Create Personalized Images in the Gemini App

Imagine an AI that doesn't just know what a dog looks like, but knows what your dog looks like. This week, Google unveiled a major leap for the Gemini app: the marriage of Personal Intelligence and the Nano Banana 2 image model.

It’s a shift from Generic AI to Your AI. By securely connecting to your Google Photos and your broader Google ecosystem, Gemini is moving away from the era of long, complex prompts and into an era where it simply understands who you are.

1. Identity-Aware Creation

The most exciting part of this update is the ability to put yourself and your loved ones at the center of the frame.

• Real-World Grounding: Because Gemini can now talk to your Google Photos library, it uses your existing labels for people and pets to guide its generation.
• The "Me" Prompt: You can now ask: "Create a claymation version of me and my best friend at a concert," or "Sketch my cat as a royal monarch." Gemini identifies the right faces and features from your library to ensure the actors in the image actually look like you.
• No Uploads Required: Since the context is already there, you don't have to hunt for, download, and re-upload reference photos. The AI already knows your world.

2. Intelligence Over Instruction

We’ve all spent too much time trying to prompt engineer the perfect image. Personal Intelligence changes the workflow:

• Automatic Context: If you ask Gemini to design my dream living room, it doesn't just give you a Pinterest-cliché. It looks at your connected Google apps to understand your aesthetic whether you're a minimalist or a maximalist and builds the image accordingly.
• The Sources Button: If you're curious about how Gemini arrived at a design, a new Sources view shows you exactly which of your photos or interests influenced the output.

3. Privacy by Design

In an era of data concerns, Google has built this with a privacy-first architecture:

• Opt-In Only: Personal Intelligence is off by default. You choose which apps to connect and can sever the link at any time.
• No Training on Photos: Crucially, Google states that your private Google Photos library is not used to train its base AI models. Your memories stay yours.
• Creative Control: If the AI picks the wrong photo for a prompt, you can hit the “+” icon to manually swap it for a different memory.

4. Availability & Gear

• Rolling Out Now: This feature is currently hitting the apps of Google AI Plus, Pro, and Ultra subscribers in the U.S.
• The Engine: This is powered by Nano Banana 2 (also known as Gemini 3.1 Flash Image), which blends high-speed generation with Pro-level reasoning.
• Coming Soon: Support for Chrome desktops and a wider global rollout are expected in the following months.

Technology is most helpful when it stops acting like a stranger. By bridging the gap between your digital memories and AI's creative power, Google has turned Gemini into a true personal artist. If you're in the U.S., your next masterpiece isn't just a prompt away, it’s already sitting in your Google Photos. What are you going to create first?