New engine: rewrites word by word
Sounds like you wrote it. Says what you meant.
Samesay rewrites AI drafts into plain, natural prose, then checks that every figure, name and claim survived. Free account, no card needed.
Medium: closest to your draft, 3 passes of the HIP engine over your text. Medium keeps the most meaning; Max changes the most and needs a careful read. Versions are ranked by the meaning check, best first. English prose only; headings stay as they are.
- Human-like score on every rewrite
- Meaning check flags what changed
- Headings kept as they are
- Never trained on your text
How it works
From draft to your voice in four steps
Bring your draft
Paste anything: an essay from a chat assistant, a product page, a report. English drafts work best.
Pick a strength
Medium stays close to your draft, Extra goes further, Max rebuilds it. Ask for up to three versions, ranked by the meaning check.
Read the scores
Your rewrite opens with its human-like score, a style score and a check on every figure, name and qualifier.
Copy with confidence
Show what changed against your draft, fix anything you disagree with, then copy.
See the difference
Same facts. A voice people want to read.
Pick an example. The left side is a typical assistant draft, the right side what Samesay does with it.
I hope this email finds you well. I wanted to reach out to provide a comprehensive update regarding the Q3 launch. Moreover, it is important to note that the timeline has shifted by two weeks due to the fact that vendor approvals took longer than anticipated. Furthermore, the budget remains at $48,000, and we are confident that this adjustment will ultimately facilitate a seamless rollout.
Quick update on the Q3 launch: we are now two weeks behind, because vendor approvals took longer than we expected. The budget has not changed and stays at $48,000. We think the extra time will make the rollout smoother.
Every figure and qualifier kept
Examples written by our team to show the kind of change. Scores from our AI checker; your results will vary.
Why Samesay
Built for people who care what their words say
Meaning check on every rewrite
Figures, units, names, negations and words like "may" are compared with your draft. Anything lost or changed is listed before you copy.
A style score you can read
See which sentences still lean on stock phrases, flat rhythm or heavy punctuation, and why. It is a writing check, not a detector verdict.
One engine, a published method
HIP (Carnegie Mellon, 2026): a base model with the adapter its authors released rewrites your text in passes, each one starting from the last. Pick a strength, Medium, Extra or Max, and up to three versions.
Headings stay put
Titles and section headings are kept exactly as you wrote them. Only the prose under them is rewritten.
Built for English
The engine is tuned on English prose. Paste English drafts for the best results; other languages come later.
Never trained on your text
Your drafts are used for your rewrite and nothing else. Turn history off and they are gone within the hour.
See what changed
One click shows every word added and removed against your draft, so nothing slips through unnoticed.
An API for your pipeline
The same engine, limits and checks over a single REST endpoint on Pro and Ultra.
Use cases
For every kind of draft you write
Emails and reports
Status updates, proposals and summaries that sound like a colleague wrote them, not a template. Figures, dates and names stay exactly as you had them.
Tip: Ask for 2 or 3 versions, then pick the one closest to how you talk.
The basics
What is an AI humanizer?
A humanizer is a rewriting tool for text that sounds machine-made. It keeps the content and changes the delivery: wording, rhythm, sentence shape, punctuation. A good one is closer to a sharp editor than to a thesaurus.
What it changes
Stock phrases, filler connectors, sentences that all sound alike, and punctuation habits. It does not change your facts.
Who uses one
Marketers, founders, support teams, researchers writing summaries, and anyone drafting in a second language.
What it is not
It is not a fact checker, it is not a way around rules you agreed to, and it cannot guarantee any detector result. Anyone promising that is guessing.
Stock phrasing
Openers and filler that every assistant uses and every reader recognises.
In today's fast-paced world, it is important to note that…
Today, …
Flat rhythm
Sentence after sentence of the same length and shape. People write with more variety.
Sentences of 18, 19, 18 and 20 words in a row.
A short one. Then a longer one that carries the detail.
Connector overload
Moreover, furthermore, additionally: links that announce logic instead of showing it.
Moreover, the team… Furthermore, the budget…
The team… The budget, meanwhile, …
Dash habits
Em dashes everywhere. We use commas, colons, brackets or a new sentence, and keep ranges like 2023–2025.
The plan — once approved — will start.
Once approved, the plan will start.
Our promises
What you can hold us to
We show what changed
If a figure, name or qualifier goes missing, you see it flagged next to the rewrite, and one click shows every word that moved.
We do not sell detector magic
Detectors disagree and change every month. We will never promise a score, a grade or a pass.
Your words stay yours
No training on your text, no sharing with detector companies, delete everything in two clicks.
Pricing
Simple plans, priced by the word
Start free with 1 500 words a month. Upgrade when you need more.
Basic
$10 / month
or $80 a year
30,000 words a month, up to Extra strength and 2 versions per rewrite, 600 words per text.
Pro
$20 / month
or $160 a year
60,000 words a month, up to Max strength and 3 versions per rewrite, 1,500 words per text, API.
Ultra
$50 / month
or $400 a year
100,000 words a month, up to Max strength and 3 versions per rewrite, 3,000 words per text, API.
For developers
The same engine, one endpoint
Send a draft, get the rewrite back with its human-like score, style score and meaning flags. Same limits and words as your plan, nothing stored in your history.
Read the API guidePOST /v1/humanize
Authorization: Bearer hz_live_…
{ "text": "Your draft…", "rounds": 5, "n": 2 }
→ { "outputs": [ { "text": "…", "human_score": 74, "style_score": 94,
"meaning_ok": true, "meaning_flags": [] } ] }From the blog
Notes on writing with AI
FAQ
Questions people ask
What does Samesay actually do?
It rewrites a draft, often one produced with an AI assistant, so it reads like careful human writing, then checks that the rewrite still says the same thing. Every rewrite comes with a human-like score, a style score and a meaning check.
Will it get my text past AI detectors?
We do not promise that, and nobody honestly can. Detectors disagree with each other and change often. What we do is remove the habits readers notice and keep your meaning intact. Use it for writing you are allowed to get help with.
Is it allowed for school or university work?
That depends on your institution and your teacher. Many allow AI help for some tasks and not others, and some ask you to disclose it. Check the rules for your assignment first. Our terms ask you to follow them.
What is the style score?
A 0 to 100 measure of writing habits, from our own rules: stock phrases, flat sentence rhythm, dashes, filler connectors and similar. Higher means fewer habits. It is not a prediction of what any detector will say.
What does the meaning check look at?
Numbers and units, names, negations, qualifiers like "may" or "associated with", and large changes in length. It errs on the side of flagging, so read the flags before you copy.
What do Medium, Extra and Max mean?
The strength of the rewrite. The engine rewrites your text in passes, each starting from the last: Medium is 3 passes and stays closest to your draft, Extra is 5, Max is 10 and rebuilds the text (the setting of the research paper, and the one that drifts the most). Versions are independent runs at the same strength, up to three, ranked by the meaning check so the most faithful comes first. Free includes Medium, Basic adds Extra, Pro and Ultra add Max.
Which languages are supported?
English for now. The engine is tuned on English prose, so drafts in other languages may come back in English.
Do you store my text?
With history on, your drafts and rewrites stay in your account for up to 90 days so you can come back to them. With history off, neither the draft nor the versions are stored at all. We never train models on it.