers for it.

We will not pretend about one fact: the humanizer is used to lower the scores detection tools return, and that is what many people pay for. Denying it would cost this page whatever credibility it has.

But what the tool can and cannot do is a design decision, and those boundaries are in the code rather than in a statement. It does not write - every feature takes a document you have already written, and there is no "draft this for me" entry point. It does not alter facts - numeric values, units, percentages, statistical notation, exhibit numbering, in-text citations, and reference entries are protected during a rewrite, and citation format correction goes further by freezing the body outright. And it does not make your disclosure decision.

Universities and journals differ widely on AI-assisted writing, and the rules keep moving. Some prohibit it, some permit it with disclosure, some restrict particular stages.

We do not interpret those rules, and we do not declare any use "fine". Your institution’s academic integrity policy and your target journal’s author guidelines are the authority. Our journal-submission page says so already - not as cover, but because we genuinely do not hold that judgement.

One specific note: if your institution requires disclosure of AI use, a manuscript processed here is still within that requirement. The tool changes how something is worded; it does not change the fact that AI was involved.

We do not promise you will pass any detector. Detection systems keep updating; text that passes today may not after the next iteration. A "guaranteed to pass" or "refund if detected" promise is either hedged into near-unusability or is a promise about something outside the promiser’s control. It is common in this category. We do not make it.

We do not run our own detector. Setting your own exam and grading it is not persuasive. Instead we accept the Turnitin or iThenticate report your institution issued - the data comes from your side, and we only locate and process.

We do not claim head-to-head results. We have run no controlled comparison against other tools, so no page here says our pass rate is higher. Judge it on your own document.

These violate our terms: submitting work as someone else, or completing an identity-verified assessment on another person’s behalf; processing documents you neither own nor are authorized to handle; circumventing checks tied to identity, credentials, or professional licensing; and producing or processing material for fraud, defamation, or harassment.

On ghostwriting: we provide document processing, not writing services, and we do not partner with writing services. The input must be your own manuscript.

Where we find prohibited use, we act under the terms of service. What that means is what the terms say - this page will not list enforcement we have not built.

Uploaded documents and detection reports are used only for the job you submitted. A report locates the passages needing work and serves no other purpose - not training, and not shared with third parties.

Files are retained for 7 days. After that they stop being downloadable; the job record and history entry remain, because that is your own account history. The window is enforced by a scheduled sweep, not by hand.

Fuller detail on data handling is in the privacy policy.

This page changes as the product and the norms do. If you think a line is drawn wrongly, or that something we do contradicts what is written here, write to support@humanpen.net - especially the second case.