Just for fun, I decided to ask a few text-to-image artificial intelligence models to generate images using just the prompt “Alexology” with their default negative prompts. Here are some of the images that they generated (click to expand):

Even though it still messes up often, AI has come a long way in being able to generate coherent text. Something like this would have been difficult to make just a year or two ago.

Also what’s that on his head?

This one shows one of the most defining traits of AI-generated images. At first glance, everything looks fine but if you zoom in then a lot of the details don’t really make sense.

This one looks like it could be a pretty good album cover for a retro-inspired band (shame about the mangled text though).

This one could also be an album cover or a great promotional image for a video game.

What are your thoughts on AI-generated images? Love them? Hate them? Feel free to share your thoughts and creations in the comments.