ChatGPT Can Turn You Into a Toy Action Figure
Last month, ChatGPT upgraded its AI image generator to a powerful new model which has generated a lot of buzz. First, there was the Studio Ghibli style images, but a new trend now sees people becoming toy action figures in a box.
It works by the user uploading a photo of themselves to ChatGPT and typing in a prompt, something along the lines of: “Create a picture of me as a 3D action figure toy, as if it were displayed in a transparent plastic packaging, blister packaging model.”
The user can add accessories by telling ChatGPT to include supporting items like a camera, a lens, and a tripod. You can also tell it to include your name on the box as well as a title, like “Photographer.”
PetaPixel has spotted numerous photographers on the internet jumping on the trend. We’ve rounded up a few examples from Instagram below.
ChatGPT shared a prompt on how to make the AI-generated image. Which is as follows.
Prompt: Draw a picture of me as a real-life action figure. Be specific as possible based on what you know about me. And put multiple interesting things that represent who I am inside the packaging.
However, there are other prompts and styles available. You can make it way more specific to try and control the image a little better. The Make It Simple Youtube channel shared the following prompt for making a Cristiano Ronaldo figurine which you can edit accordingly.
Prompt: Create a picture of a 3D action figure toy, as if it were displayed in a transparent plastic packaging, blister packaging model. The figure is as in the photo, the style is relaxed, friendly smile. On the top of the packaging there is a large writing: “CRISTIANO RONALDO” then below it “FOOTBALLER”. Also add some supporting work items next to the figure, like a ball, shoes, and a world cup trophy. The packaging design is minimalist, cardboard color, cute toy style sold in stores. The style is cartoon, cute but still neat and clean, ala-ala mattel toys, also give the mattel logo in the top corner.”
Scary
AI remains controversial but features like this are undeniably impressive. Creating an image like this would have taken hours of work and years of expertise, now it can be done in a matter of seconds from a simple text prompt.
It is also worth bearing in mind that OpenAI may train on any data users upload to ChatGPT — including photos.
