Last Updated on April 13, 2025 4:07 pm by Jeffrey Powers

You’ve probably scrolled past a post on social media where someone transformed their face into a boxed action figure—and thought, “How did they do that?” The answer: AI. Specifically, ChatGPT 4.0 with vision capabilities.

The idea of turning yourself into an action figure may sound like a complex, high-budget task, but it’s surprisingly simple with the right tools and prompts. With a bit of creativity, a solid photo, and some specific directions, you can create everything from a YouTube influencer figure to a superhero cover straight out of a vintage comic book.

The process begins with a clean, front-facing photo—ideally in front of a plain background. If you can fix it in a photo editor, it will help with the overall look if GPT doesn’t have to analyze too much.

Give your photo a name you can use (like jeff.png), then upload it into ChatGPT. If you have other photos or even logos, you can tell the prompt which photo to reference.

Keep in mind, it’s not going to be 100% accurate. This AI will see the image as “a person with”. In my case, it’s “An older man with glasses and a long beard”.

Since GPT sees an image and creates a description, so the magic happens in how you create your prompt. One well-written paragraph can define your face, outfit, accessories, pose, packaging, and even the background environment. Take that prompt to the limit – which I don’t believe there is one.

The more detail, the better.

I put this together with the help of Owen Hemsath’s prompt. The one thing missing was the logo – which I wanted to test if you could separate images. This is the prompt I used:

Once the standard figure is complete, you can use the prompt to make adjustments, or start making other things. What if you were a Cabbage Patch Kid? What if you were a superhero action figure?

I started making comic book covers, my likeness on a candy bar, and even a cereal box. With the right prompt, you could make a graphic, diorama, and more.

And you can take your prompts beyond the person. Put action figures in scenes, holding items, and more. Even a simply photo and prompt of “Recreate this image turning the person into an action figure” makes for a fun photo.

While MidJourney could do this, it might take a lot more information and more prompts. But by uploading a photo and prompting in MidJourney will only make you a generic action figure.

While MidJourney isn’t creating the results I am looking for in this case, I still use MidJourney for many other things. The images in the video (including that Matrix image) was mostly created using MJ.

This tool is getting more sophisticated as we expected. Being able to focus on what you say in your prompts can get you not only action figures, but also scripts you can use to make your life easier (I’ve done that several times now), programs like games and more.

But think of this – what if you could prompt a way to search for your likeness across the Internet? Or prompt a program that will keep you safe on the web?

The possibilities are endless.

Reviews: Geekazine gets products in to review. Opinions are of Geekazine.com. Sponsored content will be labeled as such. Read all policies on the Geekazine review page. 

Reviews: Geekazine is also an affiliate of Amazon

The post Make an Action Figure and More Using ChatGPT appeared first on Geekazine.