Mind-Blowing AI Image Generator - Give Visual Representation to C's Concepts?

[Andrew]

Hi Everyone,

I wanted to share something I found I think is quite profound. An AI Image generator that seems to have unlimited possibilities as to what it can create.

I found it from this video:

(highly recommend watching to learn how to use) and have been absolutely blown away by what it can do.

Take a look at the attached pictures. These are all randomly generated images by an AI using descriptive keyphrases and words that can be upscaled to incredible resolutions and clarity.

I thought it would be fun to incorporate some of the concepts we've learned from the C's and well... just take a look for yourselves. This is really the tip of the iceberg on what you can do as I'm still learning. So if others find interest in this, would love to see your own creations.

Maybe this can help give some type of representation to more complicated concepts we don't have a visual for?

Also, you can hover over the image to see what I've titled them. :)

Enjoy!

 

[Andrew]

I like this one in particular. These were the descriptors to give an example: "lords of entropy fighting lords of light, spiritual battle above planet earth, highly detailed, hyper dimensional"

 

[Alejo]

Hey Andrew,

Thanks for sharing, those are indeed very pretty images. How did you get to them, I watched the video and I guess you could get to very detailed description of what you want the AI to create, right?

I think this is one of the most advanced ones I've seen thus far, and it does create a very pretty image indeed.

My only contention is that it's lacking the experience of the concepts it depicts, there's a distance from what a human would comprehend, by being part of what it is creating, that is palpable... despite how pretty the image is.

And that is the reason I use pretty and not beautiful, I think beauty requires conscious effort. I recently started reding McGillchrist, and his explanation on how the right hemisphere works with the left, particularly in the creation of art, might've ruined these AI image generators for me lol.

Perhaps, put another way, what's lacking in these images, despite their prettiness, is the artist, no one is reaching out through the pictures to connect with my me... if that makes any sense.

It is a cool image generator, I will not deny that once. And I don't mean to be a party pooper or suggest that no one should experiment with it, I think it's cool enough to give it a shot for sure.

My two cents.

 

[whitecoast]

Very interesting Andrew and Alejo. I agree that they are pretty, but there seems to be a certain something missing from them. Maybe it's the rightbrain-to-rightbrain connection that all beauty telegraphs? The beauty I see in the images tends to come from the palates themselves, if not too dark, and when drawn from specific natural phenomena (eg, the sun, stars, planets, plants, natural skylines).

 

[BHelmet]

All eye candy, no soul.

 

[Eulenspiegel]

People are really getting carried away by this silly silicon valley AI madness. At the end of the day, it's just algorithms pretending to be something more than they actually are. They cannot replace or even imitate the human soul.

 

[BHelmet]

People are really getting carried away by this silly silicon valley AI madness. At the end of the day, it's just algorithms pretending to be something more than they actually are. They cannot replace or even imitate the human soul.

You are quite right. But I don’t think that is what they are trying to do. They want humanity sucked into an even deeper state of hypnosis and a greater degree of control And that’s where this imagery and AI comes in as far as it looks, anyway.

And it’s meant for the younger, inexperienced generations that grew up in the digital age and can’t tell the difference.

 

[dredger]

Hi, i read about such AI these last days, from a french article which is the following, posted the 27/7, so recent :
Name of the project is DALL-E :

J’ai testé DALL-E 2

Grâce à Fooohx, lecteur de Korben.info et soutien sur mon Twitch, j’ai testé en live cette semaine l’intelligence artificielle DALL-E 2 en bêta. Pour ceux qui auraient loupé le phénomèn…

korben.info

( Here's the link to auto-translation to english )

And here's the site of the project : DALL·E 2
I did not tried it, there's visibly still the possibility to open an account, which is the lone way to test it.
The article is interesting to read as you'll have a couple of good example this guy directly tested.

My personal thought about is that it's ... frightening. This is one step more in the bad direction, the virtualisation of the reality or something like this (even if there would be much to say about what is really our reality)

 

[Andrew]

My personal thought about is that it's ... frightening. This is one step more in the bad direction, the virtualisation of the reality or something like this

While I agree with most everyone that it lacks soul (it is an AI afterall) I think you hit the nail on the head. It's definitely frightening on how this technology is progressing.

 

[psychegram]

Those are some rather striking images.

I'd like to push back a little against the idea that there's no soul behind them. While the images were generated by the machine, the machine was being used by Andrew - who, presumably, has a soul. I don't see any reason why an artist could not learn to use such technology as a unique medium, and express themselves through it every bit as authentically as a sculptor or painter.

An analogy with photography. The invention of the camera rendered a great deal of sketch art superfluous. Previously, the only way to communicate an image of something was to draw it. After the camera came along, journalists and scientists (for example) no longer needed to learn how to sketch. On the one hand, this led to the disappearance of this skill in those professions. On the other, photography itself rapidly grew into its own art form, in which the photographer learned to combine an eye for a good shot with skill in selecting filters, lighting, exposure times, etc. The rise of digital photography changed things again: it was no longer necessary to choose the right film or method of developing the film; on the other hand, digital photographers must become virtuosos with Photoshop.

In a similar fashion, I could easily see talented artists learning to work with these machine learning systems to develop an entirely novel visual art form, where the artist plays the role of muse to the machine by telling it what to make, selects the styles it works with, curates the images it generates, and tweaks those images until they match the artist's inner vision.

 

[Andrew]

In a similar fashion, I could easily see talented artists learning to work with these machine learning systems to develop an entirely novel visual art form, where the artist plays the role of muse to the machine by telling it what to make, selects the styles it works with, curates the images it generates, and tweaks those images until they match the artist's inner vision.

Ya, I can definitely see what you're saying. And after some consideration this could function as a great tool to get initial conceptual images to use as a base to work from - to match the inner vision as you stated. It literally spits out these wild images in seconds. Still trips me out.

Those are some rather striking images.

Indeed. I had the idea to take what the C's said word for word from some of the transcripts (the orion federation one for example) and it seems to be able to reproduce strikingly accurate representations. At least what one could envision.

 

[psychegram]

Ya, I can definitely see what you're saying. And after some consideration this could function as a great tool to get initial conceptual images to use as a base to work from - to match the inner vision as you stated. It literally spits out these wild images in seconds. Still trips me out.

Indeed. I had the idea to take what the C's said word for word from some of the transcripts (the orion federation one for example) and it seems to be able to reproduce strikingly accurate representations. At least what one could envision.

The Orion Federation images were particularly cool.

 

[psychegram]

Ya, I can definitely see what you're saying. And after some consideration this could function as a great tool to get initial conceptual images to use as a base to work from - to match the inner vision as you stated. It literally spits out these wild images in seconds. Still trips me out.

I was also thinking of what writers could do with this. Most authors don't have the talent to properly illustrate their ideas. With a tool like this, illustrations could be added to their work relatively easily, which could enhance their ability to communicate their ideas. Of course, one can always hire an artist, but even a decent book cover is pretty expensive and an independent author doesn't usually have the money for that.

 

[Andrew]

I was also thinking of what writers could do with this. Most authors don't have the talent to properly illustrate their ideas. With a tool like this, illustrations could be added to their work relatively easily

Exactly. I had this same discussion with my girlfriend. And I believe all these images can be used commercially if you wanted to. It can also do logos, shirt designs, graphics, comic strips, etc.

 

[Approaching Infinity]

If AI can develop enough complexity to hold a hyperdimensional soul imprint, then what are the chances that some of this art could conceivably (now or in the future) represent a glimpse into other worlds (or an alien mind)? I don't know, but I would like to see that movie!