Since its release date, ChatGPT has touched the lives of many people. For some of us, it has been a milestone that made us realize the power of artificial intelligence. Today, we will try an extraordinary experiment with ChatGPT. And we will try to create a Mona Lisa painting with ChatGPT.
Can ChatGPT even create an image?
ChatGPT is a language model. It can’t actually draw pictures. Its greatest skill is being able to have conversations like a human. However, when it comes to drawing things, especially complex images, ChatGPT can’t give us great results. But ChatGPT can create images in ASCII and SVG formats. You can learn more about how it generates these images in our relevant topic.
Can ChatGPT draw Mona Lisa or any painting?
Don’t pay too much attention to the statement in the title. ChatGPT can’t actually draw the Mona Lisa painting, of course. However, when it comes to having conversations, ChatGPT has impressive abilities that we appreciate. Today, we’ll conduct a fun experiment to push its limits and see how it can draw a Mona Lisa painting or create a version of it. It will be an enjoyable activity to explore.
SVG format
First, let’s explore what SVG format is. SVG stands for Scalable Vector Graphics, which is the name given to a drawing made using vectors. It is, in a way, the expression of the image using mathematical vectors, and due to the nature of the drawing, it can be scaled to any size without losing resolution.
Today, we will ask ChatGPT to draw the Mona Lisa painting using SVG format. Because it is the most suitable option.
Make ChatGPT draw in SVG format
Now, before we begin, as a simple example, let’s have ChatGPT draw a basic circle in SVG format.
When we ask ChatGPT to do that, it will provide us with the code for the image generated in SVG format.
These codes don’t directly mean anything to us. To be able to see the images represented by these codes, we need to use an online viewer. ChatGPT doesn’t have this feature built-in, so we’ll paste the code into an online viewer to see the image it represents.
You can view the preview of this code using an online tool by looking at the screenshot below. You can also try it yourself on an online SVG viewer by searching for an online tool.
How to create a Mona Lisa painting with ChatGPT?
Now that we have seen how a simple image can be drawn with ChatGPT, we can move on to creating the more complex Mona Lisa painting.
I used the following prompt. You can try it by yourself by copying and using it.
Using the SVG format, output a schematic reproduction of the Mona Lisa painting. [I want to see a frame, a head, two eyes, a nose, a mysterious smile, and a body]. Put the output in the code block. Use colors.
When I used this prompt, ChatGPT provided me with a code in SVG format, just like before. When I previewed this code using an online tool, I obtained the following result.
I really liked the performance ChatGPT showed on the first attempt. But I knew as an advanced language model, it could do more. That’s why I asked ChatGPT to do more detailed work.
And here is a work by ChatGPT, who doesn’t have a great skill in drawing, but in only the second attempt, managed to create a piece that preserves all the artistic meanings of the Mona Lisa painting.
Final Words on ChatGPT and Mona Lisa
Trying to have ChatGPT draw Mona Lisa was quite enjoyable for me. Of course, a much more detailed and beautiful work could have been produced. For this, some corrections need to be made and various feedbacks need to be given. At the end of the day, I think that this text, written solely for entertainment purposes, has fulfilled its purpose.
By applying what has been described in this article, you can also ask ChatGPT to draw a picture and even a much more detailed Mona Lisa painting. Regardless of its drawing ability, we still love ChatGPT nonetheless.
In memory of Leonardo Da Vinci…