Gemini Can Now Create Personalized Fake Images Using Google Photos

Google has introduced a new Gemini feature that allows users to generate more personalized images by combining Nano Banana 2 with their Google Photos library.

Users who connect Google Photos to Gemini can now create images with less detailed prompts. Instead of uploading reference photos manually, Gemini can use context from the user’s photo library.

This means prompts such as “design my dream house” or “create a picture of my desert island essentials” can generate results based on personal preferences and stored photo history.

If users have tagged people in Google Photos, Gemini can also generate more specific scenes.

For example, users can request an image such as “create a claymation image of my family and me enjoying our favorite activity,” and Gemini can use available context to identify family members and likely activities.

Google said results may not always match expectations on the first try, as the feature is new. Users can refine prompts, correct mistakes, or upload reference photos if needed.

Gemini also includes a Sources button that shows how personal context was applied. Users can ask Gemini directly which sources or references were used for a generated image.

Google said it does not directly train AI models on private Google Photos libraries. However, the company noted that limited data, such as prompts and model responses, may be used to improve functionality over time.

The personalized image creation feature is rolling out over the next few days in the Gemini app for eligible subscribers to Google AI Plus, Pro, and Ultra plans in the US.

Google said support for Gemini in Chrome and wider availability will follow soon.