How to Turn Off Watermark in Gemini for Free? (Step by Step)

That sparkle in the corner of your AI-generated image.

You know the one. It sits there like a permanent reminder that what you just made came from a machine, not your own hand.

For designers, marketers, and anyone who values clean visual assets, that little star has been a persistent annoyance since Gemini first started generating images.

Google finally heard you.

On August 14, Google rolled out a new setting that lets you turn off visible watermarks on AI-generated content across Gemini.

No more cropping.

No more third-party removal tools.

Just a simple toggle that gives you control over whether your creations carry that identifying mark.

What the Gemini Watermark Actually Is

Before we walk through the settings, let's clarify what we are dealing with.

When you generate an image using Gemini's Nano Banana model, create a video with Omni, or produce music with Lyria, Google automatically adds a visible watermark.

• For images, it appears as a small sparkle or star icon in the bottom right corner.
• For videos, the watermark persists throughout playback.
• For music, it shows on the visual player associated with the track.

This visible mark was designed for one purpose: quick, at-a-glance identification of AI-generated content. Anyone viewing the file can immediately tell it came from Gemini.

But here is where it gets interesting. The visible watermark is only half the story.

The Invisible Layer That Never Goes Away

Turning off the visible watermark does not make your content untraceable. Google keeps two invisible identification systems embedded in every file you generate, regardless of your settings:

SynthID is an invisible digital watermark that Google developed specifically for AI-generated content. It embeds identification signals into images, videos, and audio without affecting how they look or sound. You cannot see it, but Google's verification tools can detect it.

C2PA Content Credentials provide metadata about the content's origin. This is essentially a digital nutrition label that travels with the file wherever it goes. It records when and how the content was created.

This dual-layer approach means anyone can still verify whether a file came from Gemini. Upload an image to Gemini and ask if it was AI-generated. Use Google Search's "About this image" feature. Run it through Lens. The invisible markers remain detectable even when the visible sparkle is gone.

Josh Woodward, Google's VP of Google Labs, put it this way:

"We're striking a balance here between creative control and safety: while the visible watermarks are now optional, invisible SynthID watermarks and C2PA metadata are still being used for transparency".

How to Turn Off the Gemini Visible Watermark (Step by Step)

The process is straightforward. Google designed this setting to be accessible in just a few clicks.

Here's how:

On the Web (Desktop)

1. Open Gemini in your browser and log in to your Google account
2. Look for the cog icon (gear/settings) in the lower left corner of the interface
3. Click on it to open the Settings menu
4. Find and select Media Watermark from the options
5. You will see two choices: On and Off
6. Toggle it to Off to disable visible watermarks on all future content

The setting applies globally. Everything you create going forward from images to videos to music will carry no visible watermark.

On Mobile (App and Browser)

As of August 2026, the toggle may not be available in the mobile apps immediately. Google is rolling it out across platforms, and it should appear in mobile settings soon.

When it does arrive, the steps will be similar:

1. Open the Gemini app or navigate to gemini.google.com on your mobile browser
2. Tap the menu icon (three lines or similar) in the left corner
3. Select Settings from the menu
4. Find Media Watermark in the settings list
5. Choose Off to disable visible watermarks

Just keep an eye and whenever the new update of Gemini app comes, update it to see above settings.

On Flow (Google's Video Editor)

The same Media Watermark setting applies to Flow, Google's AI video editing tool. If you generate videos through Flow using the Omni model, you can toggle the visible watermark on or off through the same settings path.

Important Limitations You Need to Know

Not everyone gets this option. Google has placed several restrictions on who can turn off the visible watermark.

Geographic Restrictions

If you live in India, South Korea, or Vietnam, you will only see the Media Watermark toggle if you have a paid AI Ultra subscription. Free users in these regions cannot disable the visible watermark.

Countries in the European Union may also face restrictions due to the EU AI Act, which requires clear labeling of AI-generated content. Google has stated that visible watermarks remain mandatory in countries where the law requires them. China is another country where the option is not available. Pakistani users of Gemini can also turn off watermarking.

Work and School Accounts

If you use Gemini through a Google Workspace work account or a school/education account, the Media Watermark setting does not appear at all. Visible watermarks remain mandatory for these accounts regardless of geographic location.

Rollout Timing

The feature is rolling out over several days starting from August 14, 2026. If you do not see the Media Watermark option in your settings yet, your account simply has not received the update. Check back in a few days.

Existing Content

The toggle only affects content you generate after changing the setting. Any images, videos, or music you created before turning off the watermark will keep their visible marks. You cannot retroactively remove watermarks from old files through this setting.

Why Google Made This Change?

This did not come out of nowhere. Google VP Josh Woodward asked users for their biggest Gemini frustrations, and thousands responded. Removing the visible Nano Banana watermark landed in the resulting top 10 requests.

Woodward described the tweak as a "papercut fixed", suggesting that the overlaid logos had been frustrating users who wanted a clean AI content creation workflow.

Credit: Google/X

The move also puts Gemini on more equal footing with competitors like ChatGPT, which does not apply visible watermarks to its generated images.

What This Means for Your Workflow

For designers, this is significant. You no longer need to crop, edit, or use third-party tools to remove watermarks from images intended for presentations, social media, or client work. The setting streamlines your workflow and saves time.

For content creators, the change offers flexibility. You can decide on a case-by-case basis whether you want the visible AI marker or prefer a clean output. Need to demonstrate that content came from AI? Leave the watermark on. Want a polished final product? Turn it off.

For transparency advocates, the invisible layers remain. The content is still traceable. The verification tools still work. Google has not abandoned accountability. They have simply moved the label from the surface to the background.

The Tradeoff

Visible watermarks let anyone instantly recognize AI-generated content without any special tools. Invisible systems require someone to actively check. That distinction matters.

When you turn off the visible watermark, you are making a choice. The content becomes harder to identify at a glance. People scrolling through social media or viewing a presentation may not realize an image came from AI unless they take the time to verify it.

Google's position is that the verification tools remain available and accessible. You can upload a file to Gemini and ask if it was AI-generated. You can use Search tools. The capability exists. But the responsibility for checking shifts from passive observation to active inquiry.

Looking Forward

Google has open-sourced Credentio, a C++ library that lets developers verify C2PA credentials locally within their applications. This means verification capabilities can be built directly into other platforms and tools, making it easier to check content origins without relying on Google's servers.

The company is also working on expanding content-credential generation and embedding capabilities. The infrastructure for AI content verification is growing, even as the visible markers become optional.

The Media Watermark toggle represents a broader shift in how AI companies think about labeling. Instead of forcing every user to display an obvious marker, Google is putting more weight on behind-the-scenes verification. The label is still there. You just have to know where to look.

So, you now have control over that sparkle in the corner.

The setting exists.

The toggle works.

Go turn it off if that is what you need.

Just remember: the invisible markers stay. The content remains traceable.

The choice is yours, but the accountability does not disappear.