I just tried Gemini’s new ‘Personal Intelligence’— it’s the end of AI image prompting as we know it

Generating the perfect AI image sounds easy enough until you realize the prompt and the actual image don't seem to match. Luckily, the days of long, complex descriptions are gone.

Today, Google is changing the game by introducing Personal Intelligence with Nano Banana image generation. This new integration allows Gemini to use your interests, preferences and even your private Google Photos library to "fill in the blanks" of your imagination automatically. Here's how it works.

Skip the prompting

Even as a prompt engineer, I can honestly say the biggest hurdle in AI has always been the prompt. But with Personal Intelligence, Gemini now has an inherent understanding of your life from your Google Docs, Photos or even Gmail.

Instead of writing a paragraph to describe your aesthetic, you can now use simple commands:

• "Design my dream house" — Gemini will automatically reflect your specific tastes and lifestyle gleaned from your connected Google apps.
• "Create a picture of my desert island essentials" — The AI draws on your established interests to populate the image.

Photos starring you and your loved ones

The most impressive update is the deep integration with Google Photos. Gemini can now use actual images of you, your family and even your pets to guide the generation process.

Since you can already label people and pets in your library, Gemini uses those labels to make your inner circle the stars of the show. You can ask for a "claymation image of me and my family enjoying our favorite activity," and Gemini handles the rest. Beyond claymation, you can experiment with:

• Watercolors
• Charcoal sketches
• Oil paintings
• Cartoons/comic

The privacy factor

Personal Intelligence is something users can enable or toggle off at any time. But even still, giving an AI access to your photo library sounds like a privacy nightmare. However, Google is leaning heavily into an "opt-in" model. Here's what that looks like:

• No direct training: Google states the Gemini app does not directly train its models on your private Google Photos library.
• Total transparency: A new "Sources" button allows you to see exactly which photo was auto-selected to guide a creation.
• Manual control: If Gemini picks the wrong reference, you can click a ‘+’ icon to select a different photo or simply tell the AI what it got wrong.

How to get it

This feature is powered by the new Nano Banana 2 model and is rolling out now to Google Plus, Pro, and Ultra subscribers in the U.S..

While it’s starting on mobile, Google plans to bring these personalized tools to Chrome on desktop soon.

Give it a try and let me know in the comments what you think. And feel free to share some of your photos with me. I'd love to see what you create.

Follow Tom's Guide on Google News and add us as a preferred source to get our up-to-date news, analysis, and reviews in your feeds.

More from Tom's Guide

• ChatGPT’s new ‘Thinking’ mode just hit a 94% reasoning score — 7 prompts it can solve that standard AI can’t
• I tested Anthropic’s new Claude Opus 4.7 — and it’s the first AI that actually ‘reasons’ through tasks
• I replaced ChatGPT with Google's offline AI on my phone for 24 hours — here's my verdict