I’ve been playing with AI-generated art via Midjourney and DALL-E, which take text prompts and turn them into images. That may not sound impressive, but the results show otherwise:
Midjourney – Prompt: cyborg Sloth at night, neon city alley, neotokyo, cinematic, 
Midjourney – Prompt: a portrait of bacon by Banksy 
DALL-E Prompt: origami Winston Churchill during the Blitz 
Midjourney – Prompt: stained glass Ewok savior, symmetrical 
Midjourney – Prompt: Tom Waits as a grizzled Jedi in the style of a scifi book cover 
Midjourney – Prompt: punk Cicero 
Midjourney – Prompt: origami samurai walking, neotokyo, blade runner, neon, alley, night, light rain, highly detailed 
Midjourney – Prompt: ???? viking with ???? 
Midjourney – Prompt: Hank Hill in Fight Club
I’ve become infatuated with this new set of capabilities and the opportunity to create in an utterly new way. While I’ve generated many random ideas, including some with my 10-year-old, I’ve also used it as an opportunity to be intentional in order to craft a series of images.
As the son of an avid reader of science fiction, who grew up to be one myself, I was surrounded by amazing cover art in my formative years. So, in that spirit, I present:
Bear Force
Conceptually, I wanted to build out a set of portraits of a bear-like race known for their prowess as pilots in the deep reaches of space. It may sound silly, but it proved a wonderful starting point. (Don’t worry, I have no intention of trying to write a book about them).
For the curious, I built these out through many iterations within Midjourney, starting with the initial prompt Panda fighter pilot, cinematic and working from there. When I was happy with my first image, it became the starting point for those that followed, so the following prompts included the image URL followed by Black Bear fighter pilot, portrait, cinematic --ar 16:9 etc.
What’s next?
Well, beyond allowing my imagination to run wild, I’m also working through how to craft images to use within articles and even presentations. It is an amazingly flexible system to work with.
Sound interesting?
I highly recommend trying out Midjourney, as it allows you to generate a bunch of free images to start, and the paid plans are pretty reasonable thereafter if you get hooked. There’s a vibrant community in discord and a lot to learn (in a fun way!).
If you do sign up, let me know! My profile is here (you have to be a member to see it), and I’d love to see what you make!