Does RewriteIQ bypass Pangram? Nobody has tested it — and the more useful finding is that the site’s own pages refute its own guarantee.
The only tool that bypassed Pangram and Turnitin in 2026.
Most humanizers clear one detector and get caught by the other. StealthWriter is the one that gets past both — run Ghost 5.2 Pro at level 7–8, section by section, and re-check before you submit.
Try StealthWriter →Key Takeaways
- RewriteIQ sells a “99.95% Bypass Success Rate” and a “100% Human Score Guarantee”. Two clicks away, its Responsible Use page says “100% Algorithmic Humanization is a Myth.”
- Its Refund Policy puts detector performance outside any remedy: refunds for dissatisfaction with “detector scores… are not guaranteed.”
- Pangram appears zero times across all 18 site pages, all 18 blog posts, and the separate rewriteiq.ai domain — while its own free detector page names GPTZero 30 times and Turnitin 33.
- The footer “Methodology” link is a dead button with no destination, and
/methodologyreturns a 404. The only method ever published is an eight-word footnote. - It sells “zero-knowledge processing… never stored or logged” while its Privacy Policy says it collects your inputs and outputs and that users cannot delete them.
- Zero Reddit posts and zero Reddit comments mention the product, all-time — against claims of 50M+ words a month and 800+ reviews.
- No legal entity, no jurisdiction, no address, and no governing-law clause anywhere. The Turnitin “proof” is a 659-byte icon-pack PNG.
The only tool that bypassed Pangram and Turnitin in 2026.
Most humanizers clear one detector and get caught by the other. StealthWriter is the one that gets past both — run Ghost 5.2 Pro at level 7–8, section by section, and re-check before you submit.
Try StealthWriter →Does RewriteIQ bypass Pangram?
The question has never been asked by anyone, including RewriteIQ. We fetched every URL in its sitemap — 18 pages — plus all 18 posts on its blog subdomain and the separate rewriteiq.ai site. Pangram appears zero times in any of them.
That absence is pointed rather than incidental. RewriteIQ runs its own free AI detector at /detector, and that page names GPTZero 30 times, Turnitin 33, Originality 27 and Copyleaks 12 — while claiming it will “ensure your content passes all major AI detection tools.” Pangram: nought.
This page reports no first-party test. Everything below was fetched on 28 August 2026.
Where does the site contradict itself?
The homepage sells “99.95% Bypass Success Rate”, “0% AI Detected — Perfect Score”, “100% Human Content” and a “100% Human Score Guarantee”. The bypass-Turnitin page asks “Can Turnitin detection be completely avoided?” and answers “Yes.”
The Responsible Use page says: “100% Algorithmic Humanization is a Myth. While our technology is advanced, no AI can perfectly replicate the depth of human experience 100% of the time. AI models are probabilistic, not flawless.”
The Refund Policy closes the loop by putting the outcome outside any remedy: “Refunds for dissatisfaction with output quality (e.g., detector scores, plagiarism results, or writing style) are not guaranteed.”
All three pages carry the same “Last Updated” date of 4 June 2026. A guarantee that the same site calls a myth, and that the refund policy specifically declines to honour, is not a guarantee.
Where is the methodology?
There is a Methodology link in the footer. It is a <button> with no destination, and /methodology returns a 404 and does not appear in the sitemap.
What does exist is a bar chart grading six named competitors — WriteHuman 60%, Phrasly 45%, GPTinf 38%, StealthGPT 34%, Walter Writes 32%, Undetectable.ai 28% AI detected — with RewriteIQ shown separately at 0%. The entire methodological disclosure is one footnote: “Tested across GPTZero, Turnitin, Originality.ai, and Copyleaks detection platforms • Jul 2026.”
A detector list and a month. No sample size, no corpus, no word count, no source model, no trial count, no per-detector breakdown, no raw data. That is the sum total behind 99.95%.
| Where the claim appears | What it says | Evidence |
|---|---|---|
| Homepage | 99.95% bypass success rate; 100% Human Score Guarantee | An eight-word footnote |
| Bypass-Turnitin page | “Can Turnitin detection be completely avoided?” — “Yes” | None |
| Responsible Use page | “100% Algorithmic Humanization is a Myth” | — |
| Refund Policy §4 | Detector scores “not guaranteed” | — |
| Footer “Methodology” | Dead button; /methodology returns 404 | — |
date_gmt values. We swept all 18 pages for the words that would identify a legal entity, jurisdiction or governing law. We searched the DAMAGE paper, Pangram’s August 2025 benchmark and the Pangram 4 report for RewriteIQ by name, and searched Reddit posts and comments all-time. Where a source has a commercial stake, we say so.What is on the site that could count as proof?
Nothing. We enumerated every image reference across all 18 pages and fetched each file.
The complete inventory is four testimonial avatars — which appear with no names, no titles and no review text attached, on a site claiming “4.8/5.0 based on 800+ Feedbacks” — plus four detector logo files and some vendor logos. There is not a single screenshot of a detector result page anywhere. The Turnitin “proof” is icons8-turnitin-48.png, a 659-byte 48-pixel icon-pack file.
Every number on the site — 99.95%, 0% AI, 100% Human, and all six competitor bars — is rendered as HTML text and CSS divs, not captured output.
And the privacy claim?
It is contradicted by the privacy policy. The partner page sells “Complete Anonymity. Zero-knowledge processing. Your content is never stored or logged”, and the bypass-Turnitin page repeats “We never store your data.”
The Privacy Policy, section 2.2: “We collect the content you input into our Service (‘Input’) and the content generated by the Service (‘Output’).” Section 5: “users cannot manually delete their usage history (Inputs and Outputs) from our systems.”
That is a zero-knowledge promise sold to people buying discretion, refuted by the operator’s own policy — and there is no operator to hold to it. Across all 18 pages there is no legal entity name, no jurisdiction, no registration number, no address and no governing-law clause of any kind. The Terms contain no choice of law, no venue and no arbitration clause, so there is no stated forum in which any of it could be enforced.
Has Pangram named it, and does anyone use it?
No, and apparently not. RewriteIQ is absent from the DAMAGE paper‘s 19 named tools and the August 2025 benchmark‘s 20. The Pangram 4 technical report of July 2026 tests 13 commercial humanizers anonymised as “Commercial A” to “Commercial M”, reporting 97.69% overall AI recall with the weakest system still caught 92.78% of the time. RewriteIQ might be one of the thirteen; nothing published lets anyone narrow it.
Reddit returns zero posts and zero comments mentioning the product, all-time. Not seeded promotion — nothing at all. No Chrome extension exists. The YouTube channel’s feed returns no videos, despite one being embedded on the homepage. The domain was registered on 12 November 2025 for a single year, and the entire blog is 21 days old and still carries its un-deleted WordPress theme demo pages.
What should you take from this?
RewriteIQ’s most useful document is its own Responsible Use page. It says a 100% result is a myth, and the refund policy says detector scores are not a remediable complaint. Read together with the homepage guarantee, that is a product whose strongest claim is disclaimed by its own publisher.
On Pangram specifically, the record is empty in every direction, and the honest answer is that nobody knows. What the record does support is Pangram’s published ceiling: 97.69% of humanized text caught across 13 unnamed tools, with the best evader still caught 92.78% of the time.
If the real worry is being wrongly flagged on your own writing, keep your drafting history — it costs nothing and beats any rewriting tool as a defence, and it matters most for the writers detectors treat worst: ESL writers and AI detection and false positives for neurodivergent students. Check before you submit with the best pre-submission check and the Turnitin self-check, and strip the obvious tells by hand first via what to remove before using an AI humanizer.
For the product itself see our RewriteIQ review. For the Pangram cells with real evidence, see Undetectable AI, StealthGPT, QuillBot, StealthWriter, Phrasly, Humbot, Rephrasy, Ryne AI, Clarity Bubble, SuperHumanizer, UnAIMyText, Text Polish, AuraWrite, HueWrite and AiTextools, plus the Pangram review.
The one tool that got past both
Pangram and Turnitin catch most humanizers — one or the other, and usually the one you did not think to test. StealthWriter is the only tool that bypassed both in 2026. Run Ghost 5.2 Pro at level 7–8, section by section, and re-check before you submit.
Free plan to start, no card. Or get the free prompt pack first →
Affiliate link — I earn a commission if you upgrade.
Frequently asked questions
Does RewriteIQ bypass Pangram?
Nobody has ever tested it, RewriteIQ included. Across all 18 pages of rewriteiq.com, all 18 blog posts and the separate rewriteiq.ai domain, Pangram appears zero times. It is absent from every Pangram publication, and no third party has run the pair. The question has simply never been asked.
But it claims a 99.95% bypass rate.
It does, on the homepage, alongside “100% Human Content”, “0% AI Detected” and a “100% Human Score Guarantee”. Its own Responsible Use page then says: “100% Algorithmic Humanization is a Myth. While our technology is advanced, no AI can perfectly replicate the depth of human experience 100% of the time.” Both pages are dated 4 June 2026.
Is there any methodology behind the numbers?
Eight words. The competitor benchmark chart carries one footnote: “Tested across GPTZero, Turnitin, Originality.ai, and Copyleaks detection platforms • Jul 2026.” No sample size, no corpus, no word count, no model, no trial count, no raw data. The footer’s “Methodology” link is a button with no destination, and /methodology returns a 404.
What does the refund policy actually cover?
Not detector performance. Section 3: “All payments are final.” Section 4: “Refunds for dissatisfaction with output quality (e.g., detector scores, plagiarism results, or writing style) are not guaranteed.” There is no 7-day, 14-day or 30-day window anywhere — the default is no refund — and the Terms note you will not be warned before a free trial converts to a charge.
Is RewriteIQ in Pangram’s published research?
No. It is absent from the DAMAGE paper’s 19 named tools and from the August 2025 benchmark’s 20. The July 2026 Pangram 4 report tests 13 humanizers anonymised as “Commercial A” to “Commercial M”, so presence there can be neither confirmed nor excluded — and nothing published by anyone can narrow it further.
What proof does the site publish?
None. There is not a single screenshot of a detector result page anywhere on the site. The image inventory is four unnamed testimonial avatars and four detector logo files — the Turnitin one being a 659-byte 48-pixel icon-pack PNG. Every figure, including the six competitor bars, is rendered as HTML text.
Is anyone actually using it?
Reddit returns zero posts and zero comments mentioning RewriteIQ, all-time. No Chrome extension exists. The YouTube channel’s feed returns no videos despite one being embedded on the homepage. This sits against claims of “50M+ words humanized monthly”, “10,000+ documents/month”, “15+ countries” and “4.8/5.0 based on 800+ feedbacks” — not one of which is attributable.
Are rewriteiq.com and rewriteiq.ai the same company?
They appear to be different products with opposite positioning. rewriteiq.com sells guaranteed Turnitin bypass to essay-writing agencies. rewriteiq.ai is a separate waitlist product registered in May 2026 that states: “Built for marketing, not academic fraud… Our terms of service prohibit use on coursework, theses, or any submission to an educational institution.” Every legal link on that second site returns a 404, including the terms it cites.
