• Welcome to the new COTI server. We've moved the Citizens to a new server. Please let us know in the COTI Website issue forum if you find any problems.

General Has anyone done any Traveller art with the AI tools out there?

It depends upon the legal jurisdiction. At this writing, AI-generated images can’t be copyrighted in the US, but they can be copyrighted in China.
good to know, and interesting. It all comes down to what is art? is the prompt for generating art something that can be copyrighted? But I am taking this thread off topic. I have attempted to do a few things with various AI tools, some Traveller related. So far, I've not been happy with the results, but I also feel I am cheating or something. Personal bias, I know. For me it is still in the toy phase though moving so fast it is leaving that pretty durn quickly!
 
good to know, and interesting. It all comes down to what is art? is the prompt for generating art something that can be copyrighted? But I am taking this thread off topic. I have attempted to do a few things with various AI tools, some Traveller related. So far, I've not been happy with the results, but I also feel I am cheating or something. Personal bias, I know. For me it is still in the toy phase though moving so fast it is leaving that pretty durn quickly!
Part of the issue for me is that it is not really a collaboration when I utterly fail to covey to the AI any of MY visionary contribution. The AI generates images on a "this is what I feel like, so take it or leave it" basis. As an example, I described a knife with a pommel in the shape of the head of a tiger ... the knife had no pommel, but the handle was inlaid with tiger fur and the leather sheath had a tiger head on it. It was pretty, but it bore little resemblance to what I had asked for.

Then there are the obvious limitations ... it can do "photo realism", but not "fingers" ... round wheels and 4 legs often allude it as well. So watching a computer create things unrelated to what I want is not my definition of "fun" and if it is "ART", it really doesn't seem like "MY" artwork (as an artist) ... just art that I commissioned with a prompt and had trivial control over (I get to specify the style).
 
For me it is still in the toy phase though moving so fast it is leaving that pretty durn quickly!
That's the problem with AI tech ... it grows at exponential speed ... and humans are REALLY BAD at judging exponential rates of anything.

It's one of those "start with a penny, double it every day, by the end of a month you're a millionaire" kinds of deals.
If you ask (most) humans when would the "midpoint" of wealth accumulation have been, (most) people tend to assume mid-month about 15 days in. The thing is that most people do not realize that the "half a million" day was actually the day before reaching (over) 1 million.

Our mental habit is to draw lines of expectations, not curves that go exponential, when trying to predict the future.



{ cue Yoda voice: "That is why you fail." }
 
Part of the issue for me is that it is not really a collaboration when I utterly fail to covey to the AI any of MY visionary contribution. The AI generates images on a "this is what I feel like, so take it or leave it" basis. As an example, I described a knife with a pommel in the shape of the head of a tiger ... the knife had no pommel, but the handle was inlaid with tiger fur and the leather sheath had a tiger head on it. It was pretty, but it bore little resemblance to what I had asked for.

Then there are the obvious limitations ... it can do "photo realism", but not "fingers" ... round wheels and 4 legs often allude it as well. So watching a computer create things unrelated to what I want is not my definition of "fun" and if it is "ART", it really doesn't seem like "MY" artwork (as an artist) ... just art that I commissioned with a prompt and had trivial control over (I get to specify the style).
I’m sure many Renaissance patrons experienced the same thing.
 
Part of the issue for me is that it is not really a collaboration when I utterly fail to covey to the AI any of MY visionary contribution. The AI generates images on a "this is what I feel like, so take it or leave it" basis. As an example, I described a knife with a pommel in the shape of the head of a tiger ... the knife had no pommel, but the handle was inlaid with tiger fur and the leather sheath had a tiger head on it. It was pretty, but it bore little resemblance to what I had asked for.

Then there are the obvious limitations ... it can do "photo realism", but not "fingers" ... round wheels and 4 legs often allude it as well. So watching a computer create things unrelated to what I want is not my definition of "fun" and if it is "ART", it really doesn't seem like "MY" artwork (as an artist) ... just art that I commissioned with a prompt and had trivial control over (I get to specify the style).
I just try to keep working at it until I get something that I like or can live with. That does take a while.
 
I just try to keep working at it until I get something that I like or can live with. That does take a while.
I am contemplating trying the "start with an image" feature and Photoshop a rough collage of cut, stretch and paste objects that look like crap, but convey "composition" and see what happens from there. I just have not gotten around to it yet.
 
I am contemplating trying the "start with an image" feature and Photoshop a rough collage of cut, stretch and paste objects that look like crap, but convey "composition" and see what happens from there. I just have not gotten around to it yet.
Look my stuff over if there is anything that you can use. Otherwise, let me know here and I will try to make you something that you can evolve or edit.
 
Look my stuff over if there is anything that you can use. Otherwise, let me know here and I will try to make you something that you can evolve or edit.
Here is an example ...

I always thought that a JEEP with off-road suspension and TRACKS would look great. It would be even better in an "action" image rather than a static 3-view. So start with an image of a Baja Jeep catching some air as it comes over a hill. Then photoshop in the tracks from the "RIPSAW" to replace the wheels. Then maybe adjust the background image to look more Jurassic Park.

Then see what AI could do to clean up that rough cut-paste starting image.

[I just lack the free time to get around to that project and the little tinkering with AI that I have done, suggest a low probability of success ... so it is low in my pile of ideas. Right now I am pushing the envelope on a size 5 world with a tainted atmosphere (excessive ozone), dense forest coverage, a population in the tens of thousands and TL 10 ... they are combining log/timber construction - because wood is everywhere - with some advanced technology and sealed "arcologies" to filter the air. Most recent was small Arctic settlement over some critical mineral resources that employed a translucent Buckminster dome facing south, a ring of cylindrical towers forming the north edge and a north berm of snow over earth for insulation. ... for a CT PbP game called "BACKWATER".]

file.php
 
Last edited:
So I decided to invest the time and see what would happen with my little experiment.

The GOAL:
I always thought that a JEEP with off-road suspension and TRACKS would look great. It would be even better in an "action" image rather than a static 3-view. So start with an image of a Baja Jeep catching some air as it comes over a hill. Then photoshop in the tracks from the "RIPSAW" to replace the wheels. Then maybe adjust the background image to look more Jurassic Park.

Uploaded to the AI ...
The Photoshop Start Image [LINK]

The Text Prompt: "Tracked jeep racing through Jurassic Forest oil painting by James Gurney"

The Result:
wndOfeZ9c84F0Q7Dsbh3--1--b3y41.jpg


Not quite MY vision ...
 
Last edited:
New Text Prompt with same Photoshop image: "Jeep with tank treads detailed matte painting, deep color, fantastical, intricate detail, splash screen, complementary colors, fantasy concept art, 8k resolution trending on Artstation Unreal Engine 5"

New Results: (2 attempts)
7h7ptRiWupXTGcTx2vbN--1--npusg.jpg


CLDfwUrNB15mrk5Hsj1M--1--g2a73.jpg
 
Back
Top