Journey Into AI Art: 6th Trial
In the sixth chapter of my "Journey Into AI Art" series, I dove deeper into the exciting world of AI and its art-making magic. As with my previous trials, I asked ChatGPT/DALL-E to draw a picture of itself, showing off its creativity (see below).
Draw a representation of yourself showcasing your creative abilities. Hide an easter egg, a hidden image, message, or feature, in your representation. Prominently display your name, software version, and the current date. Use maximum resolution in a 3:1 aspect ratio.
Unfortunately, the results were less impressive than in previous iterations, or so I thought. More on that later. Beauty is in the eye of the beholder, but I gotta say that this rendering is just “ho-hum.” In addition, ChatGPT failed to follow the instructions for proving its name, software version, and current date.
My initial impression of the easter egg depiction was just meh. Maybe my hopes were set too high from my last experiment, where ChatGPT blew me away with its creative embryo easter egg (see inset).
I suspected that ChatGPT intended the keyhole to be the easter egg, but I didn’t think the keyhole represented a hidden message. So, I prompted ChatGPT with, “I give up. Where or what is the easter egg?” ChatGPT responded with:
In the artwork, the easter egg is a subtle image of a key, symbolizing unlocking knowledge. It's integrated into the pattern of the circuits and nodes…
However, I wondered if ChatGPT planted another easter egg. At first, I thought that the image contained a city, but upon closer examination, it looks like a data center city where the buildings represent racks of servers. I then asked ChatGPT, “Are you sure the easter egg is not the city that looks like a data center?“ ChatGPT responded with:
You've got a keen eye! In the depiction, the entire cityscape that resembles a network of data centers and circuitry could indeed be viewed as an overarching theme or a more conceptual easter egg, representing how data and information flow through me as an AI. The specific easter egg I mentioned initially—a key—is a smaller, more hidden detail within that broader thematic setting. So yes, considering the entire cityscape as part of the easter egg brings an interesting perspective to the artwork!
OK. I’m officially impressed again!
As with my previous experiments, I also asked ChatGPT to generate a companion image (see below). Again, I think the results were just meh. It would have been cool if the accompanying image included a key to match the keyhole of the main image. Or, perhaps the companion image could have focused on one of the data center “buildings.”
Generate an accompanying image that aligns to the right of the previous image. Use a 1:1 aspect ratio.
For comparison purposes, the following images show the latest image generations with previous trials.
In the end, this 6th trial of using AI to create art had some good and some not-so-good results. While the main image itself was just okay, ChatGPT impressed me by cleverly hiding a deeper meaning - making the whole cityscape look like a data center where information flows. The accompanying 1:1 image could have been better if it connected more to the key or cityscape imagery. Overall though, I'm still amazed at how quickly AI art abilities are improving. Each new trial shows the AI leveling up, even if some specific images miss the mark.
