My ever-evolving theory on artists using AI growth! I’ll make a bigger article soon… but you can read my one from only 7 days ago AND you can look at the differences between those images and these images. It’s amazing how much progress can be made in just a week! The article focuses on how artists can team up with an AI to make some pretty amazing stuff.
Here are a few snippets from the previous article:
I have an interesting theory on AI in this arena – I believe a very interesting future for all artists is to “team up” with AIs to form an UberArtist.
Chess Grandmasters who routinely get destroyed by AIs sometimes team up with an AI in competitions against other teams. Anyway, with photography and pretty much all other forms of art (and soon video), it’s clear to me that AI will be able to make some extremely interesting creations, often surpassing what the most creative humans can produce. And it’s only going to get better. So, rather than “fight” the system, it only makes sense to join forces as artists curate and coax their AI into creating things they think are interesting.
…
After you do a little mind-flip on the benefits of AI, it’s like having a very creative assistant around you all the time, making wild suggestions that you may never think of. It’s like having a coach in a way… or the way Tiger Woods has a caddy to make suggestions… Many great people out there have coaches and listen to ideas and thoughts that may never have occurred to them.
I use about five different AI tools, and my new favorite is Midjourney Version 4. It doesn’t have much of a user interface, and you use Discord and command-line language with switches to create the images. I’m a Computer Science nerd, so I’m all into this stuff, but it may be a bit esoteric to others. Basically, you tell the AI what you want. I’ll give you a sample prompt I used for one of the images below:
“A huge semi-buried Burning Man sculpture in an inhospitable white desert, night time, dystopian, awe-inspiring, Trey Ratcliff style, lonely feeling, hyper-realistic, ultra-detailed, ultra-realistic, UHD, insanely beautiful, 32k, unreal engine, high octane, perfect composition, cinematic lighting, lonely feeling, minimalist, humbling scene, Joseph Campbell quest, a feeling of Rumi’s the cave you fear holds the treasure you seek –v 4 –q 2 –uplight –ar 1:1 ”
You may notice that I used the words “Trey Ratcliff Style” in the command line. This was an unexpected joy! The AI actually has gone out and seen all my Burning Man photos (and other photos – see my Full Portfolio) that I’ve been uploading to the internet for the last 15+ years. This makes the final rendering look quite similar to the sort of photo I might actually take. So, it turns out I’ve been accidentally training the AI for the last 15 years! Enjoy the photos, and I’ll have another series of articles and sample images in future posts… in the meanwhile, feel free to follow me on Instagram or Twitter.
Also, at the bottom after all the Burning Man works, I added some different sorts of creations I have made with this AI. I’ll post another article soon that explores all of these innovative ways to use AI to make visually interesting creations. I know I get quite a bit of negative feedback from photographers (or other visual artists) on this stuff because it is not hand-crafted, but that’s okay. I’ve got through over a decade of HDR hate and stuff. ️
Other Creations with the AI
Below is a sampling of other fun creations I’ve created. I’ll post a longer article later that really does an analysis of what is going on here and some fun implications for the art world.
I’ll give you the prompts here for the first couple… and you can have fun trying to guess the next ones. I’m not keeping secrets, I’m just toooo lazy now to copy and paste all the prompts.
Below:
“a stunning interpretation of a leopard Human girl hybrid, highly detailed and intricate, hypermaximalist, blonde, big blue intense eyes, mouth slightly open, ornate, luxury, elite, welcoming, loving, matte painting, cinematic, cgsociety, James jean, Brian froud, ross tran, fantasy forest background -v 4 –q 2 –uplight”
Below:
“insanely detailed photorealistic Japanese treehouse in a tall cherry blossom tree, hybrid, highly detailed and intricate, mysterious foggy night, closeup of treehouse, glowing treehouse, many cherry blossoms around treehouse, treehouse in focus, dark green forest in background out of focus, tiny glowing fairies in the background, Miyazaki style, 4k, octane render, Rembrandt lighting -v 4 –q 2 –uplight”
My ever-evolving theory on artists using AI growth! I’ll make a bigger article soon… but you can read my one from only 7 days ago AND you can look at the differences between those images and these images. It’s amazing how much progress can be made in just a week! The article focuses on how artists can team up with an AI to make some pretty amazing stuff.
I use about five different AI tools, and my new favorite is Midjourney Version 4. It doesn’t have much of a user interface, and you use Discord and command-line language with switches to create the images. I’m a Computer Science nerd, so I’m all into this stuff, but it may be a bit esoteric to others. Basically, you tell the AI what you want. I’ll give you a sample prompt I used for one of the images below:
“A huge semi-buried Burning Man sculpture in an inhospitable white desert, night time, dystopian, awe-inspiring, Trey Ratcliff style, lonely feeling, hyper-realistic, ultra-detailed, ultra-realistic, UHD, insanely beautiful, 32k, unreal engine, high octane, perfect composition, cinematic lighting, lonely feeling, minimalist, humbling scene, Joseph Campbell quest, a feeling of Rumi’s the cave you fear holds the treasure you seek –v 4 –q 2 –uplight –ar 1:1 ”
You may notice that I used the words “Trey Ratcliff Style” in the command line. This was an unexpected joy! The AI actually has gone out and seen all my Burning Man photos (and other photos – see my Full Portfolio) that I’ve been uploading to the internet for the last 15+ years. This makes the final rendering look quite similar to the sort of photo I might actually take. So, it turns out I’ve been accidentally training the AI for the last 15 years! Enjoy the photos, and I’ll have another series of articles and sample images in future posts… in the meanwhile, feel free to follow me on Instagram or Twitter.
Also, at the bottom after all the Burning Man works, I added some different sorts of creations I have made with this AI. I’ll post another article soon that explores all of these innovative ways to use AI to make visually interesting creations. I know I get quite a bit of negative feedback from photographers (or other visual artists) on this stuff because it is not hand-crafted, but that’s okay. I’ve got through over a decade of HDR hate and stuff. ️
Other Creations with the AI
Below is a sampling of other fun creations I’ve created. I’ll post a longer article later that really does an analysis of what is going on here and some fun implications for the art world.
I’ll give you the prompts here for the first couple… and you can have fun trying to guess the next ones. I’m not keeping secrets, I’m just toooo lazy now to copy and paste all the prompts.
Below:
“a stunning interpretation of a leopard Human girl hybrid, highly detailed and intricate, hypermaximalist, blonde, big blue intense eyes, mouth slightly open, ornate, luxury, elite, welcoming, loving, matte painting, cinematic, cgsociety, James jean, Brian froud, ross tran, fantasy forest background -v 4 –q 2 –uplight”
Below:
“insanely detailed photorealistic Japanese treehouse in a tall cherry blossom tree, hybrid, highly detailed and intricate, mysterious foggy night, closeup of treehouse, glowing treehouse, many cherry blossoms around treehouse, treehouse in focus, dark green forest in background out of focus, tiny glowing fairies in the background, Miyazaki style, 4k, octane render, Rembrandt lighting -v 4 –q 2 –uplight”
Below (yes the whole thing is created by AI – quite meta, hey? I borrowed this idea!):
“a bookish redditor shows off her painting in a niche subreddit.”
More soon!
More soon!