Nano Banana now lets you draw prompts directly on your photos — Sam Altman is going to hate this

When Nano Banana first launched, it almost broke the internet with viral trends. Now, Google is back at it again with a whole new update for its AI image generator, powered by Gemini 3.0 — and it might change how you think about photo editing altogether.

Just days after OpenAI updated ChatGPT Images, Google seems to be taking the wind out of Sam Altman's sails once again. Because now, users can draw or annotate directly on their images inside the Gemini app. With very little effort, users can show the AI exactly what to do, where to do it and how you want it to look. In other words: you don’t need to describe your vision in words anymore. You can literally sketch it.

Google AI Labs has implemented this strategy before on its lesser known AI tool called, Whisk. But adding this feature to Nano Banana, is the ultimate game-changer.

What this meeans for editing with AI

Now, instead of typing, “Remove this person in the background” or “Make the sky brighter on the left side,” you can circle the exact spot and let Gemini handle it. This completely changes AI editing by putting the power in the hands of every user — literally.

Visual prompting like this is a big deal, especially for anyone who gets easily frustrated trying to describe small visual tweaks or doesn't want to rewrite a prompt five times just to get the right crop or object removed.

Ultimately, this update means users can now do precision editing without Photoshop skills, which turns Gemini into a kind of a unicorn. This is something competing tools like ChatGPT and even Photoshop Generative Fill don’t fully offer yet — at least not with this level of mobile-first simplicity.

And yes, it works right inside the Gemini app, not just in the web UI.

Potential use cases

When users have the ability to draw their edits and bring them to life, things get exciting fast. This type of editing could be useful for:

• Fixing headshots: Circle the lighting or background you want changed
• Product mockups: Annotate layout shifts or add callouts
• Social posts: Touch up or blur things before posting
• Quick memes or edits: Add/remove stuff without opening multiple apps

How to try it

• Download the Gemini app or go to: http://gemini.google
• Select “Create image”
• Start by uploading an image you want to edit and tap on it to draw or annotate what you want directly on the image.
• Although no prompt is needed, you can always add one if you'd like.
• Once you're ready, hit submit to apply your edits.

Bottom line

Causual users, non-designers and anyone who wants to make edits on the fly — without design skills or even knowing the right terminology, can make it happen.

Gemini’s new image annotation feature brings a whole new layer of control to prompting. You don’t need to be good at describing edits anymore, you just need a finger and a vision.

Follow Tom's Guide on Google News and add us as a preferred source to get our up-to-date news, analysis, and reviews in your feeds.

More from Tom's Guide

• Google just launched a new AI agent that runs your inbox: Here’s what CC actually does
• I almost bought an e-paper tablet — ChatGPT already does exactly what I need
• 5 reasons the Google app is a must-download on your iPhone — especially before your next flight