--- title: Google lets users turn off visible AI watermarks, keeps... description: Google is rolling out a toggle to remove visible watermarks from AI-generated images, video, and music made with Nano Banana, Omni, and Lyria, starting in... canonical: https://daily.dev/posts/google-lets-users-turn-off-visible-ai-watermarks-keeps-hidden-ones-intact-ejapr6jj5 twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Google lets users turn off visible AI watermarks, keeps hidden ones intact | daily.dev og:description: Google is rolling out a toggle to remove visible watermarks from AI-generated images, video, and music made with Nano Banana, Omni, and Lyria, starting in... og:url: https://daily.dev/posts/google-lets-users-turn-off-visible-ai-watermarks-keeps-hidden-ones-intact-ejapr6jj5 og:image: https://api.daily.dev/og/posts/EjApR6jJ5.png og:image:alt: Google lets users turn off visible AI watermarks, keeps hidden ones intact og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Google lets users turn off visible AI watermarks, keeps hidden ones intact **[Collections](https://daily.dev/sources/collections)** · 2 min read · 0 upvotes · 0 comments ## Summary Google is rolling out a toggle to remove visible watermarks from AI-generated images, video, and music made with Nano Banana, Omni, and Lyria, starting in Gemini and Flow, with Search support to follow. The invisible SynthID watermark and C2PA provenance metadata remain embedded regardless, so content can still be flagged as AI-generated later. Google is also open-sourcing Credentio, a C library for building local C2PA validation into third-party apps. The move contrasts with Anthropic's recent decision to watermark Claude-generated text to comply with EU rules. ## Content Google is rolling out a setting that lets people remove the visible watermark stamped on AI-generated images, videos, and music. The toggle covers content made with Nano Banana (images), Omni (video), and Lyria (music), and it's showing up first in Gemini and Flow, with Search support coming later. You'll find it under Settings > Media Watermark. Worth noting: this only applies to media, not text. Claude's recent move to watermark AI-written text for EU compliance is a separate story. Here's the part that matters most: turning off the visible mark doesn't touch the invisible stuff. SynthID, Google's imperceptible watermark, stays embedded in the file whether you can see a logo or not. It's designed to survive the usual abuse - resizing, recoloring, compression - so Google (and tools built on Gemini or Search) can still flag the content as AI-generated later. C2PA metadata, the industry-standard provenance record, also stays put. So what actually changes? Just the visible cue. That little watermark was useful as an at-a-glance signal, but it also sat right on top of content people wanted to publish, print, or edit without a logo baked in. Now you get to strip it while the underlying provenance tracking keeps working in the background. Whether that's a reasonable trade-off or a step that makes AI content harder to spot at a glance is something I think reasonable people will disagree on. Alongside this, Google open-sourced Credentio, a C library (also available as C++) for validating C2PA Content Credentials directly inside an app, no cloud round-trip required. It's not a new project cobbled together for this announcement - it's the same code already running in roughly 40 Google products that collectively handle tens of billions of media files. Supporting C2PA spec versions 2.2 and 2.4, Credentio does local validation with a small memory footprint, which matters if you're deploying on edge devices, servers, or anything resource-constrained. Skipping the cloud also sidesteps the privacy and latency issues that come with sending media off for verification. Put together, the two moves tell a fairly consistent story: Google wants to give people more control over how their AI-generated content looks in public, while making sure the ability to trace that content back to an AI model doesn't go away. Whether that balance holds up once visible watermarks become optional at scale is the open question. ## Questions this post answers ### Can I remove the visible AI watermark on images made with Google's Nano Banana or video made with Omni? Yes, Google added a setting under Settings > Media Watermark in Gemini and Flow that lets users turn off the visible watermark stamped on content made with Nano Banana (images), Omni (video), and Lyria (music). Search support is coming later. The invisible SynthID watermark and C2PA metadata remain embedded in the file regardless of this setting. _daily.dev helps developers track how AI content labeling policies like this evolve across platforms._ ### Does turning off Google's visible AI watermark also remove SynthID or C2PA metadata from the file? No, disabling the visible watermark only removes the visual cue; SynthID, Google's imperceptible watermark, stays embedded and is designed to survive resizing, recoloring, and compression, and C2PA metadata also remains in the file so provenance tracking keeps working in the background. _Teams building content-verification pipelines can follow watermarking changes like this via daily.dev._ ### What is Credentio and which C2PA spec versions does it support? Credentio is a C library, also available in C++, that Google open-sourced for validating C2PA Content Credentials locally inside an app without a cloud round-trip. It supports C2PA spec versions 2.2 and 2.4, has a small memory footprint suited to edge devices and resource-constrained servers, and is the same code already running in roughly 40 Google products handling tens of billions of media files. _Developers evaluating on-device provenance validation tools can keep up with releases like Credentio on daily.dev._ ## Similar posts on daily.dev - [How we’re bringing AI image verification to the Gemini app](https://daily.dev/posts/how-we-re-bringing-ai-image-verification-to-the-gemini-app-ijxaog12c) · DeepMind · 0 upvotes · 0 comments - [Text AI watermarks will always be trivial to remove](https://daily.dev/posts/text-ai-watermarks-will-always-be-trivial-to-remove-borbz7x9u) · sean goedecke · 6 upvotes · 4 comments --- Tags: [#google](https://daily.dev/tags/google) [View this post on daily.dev](https://daily.dev/posts/google-lets-users-turn-off-visible-ai-watermarks-keeps-hidden-ones-intact-ejapr6jj5) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"Google lets users turn off visible AI watermarks, keeps hidden ones intact","url":"https://daily.dev/posts/google-lets-users-turn-off-visible-ai-watermarks-keeps-hidden-ones-intact-ejapr6jj5","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/google-lets-users-turn-off-visible-ai-watermarks-keeps-hidden-ones-intact-ejapr6jj5"},"datePublished":"2026-08-14T16:46:00.809Z","dateModified":"2026-09-13T19:25:46.242Z","description":"Google is rolling out a toggle to remove visible watermarks from AI-generated images, video, and music made with Nano Banana, Omni, and Lyria, starting in...","image":"https://pbs.twimg.com/media/HPshKsVaMAA_wSW.png","thumbnailUrl":"https://pbs.twimg.com/media/HPshKsVaMAA_wSW.png","isAccessibleForFree":true,"articleSection":"Collections","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Collections","logo":"https://media.daily.dev/image/upload/s--fk_6ycEi--/f_auto,q_auto/v1780996001/logos/collections?_a=BAMAMiWQ0","url":"https://daily.dev/sources/collections"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/google-lets-users-turn-off-visible-ai-watermarks-keeps-hidden-ones-intact-ejapr6jj5","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"google","timeRequired":"PT2M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Collections","item":"https://daily.dev/sources/collections"},{"@type":"ListItem","position":3,"name":"Google lets users turn off visible AI watermarks, keeps hidden ones intact"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/google-lets-users-turn-off-visible-ai-watermarks-keeps-hidden-ones-intact-ejapr6jj5#faq","mainEntity":[{"@type":"Question","name":"Can I remove the visible AI watermark on images made with Google's Nano Banana or video made with Omni?","acceptedAnswer":{"@type":"Answer","text":"Yes, Google added a setting under Settings > Media Watermark in Gemini and Flow that lets users turn off the visible watermark stamped on content made with Nano Banana (images), Omni (video), and Lyria (music). Search support is coming later. The invisible SynthID watermark and C2PA metadata remain embedded in the file regardless of this setting. daily.dev helps developers track how AI content labeling policies like this evolve across platforms."}},{"@type":"Question","name":"Does turning off Google's visible AI watermark also remove SynthID or C2PA metadata from the file?","acceptedAnswer":{"@type":"Answer","text":"No, disabling the visible watermark only removes the visual cue; SynthID, Google's imperceptible watermark, stays embedded and is designed to survive resizing, recoloring, and compression, and C2PA metadata also remains in the file so provenance tracking keeps working in the background. Teams building content-verification pipelines can follow watermarking changes like this via daily.dev."}},{"@type":"Question","name":"What is Credentio and which C2PA spec versions does it support?","acceptedAnswer":{"@type":"Answer","text":"Credentio is a C library, also available in C++, that Google open-sourced for validating C2PA Content Credentials locally inside an app without a cloud round-trip. It supports C2PA spec versions 2.2 and 2.4, has a small memory footprint suited to edge devices and resource-constrained servers, and is the same code already running in roughly 40 Google products handling tens of billions of media files. Developers evaluating on-device provenance validation tools can keep up with releases like Credentio on daily.dev."}}]} ```