Gemini Nano Banana Watermark Removal May Arrive in Future Update

Google May Let Users Remove Gemini Nano Banana Watermark in Future Update

Google appears to be working on a new option that could allow users to remove the visible watermark from images created with Gemini Nano Banana watermark technology, according to findings from an APK teardown of the latest Google app. While the feature has not been officially announced, early evidence suggests Google is testing a dedicated watermark setting without affecting the invisible security systems used to identify AI-generated content.

If introduced, the feature would address one of the most common complaints from Gemini users while preserving Google’s existing methods for tracking AI-generated images.

Google App Update Hints at New Watermark Setting

The discovery comes from an APK teardown of Google app version 17.43.11, where developers identified references to a new “Watermark settings” option.

The setting reportedly appears alongside existing account controls within Gemini and is also linked to the internal codename “Robin,” which is associated with Gemini’s development.

Because the feature has only been found in application code, there is no confirmation that it will be released publicly or when it could become available.

Google has not officially commented on the feature.

User Feedback May Have Influenced Development

Earlier this month, Josh Woodward, Google’s Vice President responsible for Gemini, Google AI Studio, and Google Labs, invited users to share their biggest frustrations with the Gemini platform.

Thousands of users responded with suggestions, and the request to remove the visible Nano Banana watermark ranked among the ten most frequently mentioned improvements.

At the time, Woodward said the development team would review the feedback. Less than two weeks later, code discovered in the latest Google app suggests Google may already be exploring the requested feature.

Although the timing is notable, Google has not confirmed whether the development is directly connected to user feedback.

Visible Watermark Could Be Optional

The reported feature appears to focus only on the visible Gemini logo placed in the lower-right corner of AI-generated images.

Many users have argued that the watermark can cover small parts of an image, especially when artwork contains important visual details near the edge.

If Google enables the setting, users may be able to remove only the visible branding while leaving the actual image unchanged.

It remains unclear whether the option would apply only to newly generated images or if it would also work on images created before the feature becomes available.

Until Google makes an official announcement, those details remain uncertain.

Invisible AI Tracking Would Continue

Even if the visible watermark becomes optional, Google’s invisible content verification systems are expected to remain in place.

The reported change would not affect SynthID, Google’s digital watermarking technology designed to identify AI-generated content without altering its appearance.

Likewise, C2PA metadata, which stores image provenance information within supported files using EXIF-based standards, would reportedly continue functioning as normal.

These technologies help platforms, publishers, and researchers determine whether an image has been generated or modified using artificial intelligence, even when no visible logo appears on the image.

Maintaining these systems would allow Google to continue supporting transparency and content authenticity while giving users more flexibility over image presentation.

Industry Trend Is Moving Away from Visible Watermarks

Google would not be the first AI company to move away from visible branding on generated images.

Several competing AI image generation platforms have already eliminated prominent visible watermarks, allowing users to download clean images while relying on metadata or invisible watermarking for authenticity.

Platforms such as ChatGPT, Midjourney, and many Stable Diffusion-based image generators no longer place visible branding directly on exported images.

Google’s own AI Studio has also reportedly stopped applying visible watermarks to generated content, making Gemini one of the remaining Google AI products that still displays a logo by default.

If Gemini adopts a similar approach, it would bring greater consistency across Google’s AI ecosystem.

Why the Feature Matters

Visible watermarks have long divided AI users.

Some creators appreciate visible branding because it clearly identifies AI-generated content.

Others argue that large or fixed-position logos can interfere with design work, presentations, social media graphics, marketing materials, or creative projects where image composition matters.

Google appears to be balancing both concerns by potentially removing only the visible logo while continuing to rely on invisible verification technologies.

This approach would preserve AI content traceability without affecting the visual quality of generated images.

As governments and technology companies continue discussing standards for AI transparency, invisible watermarking systems such as SynthID are becoming increasingly important.

Removing visible branding while keeping underlying verification methods intact may offer a compromise between user convenience and responsible AI practices.

Google Has Yet to Confirm Release Plans

At this stage, the reported watermark option exists only within code discovered through an APK teardown.

Features identified during application teardowns often provide an early indication of development, but they do not guarantee a public release. Companies frequently test or remove experimental features before they become available to users.

Google has not announced a launch date, pricing changes, or availability for the reported watermark setting.

Users will likely need to wait for an official confirmation before knowing whether the feature will be included in a future Gemini update and how it will work across existing and newly generated images.