My husband works in brand design and creative services, so it’s safe to say he’s been keeping an eye on the AI wave. Like a lot of creatives, he’s experimented with different tools but often ended up frustrated. The outputs were either too generic, too distorted, or just not useful in a professional context.

That changed when Google released Nano Bana (Gemini).

For the first time, he found a platform that wasn’t just about generating random images. It was about refining and editing existing ones. And in the world of brand and creative work, where the details matter, that’s everything.

The Test Project: Tote Bag Character

He wanted to create a set of tote bags featuring a quirky illustrated character for a work project. The character already existed, so the task wasn’t to invent something new. It was to reposition the image in the same style as the original reference.

With Nano Bana, he uploaded the reference, prompted a few variations, and watched the tool iterate. The first result was rough, but a couple of follow-up prompts got it closer and closer until it finally clicked (see prompts and output below). 

After a quick polish in Adobe Illustrator, adding text and cleaning up lines, he had a design that was ready to ship (see final version below).

Why Nano Bana Stands Out

There are plenty of image generation tools available today. What makes Nano Bana different is its ability to edit with precision. Instead of creating something entirely new (use Midjourney for this!), it can take an existing design and refine it into something more polished and usable.

That’s the kind of AI that has real staying power. Most corporate creative work is less about inventing from scratch and more about adjusting, iterating, and improving. A tool that helps do that quickly while preserving brand integrity isn’t just interesting. It’s genuinely useful.