ChatGPT takes on Google Gemini Nano Banana AI with its own AI Action Portrait images: Here’s how to create it, prompt and other details

OpenAI has introduced a new creative feature called AI Action Portrait images, a response to Google’s Gemini Nano Banana AI image generation experiments. The feature allows users to create stylized, superhero-like action portraits from their own photos. Unlike conventional photo filters, this tool leverages generative AI to build 3D-like illustrations with customizable labels, costumes, and even sidekicks.

How it works
The AI Action Portrait images rely on OpenAI’s multimodal model capabilities. A user uploads a photo of themselves, and the AI transforms it into a stylized action figure portrait. The process includes options to personalise the box packaging, add text labels, and even include pets or symbolic side characters. While Google’s Gemini Nano Banana AI focuses on lightweight mobile image generation, OpenAI’s approach integrates directly into ChatGPT’s ecosystem, keeping the workflow simple.

How to create AI Action Portrait images
Users can generate these portraits in a few simple steps:
1. Upload a clear photo of yourself into ChatGPT.
2. Provide details such as your preferred superhero name, costume design, or any specific themes.
3. Optionally, include your pet or a symbolic sidekick to be placed alongside you in the action portrait.
4. ChatGPT processes the request and generates a 3D-style action figure image inside a virtual box.

The images are created instantly, with options to refine or adjust prompts for better accuracy. Since the AI processes user input directly, users retain creative control over how their portrait looks.