Google Gemini can use Google Photos to generate personalized images

In short: Google says Gemini can now connect to Google Photos and use what it learns there to create more personal AI generated images.

What happened

Google is adding a new option to Gemini, its AI assistant, through a feature it calls Personal Intelligence. If you choose to connect it to Google Photos, Gemini can generate images that are meant to match your “tastes and lifestyle,” according to Google.

Google says you can type prompts like “Design my dream house” or “Create a picture of my desert island essentials.” The images Gemini makes will try to reflect your real life preferences, based on information from the Google apps you have connected.

A Google spokesperson told The Verge that the system uses labels in Google Photos to spot people connected to you, like friends and family. Then it uses Google’s image model called Nano Banana 2 to create the new picture. An image model is the part of an AI system that makes pictures from text, like a very fast illustrator that follows written instructions.

Google says that if you opt in, it will not directly train its AI models on your private Google Photos library. However, it does train on “limited info,” including the prompts you type into Gemini and the model’s responses.

Google says the feature will roll out over the next few days for eligible AI Plus, Pro, and Ultra subscribers in the US. It is also planned for Gemini on Chrome desktops and for more users later.

Why it matters

This makes AI image tools more personal, but it also raises new privacy questions. Many people keep years of family photos in Google Photos, so even using labels and prompts can feel sensitive, like giving an assistant access to your photo albums and notes.

Source: The Verge AI