Google has been stamping Gemini's text with SynthID since 2024, an invisible watermark woven into the words themselves. It's on by default in the Gemini app and on the web, and there's no setting to switch it off. StealthGPT's Watermark Remover strips it out at the sentence level, not with a synonym swap, so your words stay yours and the signal doesn't survive the rewrite. Paste in a draft, get back clean output, and publish. It's that easy.
Words limit: 0 / 500
Everything you need to catch, strip, and verify Gemini's SynthID mark, whether it's one page or your whole publishing queue.
One document at a time. Paste in text straight from Gemini, get back a structural rewrite with SynthID gone and the meaning intact. No account setup required to try it, and no limit on how many times you run a piece through before you're happy with it.
No guessing whether a piece is actually clean. Every rewrite comes back with a detection score attached, checked against the leading AI detectors, so you know before you hit publish instead of finding out after.
Route Gemini-generated content through the Watermark Remover API before it hits your CMS or publish button. Built for teams generating faster than a human editor can rewrite by hand, with no manual review pass per document and turnaround fast enough to sit inside a real production pipeline.
Wait, what's this?
SynthID is Google DeepMind's watermark for AI content, and in text it works nothing like a logo in the corner of an image. As Gemini writes, it scores every possible next word, and SynthID uses a secret key to tilt those scores in a pattern only Google's detector can recognize. Nothing is added to the page: no hidden characters, no metadata. The mark lives in which words Gemini chose, so it travels with the text through copy and paste and holds up through cropping, a few changed words, and mild paraphrasing.
Google's own developer documentation outlines where it breaks: detector confidence drops sharply when text is thoroughly rewritten or translated, and the mark is weaker on short, factual answers where Gemini has fewer word choices to play with. Anthropic's Claude Watermark is also reportedly derived from SynthID text. The Watermark Remover is built for this purpose. It doesn't swap synonyms and hope. It rewrites sentence structure the way a real edit would, which is the type of change SynthID isn't built to survive.
Google built SynthID to survive a few changed words and mild paraphrasing, which is exactly what synonym swapping does.
Rewrite depth
Invisible-character cleanup
Detection score
Volume
Update cadence
Output readability
Still have questions?
Learn more in our support page.
SynthID is Google DeepMind's invisible watermarking technology for AI-generated content, covering images, audio, video, and text. In text, it marks what the Gemini app and web experience write so Google's detector can later identify a passage as Gemini-generated. It's on by default, and there's no setting in Gemini to turn it off.
Gemini writes one word (token) at a time, scoring every possible next word by how likely it is. SynthID uses a secret key to affect those scores in a consistent pattern, so the words Gemini picks carry a statistical signature across the whole passage. Nothing is added to the text itself, and it reads exactly like normal Gemini output. A detector holding the key can then score any passage as watermarked, not watermarked, or uncertain.
Light editing usually does not remove the watermark. Google designed SynthID to survive copy and paste, cropping, a few changed words, and mild paraphrasing. Google's own documentation says detector confidence drops sharply when text is thoroughly rewritten or translated. That's what the Watermark Remover is built for, rewriting structure and not just vocabulary.
Yes. StealthGPT's Watermark Remover rewrites Gemini output at the sentence and clause level, which is the thorough rewriting Google says SynthID doesn't hold up against, while keeping your meaning and voice intact. Every rewrite comes back with a detection score so you can confirm it's clean before you publish, and teams can run the same process at volume through the StealthGPT API.
One person with one document, a team cleaning up internal documentation, or a whole content pipeline built on Gemini, SynthID comes out the same way. Enterprise options are available for teams that need it running quietly in the background.
