431
OpenAI introduces Sora, its text-to-video AI model
(www.theverge.com)
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
this post was submitted on 15 Feb 2024
431 points (95.8% liked)
Technology
59654 readers
2732 users here now
This is a most excellent place for technology news and articles.
Our Rules
- Follow the lemmy.world rules.
- Only tech related content.
- Be excellent to each another!
- Mod approved content bots can post up to 10 articles per day.
- Threads asking for personal tech support may be deleted.
- Politics threads may be removed.
- No memes allowed as posts, OK to post as comments.
- Only approved bots from the list below, to ask if your bot can be added please contact us.
- Check for duplicates before posting, duplicates may be removed
Approved Bots
founded 1 year ago
MODERATORS
If the wind blows, cut up pieces of art magazines around and then land in a pile. That isn't art. It's just cut-up pieces of someone else's art.
If a person cuts up a magazine and pieces the parts together with intention and meaning. That can be art.
Art is not "I like this visially", art is not "you did this well." Art is human expression.
I can't really agree with this example. I think you're suggesting the AI is completely independent of human expression and is completely random in its application of its training data (the cut up pieces I suppose?)
Generative AI is driven by a human prompt (description) and refined by further prompts which pushes the result in the direction of the prompters vision.
This is in essence what is occuring above. I view this process as someone being provided a chisel and a block of stone:
-Michelangelo
As I suggested above AI is a tool that makes accessing art and expression available to anyone. The Ai is the chisel. They cut the stone with words.. It isn't just random clipart being thrown around either: The 'stone' is the culmination of all of the art the model has 'seen.' It has taken that data and found the patterns that different styles contain. You might describe this as the distillation of human expression into something new.
The source is art - human expression The prompt gives it form - human expression Further prompts drive the form to fit the users vision - human expression
There is intent and meaning.
Is it art in the traditional sense? Perhaps not in the same vein as ink and canvas but ... I believe, while it is certainly rough and unrefined, it can still be considered a tool to create art.
If you want, you can say that "prompt engineering" is an art. The act of engineering that prompts to get a picture, maybe that has a skill we might call art.
But no, the jpeg isn't art. It's a million cut-up images formed to make our monkey brains go "I enjoy".
Do you do this prompt engineering? The last time I had this conversation it turned out I was talking to someone that called themselves an artist because they put words into an ai.
Let's see if we can keep this civil, shall we?
First and foremost the model isn't compositing bits and pieces of other pictures - it's predicting what the next pixel should look like based on its training data. It is generating the image. In laymans terms: it's drawing based on what it has 'learned' by looking at other art. It's pretty interesting honestly.
I do have a background in art, though it is not my profession. Regardless of that- there are no requirements to create nor appreciate art.
A few good excerpts from wikipedia:
Everyone is entitled to their opinions. Ours seem to differ- and that's fine. My views are simple: if someone can express themselves through a medium- it is a form of art.
Prompt engineering may be a form of expression, but the ai generated images composed of copying what it saw previously and repeating is not art. It has no humanity in it. The brush strokes have nothing to say. It is really no different than the wind blowing about cut-up versions of other people's art.
If someone intentionally opens a window that allows the wind to blow in and shuffle up the cut-up art into a new image, that may be performance art, the act of opening the window. But the final result, which has no humanity to it, is not art. Will never be art.
Prompt engineering is no different than opening that window and then letting the ai wind shuffle up everything it knew over and over until you find an esthetic you like. It's not art. It will never be art. Because it fundamentally can not be art.
I know I expressed this already but the wind analogy doesn't work here. It isn't random nor undirected.
As far as copying goes - considering your staunch stance on what is and isn't art I think it's fair to say you have some involvement with it.
Regardless of the medium we all start the same way. Imitation. In traditional art we are trained by observing what the masses find pleasing. When we observe most artists work we can identify these roots. Very few artists art is not based in the works of those before them.
This article does a fine job of expressing the above.
AI assisted (generative) art is a tool that provides a user access to a compendium of learned styles. It lowers the barrier of entry to express yourself through art.
I posit that this is such a divisive topic because there is so little difference between how we learn and how these models do. It garners a lot of the same negativity that a prodigy might. "Why is it so easy for them when I worked so hard. They don't appreciate it as much as me."
In the end art belongs to nobody and everybody. Art is amorphous; formless. Art and artistic expression can exist anywhere- even here. I personally am not so high minded to gatekeep such a broad field.
I have no involvement with art. I have a lot of involvement with ai. I know what it makes is not art because I make the sausage, and that sausage is nothing but blown about and recombined shredded versions of everyone else's art.
Ai doesn't learn. That's marketing. It can't reason. It can't create. It can only generate an endless stream of anything but art. It's the antithesis of art.
I won't respond to your next comment for what it's worth. This conversation stopped having value five replies back.
Enough value to respond though. Interesting.
Be that as it may: considering your involvement you should have known the differences between random copy and paste and pixel prediction. I'm afraid I doubt your claims. Your views on art were pretty polarized - I'm pleased to have provided contrast to them.