Two new AI meant for generation of art have been getting the spotlight lately. One is called MidJourney and the other called Dall-E. Both are ultra-refined photo-bashing software which algorithmically take keywords (prompts) and merge photos, textures and 3d models from all over the internet into something unique, using various different styles both traditional and digital.
AI generated art has been a thing for a while now. We even have another thread over here by @pearl AI art generated Catholicism WomboAI was one of the first more refined art AI and I've used it too. Even made a i-ching Tarot deck with it (sample below.)
But these two new softwares have become robust enough to make high quality illustrations with very little human work. Raising concerns about the future of illustration and whether technology will become sophisticated enough to displace artist jobs. There's also concern about copyright and how images generated by AI can or will be used in the commercial art world.
Both are in Beta mode and subscription or invite model services. I haven't gotten into Dall-E yet but I have played around with MidJourney. To give you an idea of where we're at in the technology, see some of the works I was able to create just this week:
For more, you can see one of the many MidJourney art galleries over here on Instagram.
To give you an idea of how easy this is to make, here is a screen shot of the discord prompt, where you put in your keywords and then it gives you four possible results, which you can then refine through variation (v) or upsize for finished product (u).
My personal opinion is that, rather than looking at this as competition for the art field, it should be used more as a tool for inspiration and references by artists to make more refined works. Here is a (very simple) example where I created a prompt and then redrew the output (with more simple lines because it's going to be a ceramic mug transfer).
What do you guys think about this technology? Are you excited? Nervous?
AI generated art has been a thing for a while now. We even have another thread over here by @pearl AI art generated Catholicism WomboAI was one of the first more refined art AI and I've used it too. Even made a i-ching Tarot deck with it (sample below.)
But these two new softwares have become robust enough to make high quality illustrations with very little human work. Raising concerns about the future of illustration and whether technology will become sophisticated enough to displace artist jobs. There's also concern about copyright and how images generated by AI can or will be used in the commercial art world.
Both are in Beta mode and subscription or invite model services. I haven't gotten into Dall-E yet but I have played around with MidJourney. To give you an idea of where we're at in the technology, see some of the works I was able to create just this week:
For more, you can see one of the many MidJourney art galleries over here on Instagram.
To give you an idea of how easy this is to make, here is a screen shot of the discord prompt, where you put in your keywords and then it gives you four possible results, which you can then refine through variation (v) or upsize for finished product (u).
My personal opinion is that, rather than looking at this as competition for the art field, it should be used more as a tool for inspiration and references by artists to make more refined works. Here is a (very simple) example where I created a prompt and then redrew the output (with more simple lines because it's going to be a ceramic mug transfer).
What do you guys think about this technology? Are you excited? Nervous?
