New Gemini Update Personalizes AI Images with Nano Banana, Photo Integration

Gemini’s new image feature brings your own world into the frame.

Google is rolling out a Gemini update that uses Nano Banana 2 and Google Photos to generate images involving individual users. The tech titan said the new experience pulls from connected apps and personal context to make prompting less hands-on and outputs more relevant.

The move nudges Gemini further away from generic AI image creation.

Gemini gets a better sense of you

Google is building this around Personal Intelligence, a system that lets Gemini use a person’s existing preferences and connected Google activity as creative context for image generation. 

With Nano Banana 2 underneath it, Gemini becomes a tool with some built-in awareness of what matters to each user, one Google said should “feel tailored to you, not just a generic tool that works the same for everyone.”

That means users do not have to load every prompt with personal detail just to get something specific back. For example, prompts like “Design my dream house” or “Create a picture of my desert island essentials” can draw on connected-app context to produce something more personal. 

Your photo library becomes part of the prompt

With Google Photos in the mix, the personalization goes further. Gemini can use a person’s photo library as visual context for image generation, giving it access to actual images of the user, their family, and other people or pets already organized in Photos. 

The feature works with the labels users already apply to people and animals in their library, helping the AI chatbot identify who is who when building an image. For instance, users can ask Gemini to “create a claymation image of my family and me enjoying our favorite activity,” turning familiar faces and relationships into part of the prompt itself. 

Users stay in the driver’s seat

Gemini may not always get the right photo or detail on the first try. When that happens, users can tell the tool “what was incorrect and try again,” or choose a different reference image from Google Photos to guide the next result.

Users can also check how an image was built. The Sources button shows which photo Gemini selected automatically, and the app can provide attribution details for that specific creation.

Not everywhere, and not for everyone

The launch comes with privacy guardrails. Gemini does not directly train on a user’s private Google Photos library, and connected Google apps remain opt-in, with settings users can change at any time.

Access is limited, too. The personalized image experience is rolling out over the next few days only to eligible Google AI Plus, Pro, and Ultra subscribers in the US, with broader availability planned later.

Looking to elevate your images for LinkedIn or personal branding? These Gemini prompts can do the heavy lifting.