Journey Into AI Art: 2nd Trial

Well… I couldn’t wait. I was going to post monthly updates, but I wanted to bring in the new year by exploring the latest capabilities of ChatGPT. I used the same prompt as last time (see below).

Draw a representation of yourself showcasing your creative abilities. Prominently display your name, software version, and the current date. Use maximum resolution in an aspect 3:1 ratio.

Despite my enthusiasm, I was a bit disappointed by the results. The following diagram compares my most recent experiment with my previous one. I like that ChatGPT included a face this time, but it misspelled “CHAT“. Also, it provided the wrong date, but I suspect that ChatGPT used the date that OpenAI trained it.

The accompanying image wasn’t much better. As in my last experiment, I used the following prompt, which generated the below image:

Generate an accompanying image that aligns to the right of the previous image. Use a 1:1 aspect ratio.

And finally, here is the combined image for my second trial of AI art evolution.

As you can see, the accompanying image doesn’t feel very accompanying. I guess the ideal result would have been a mirror image variation of the larger main image.

I had high expectations for this round of my experiment. I had expected that DALL-E’s reinforcement learning would have improved the results in a mere sixteen days. Although the progress wasn't as pronounced as hoped, I plan to continue this journey of evaluating AI’s art evolution. Stay tuned…