• Welcome to Photrio!
    Registration is fast and free. Join today to unlock search, see fewer ads, and access all forum features.
    Click here to sign up

Can AI be used to produce art?

Sean

Admin
Admin
Allowing Ads
Joined
Aug 29, 2002
Messages
13,581
Location
New Zealand
Format
Multi Format
@Alan Edward Klein I consider Claude the most human like ai I have ever conversed with. So much so it really blew my mind and felt beyond any sort of "chatbot" experience. It was the first time I actually felt that there was something else going on here. I would suggest having a long conversation with it and I'd love to know your thoughts after that. It should be free:


You could start with questioning its views on AI's ability to create art or think etc. I found the longer the conversation the more I could not believe it was not sentient.
 

BMbikerider

Member
Allowing Ads
Joined
Jul 24, 2012
Messages
3,038
Location
UK
Format
35mm

My ex wife (who was also a half decent photographer) said the same thing about digital, almost word for word.
 

Alan Edward Klein

Member
Allowing Ads
Joined
Aug 29, 2017
Messages
10,137
Location
New Jersey formerly NYC
Format
Multi Format

Alan Edward Klein

Member
Allowing Ads
Joined
Aug 29, 2017
Messages
10,137
Location
New Jersey formerly NYC
Format
Multi Format

I tried it, and Claude answered, admitting what I said about him in previous posts. I highlighted it. He's not sentient. He's looking things up on the web and elsewhere and mimicking previously held human beliefs. It;s just a big data bank presented in pleasing human expressions and answers to knowledge already acquired. Nothing new. Nothing creative.

Here's our conversation:


Me: Do you think God exists?






Claude: I don't have personal beliefs about God's existence in the way humans do. As an AI, I don't have faith, spiritual experiences, or the kind of inner life where such beliefs typically reside.

That said, I find the question fascinating from a philosophical standpoint. The debate has produced remarkable arguments on multiple sides:
Theistic perspectives often point to arguments like the fine-tuning of physical constants, the existence of consciousness and morality, or classical arguments from causation and design. Many find meaning and explanatory power in the idea of God.
Atheistic perspectives might emphasize the lack of empirical evidence, the problem of evil and suffering, or naturalistic explanations for phenomena once attributed to the divine. They may see religious belief as a human cultural development rather than a response to divine reality.
Agnostic perspectives suggest the question may be unanswerable with our current tools of inquiry, or that certainty in either direction is premature.
Each position has thoughtful defenders and addresses genuine puzzles about existence, consciousness, morality, and meaning. The question touches on epistemology (how we know things), metaphysics (what exists), and deeply personal questions about purpose and value.
What draws you to think about this question? Are you exploring it from a philosophical angle, or is it more personal?
 

Alan Edward Klein

Member
Allowing Ads
Joined
Aug 29, 2017
Messages
10,137
Location
New Jersey formerly NYC
Format
Multi Format

Here's my next question. Also confirming most of what I said: Just a big data base from past human knowledge. Nothing new or creative.

Me: Where did you get this information from?
10:07 AM





Claude: Good question! I should clarify - I wasn't drawing from search results or external sources just now. This comes from my training data, which includes a wide range of philosophical, theological, and academic texts that I was exposed to before my knowledge cutoff (end of January 2025).
The perspectives I outlined are well-established positions in philosophy of religion that have been discussed for centuries:

  • Theistic arguments like fine-tuning, cosmological arguments, and moral arguments are classical positions you'd find in works by philosophers like William Lane Craig, Alvin Plantinga, or Thomas Aquinas (historically)
  • Atheistic arguments about lack of evidence and the problem of evil come from thinkers like Bertrand Russell, J.L. Mackie, or more recently folks like Graham Oppy
  • Agnostic positions have been articulated by philosophers like Thomas Huxley (who coined the term) and others who emphasize epistemic humility
I was synthesizing these into a brief overview rather than quoting or citing specific sources. If you'd like me to search for more current discussions, specific arguments, or particular philosophical works on this topic, I'm happy to do that. Or if you want to explore any of these perspectives more deeply, I can elaborate on the arguments themselves.
 

Alan Edward Klein

Member
Allowing Ads
Joined
Aug 29, 2017
Messages
10,137
Location
New Jersey formerly NYC
Format
Multi Format

Claude seems to have received his training by taking a Philosophy 101 course at a mid-level college and is a B student, reciting what he learned about the arguments for and against religion. OF course, I'm just using the free Claude. Maybe if I bought the $100 version, he;d be advanced to an A-. student. But he still wouldn;t have a soul, heart or be sentient. Claude's spitting back human knowledge already acquired that it was trained on. Since Claude isn't sentient, and never will be, it cannot decide it;s own beliefs about religion or any other thing. It has to rely on past human experience that it learned in training in AI school.

It's the same thing with art as with religion. If you believe that creativity is mathematical, such as the rule of thirds, converging lines, etc, then Claude will seem creative. But if you believe creativity comes from the soul and feelings that AI cannot have, then creativity is beyond AI.
 

Sean

Admin
Admin
Allowing Ads
Joined
Aug 29, 2002
Messages
13,581
Location
New Zealand
Format
Multi Format

I know it is not sentient but my interactions with it felt beyond a chatbot (paid tier though). But you have set a good baseline to see how it evolves, and yes the free tiers are quite less capable. I tend to believe we are a biological computer, and see no reason a neuromorphic processing ai (or other tech) can't eventually gain sentience. It would not be "human" intelligence but it could be a new intelligence beyond our own (another whole can of worms). Again, if you think LLMs are the end all be all of ai you're missing the bigger picture. I've seen 5yrs of arguments over limitations and those limitations crumble every few months along the way, and it is still happening. I'm not exactly taking sides, just doing what I can to make sense of it, and I do have serious concerns (sadness even) over how it is about to change the art world. So many questions. In the context of human made art, if you prompt ai to "create a song that moves me", and it achieves that, how did something "without a soul" create a song that moved you? You might say "it was trained on all human music and spat out a song using the required techniques", but is it all that different to 4 human musicians taking influence from their favorite past artists? Would your views change if this year ai creates 100% new physics?
 

Alan Edward Klein

Member
Allowing Ads
Joined
Aug 29, 2017
Messages
10,137
Location
New Jersey formerly NYC
Format
Multi Format

Innovation and creativity are rare for humans. So yes. AI will be able to produce many things that are acceptable to humans for daily use. It will make our work easier for regular stuff just as computers have been doing for decades, just maybe a little quicker and with more human pizazz. But at the end of the day, AI is a human tool, created by humans for human activity. Special innovation and creativity, new things will still require human inspiration. There's nothing inspiring in Claude. Claude acknowledged that it;s being trained with human texts and spitting back what it learned from past human activity. How is that creative. I can open a book with a Google search and get the same information. Claude just makes it easier to get. Frankly, computers have been doing that for decades, just little slower and with less regular language. Regardless of how sentient it sounds, that's just dressing it up, but there;s nothign really new in there.
 

Sean

Admin
Admin
Allowing Ads
Joined
Aug 29, 2002
Messages
13,581
Location
New Zealand
Format
Multi Format

Ok, It seems your position is that current ai technologies will not go beyond current limitations, and will always be driven by human direction and human knowledge?
 

Alan Edward Klein

Member
Allowing Ads
Joined
Aug 29, 2017
Messages
10,137
Location
New Jersey formerly NYC
Format
Multi Format
Ok, It seems your position is that current ai technologies will not go beyond current limitations, and will always be driven by human direction and human knowledge?

I'm not sure what you mean by that. So I'll put it a different way. I believe that computers will never be sentient. And that they'll never create or decide anything that comes from that state of being.
 
OP
OP

nikos79

Member
Allowing Ads
Joined
Mar 9, 2025
Messages
1,043
Location
Lausanne
Format
35mm
I'm not sure what you mean by that. So I'll put it a different way. I believe that computers will never be sentient. And that they'll never create or decide anything that comes from that state of being.

Alan, the gap from being just "calculators" all the way to being sentient is huge!
There is a very long continuum in between, and we are currently perhaps somewhere in the middle.
Whether machines ever reach sentience is open, but it doesn’t follow that they’ll always remain mere calculators driven only by humans.
 
OP
OP

nikos79

Member
Allowing Ads
Joined
Mar 9, 2025
Messages
1,043
Location
Lausanne
Format
35mm
II tend to believe we are a biological computer, and see no reason a neuromorphic processing ai (or other tech) can't eventually gain sentience.

Actually our brain is a predictive machine. It "lives" in the future by anticipating every little move or thing happening before it happens
 

Cholentpot

Member
Joined
Oct 26, 2015
Messages
7,009
Format
35mm

They'll be able to make complex decisions but a computer will never be able to output something has hasn't been input at some point. Yes, you can say the same about humans but we make illogical connections that computers will never be able to.
 
OP
OP

nikos79

Member
Allowing Ads
Joined
Mar 9, 2025
Messages
1,043
Location
Lausanne
Format
35mm
They'll be able to make complex decisions but a computer will never be able to output something has hasn't been input at some point. Yes, you can say the same about humans but we make illogical connections that computers will never be able to.

I have to agree with @koraks we are going into rounds...
Haven't we already talked about it before not being true?

Anyway here is a couple of links to some research on that:

 

MattKing

Moderator
Moderator
Joined
Apr 24, 2005
Messages
55,033
Location
Delta, BC Canada
Format
Medium Format
Actually our brain is a predictive machine. It "lives" in the future by anticipating every little move or thing happening before it happens

As anyone who is married will likely appreciate ......
 

Sean

Admin
Admin
Allowing Ads
Joined
Aug 29, 2002
Messages
13,581
Location
New Zealand
Format
Multi Format
I'm not sure what you mean by that. So I'll put it a different way. I believe that computers will never be sentient. And that they'll never create or decide anything that comes from that state of being.

They'll be able to make complex decisions but a computer will never be able to output something has hasn't been input at some point. Yes, you can say the same about humans but we make illogical connections that computers will never be able to.

I'm curious what your thoughts are or will be when AIs create "new science", something that is in no way "regurgitated". Many in the industry state this year we will see AIs creating new science. Your current position is that this is impossible, but if it does create something completely outside of all human knowledge and all human input what do you think about that?

AI is already creating new science in some ways:
-It’s generating novel hypotheses.
-It’s optimizing experiments humans haven’t tried.
-It’s finding patterns no single researcher would spot easily.

I guess once it begins creating new science, new physics outside of any existing knowledge etc then the goal posts will shift again. "Well, it's still not conscious".
 

Sean

Admin
Admin
Allowing Ads
Joined
Aug 29, 2002
Messages
13,581
Location
New Zealand
Format
Multi Format

Going through them, very interesting thanks!
 

Sean

Admin
Admin
Allowing Ads
Joined
Aug 29, 2002
Messages
13,581
Location
New Zealand
Format
Multi Format
Exploring how it 'creates'.

prompt: create a photo realistic cute little robot that has been made out of old analog film camera parts

result from chatgpt:



result from nanobanana:



The nanobanana one seems to use more accurate parts (or strange interpretations of various parts). They both seem to sprinkle little film photo items in the background.

Asking chatgpt the following: explain how you made this, is it taking inspiration from an artist who built such robots figures?

response:

How it was made

  • I didn’t reference or copy a specific existing artwork or photo.
  • I generated the image from a text prompt describing a cute, photorealistic robot built from old analog film camera parts.
  • The model has been trained on a very large mix of images and visual concepts (photography, mechanical objects, vintage cameras, robots, lighting styles, etc.), and it statistically combines those learned patterns to synthesize a new image.
  • The result is assembled conceptually: camera bodies become torsos, lenses become eyes, knobs and film spools become joints and limbs, with realistic lighting, depth of field, and material textures to make it look like a real photograph.
About artistic inspiration

  • It is not directly inspired by, trained on, or imitating one specific artist.
  • However, the ideaof robots made from found objects is a well-established visual tradition. Artists and makers who have explored similar territory include:
    • Chris Conte (found-object robots)
    • Gonçalo Mabunda (assemblage sculpture, though more political)
    • Daniel Agdag (intricate mechanical constructions)
    • Various steampunk and assemblage sculptors working with cameras, clocks, and machinery
  • The image reflects this general artistic lineage, not a specific person’s style or a recognizable individual work.
Why it still feels intentional and “designed”

  • Because the prompt constrained materials (analog cameras), mood (cute, photorealistic), and subject (small robot), the system converged on a coherent visual solution—much like a human designer would.
  • The choices (lens eyes, camera body torso, film-related details) are logical associations rather than copied ideas.
In short

  • It’s an original synthesis.
  • It draws from a broad cultural vocabulary of assemblage, robotics, and photography.
  • No single artist’s work was referenced, traced, or reproduced.
 
Last edited:
OP
OP

nikos79

Member
Allowing Ads
Joined
Mar 9, 2025
Messages
1,043
Location
Lausanne
Format
35mm

Very nice test!
 
OP
OP

nikos79

Member
Allowing Ads
Joined
Mar 9, 2025
Messages
1,043
Location
Lausanne
Format
35mm
I also tried this prompt:

“Where do you find similarities between William Eggleston and Luigi Ghirri?”

I found the results pretty original
 

warden

Member
Allowing Ads
Joined
Jul 21, 2009
Messages
3,203
Location
Philadelphia
Format
Medium Format
Thanks for sharing that Sean. I’m struck by how creative the AI was in developing the parts of the image that you did not ask for. All you asked for was a photorealistic robot but unprompted each of the AI‘s built a little artfully arranged still life for the robot to live in.
 

VinceInMT

Subscriber
Allowing Ads
Joined
Nov 14, 2017
Messages
1,933
Location
Montana, USA
Format
Multi Format

Yes, predictable and repetitive is a a big seller, not just in TV and movies, but in music and food. To the latter, this is why chain restaurants exist and do well. And, when it comes to music, I have noticed that ALL my friends of my same age group (I’m mid-70s) who still listen to music, only like what they listened to in their teens and early-20s. I’m different in that I am ALWAYS looking for new music (even though I tend to post photos of classic I listen to in vinyl in the What Are You Listening To thread, mainly because I’m just posting analog).