Google announced Thursday that it is extending Personal Intelligence, the Gemini feature that gives the AI access to your search history, Gmail, and YouTube activity, to its image generation model. The update connects Gemini's Nano Banana 2 image generator to your personal data across Google apps, enabling it to produce images that reflect your life, your preferences, and, if you opt in, the faces of people in your Google Photos library.

Here is Google's pitch for what problem this solves: "One of the biggest hurdles in AI image generation is finding the right prompt." The company claims that writing detailed descriptions and uploading reference photos was too much work. So it connected the model directly to your data. Now you can type "design my dream house" and Gemini will infer the answer from your search history, photo library, and email activity.

Your Photos Are Now Reference Material

The most significant change is Google Photos integration. If a user has opted in and applied labels to images identifying people, places, and objects, Gemini will use those photos as reference material for image generation. A prompt like "create a claymation image of me and my family enjoying our favorite activity" will pull faces from your photo library, identify your hobbies from your data, and generate the scene. Previously, this required manually uploading a reference image. Now Gemini pulls from your existing library automatically.

The feature is rolling out to Gemini app users on Google AI Plus, Pro, and Ultra plans. It will eventually come to Gemini in Chrome desktop and other platforms. Europe was excluded from the initial global rollout.

The Privacy Language Is Doing Heavy Lifting

Google says Gemini "does not directly train its models on your private Google Photos library." It adds: "We train on limited info, like specific prompts in Gemini and the model's responses, to improve functionality over time." The words "directly" and "limited" are carrying a lot of weight in that sentence. Your photos are not training data, but your prompts and Gemini's responses to your photos are.

The feature is opt-in. But the gap between "we need your permission" and "we assume your permission covers this" has been narrowing across the tech industry for years. When Google first launched Personal Intelligence earlier this year, it connected Gemini to your search and YouTube history. Now it has your photos, emails, and browsing habits. Each expansion is framed as convenience. Each expansion gives the model more of your life to work with.

The On-Device Play

Google is also bringing on-device image generation via Gemini Nano to Pixel phones and Android devices. This would enable instant, private generation without cloud dependency. The combination positions Google to cover both ends of the spectrum: cloud-powered personalised generation for complex requests and on-device generation for speed and privacy.

But the obvious concern remains. Giving an AI image generator access to your personal photos, emails, and browsing history creates risks that Google's opt-in controls may not fully address. And the fact that Europe was excluded from the global rollout tells you exactly who is asking the harder questions about this kind of integration.

As Gizmodo noted: "If you've decided that the level of cognitive offloading that you want is deferring to AI to tell you what your own dream house would be, give it a shot."

First reported by Google, with analysis from Gizmodo, TechCrunch, The Next Web, and 9to5Google.