OOTS #1281 - The Discussion Thread

[Fish]

The fortress is useless as a final location in every scenario. Except, maybe, "the Snarl somehow can now reach into all planes and also the fortress has some sort of anti-Snarl protections and also everyone got there in time and also they still need to solve this whole Snarl issue which is now a significantly larger problem than before".

There is also the endgame scenario of "Gee, this would be a great place to store a Gate that would destroy the world. Can we move them to this plane and call it a day?"

I mean, what better place, right?

[pearl jam]

No, my position is that my brother is producing art, and that he brings much more to the table than "technique", and that my suggestion/request does not change the fact that there's intention in what he's doing. My position is that my brother produced an art piece, and I didn't. I just asked for it / financed it/ whatever. Julius II wasn't an artist, and didn't make any art. Michelangelo did.

My brother is a full human, and brings much more to the table than "drawing technique", which is itself not in any way equitable to what the AI brings forward. They have nothing in common, and the fact that a similar visual effect is produced doesn't change that.



It isn't, and I don't intend to try to provide a quick and easy definition of art that'll be sufficient for this discussion. I am saying that art is an activity that always done by a conscious mind. If we're in agreement that that is not what an AI is, and we're not trying to say that the Sistine Chapel ceiling is "an art-piece by Michelangelo, the Underwear Painter, and Julius II", then there's no art produced by these images.

And again, I obviously don't deny that pope Julius had a role to play in the ceiling painting coming into being - but I honestly believe that unless you already had a horse in this race, you wouldn't dream of calling him "one of the artists who made the paintings".



There is no "if", in my view, about intention by a human painter. Again, it's just not even remotely the same process. There is no doer in AI generated imagery. The ONLY part that can be compared is me in both scenarios. My brother and the AI aren't, and can never, be discussed as equivalent. That's my point about AI "art" not being art: if I'm the only point of comparison, and I reject the notion that by saying "hey, take a picture of something" to a rando on the street I was doing art, then there is no art to be found in the ideal form of AI generated imagery.

I wonder if it isn't more accurate to think of the AI in AI generated art not as equivalent to the painter of a painting but as the brush, paints, etc. The AI image generators don't decide they want to make and image of a cat, for example, but they are prompted to generate it by a user, as far as I know. So generating an image with AI is more equivalent to choosing paint vs. pencil or some other medium of expression. Where a painter might demonstrate skill through their use of different strokes etc, perhaps someone using AI might demonstrate it by their ability to choose the right prompts to generate the outcome they desired. (I don't have any experience with AI image generators to know how much of a role this can play, but I would suspect that it might be possible.) Under this view the images generated could be viewed as art, but the artist is the person employing the AI, not the AI itself.

[Fyraltari]

This is, I think, the common interpretation of events, and one that I held myself for a long time. But I watched a video on YouTube sometime during the last few years of corona that offered another view. Although no ring bearer exercised the same level of control over the ring that Sauron might, as the current wielder of the Ring Frodo did have the ability to wield some measure of its power. When Frodo warns Gollum on the ascent of Mt. Doom that if he ever betrays him the Ring will destroy his words have power more than just a prediction that someone might make about the winner of some sporting event, but they carry a doom that means it will come true. It's only a few days later when Gollum does betray Frodo and seize the Ring, but the doom that Frodo spoke comes to pass destroying him and the Ring together. Through the pity and compassion Frodo showed to Gollum combined with the warning not to attempt to take the Ring from him, Frodo is actually the agent of the Ring's demise, though he was unable to physically throw it in the fire himself. I thought it was an interesting interpretation and certainly seems to be in keeping with the spirit of the way things happen within Tolkien's world.

That's not Frodo speaking. Sam sees Frodo as white-clad grim figure wearing a wheel of fire upon his breast, and it is from the fire that the Doom is spoken. The Ring is forbidding Gollum from intervening further. The Ring knows that Frodo cannot resist him once he is at the Cracks of Doom and will put It on rather than casting it in the Fire. This will alert Sauron to his position and allow the Ring to come back to his Master at last. This has been the Ring's intention ever since it made Frodo volunteer for the Quest all the way in the House of Elrond. However, Gollum would not take the Ring to the Cracks, he would attempt to flee Mordor again so the Ring orders him away.

Sauron is undone not because of the valour of Men and Hobbits (well, not just) but because Bilbo, Frodo and Sam all chose to spare Gollum when they had the chance and this came back to them in mysterious ways, mercy was rewarded, but Evil only knew to curse and threaten Gollum and in doing so, unwittingly caused its own demise too.

[Emberlily]

honestly I figured bloodfeast is adamant about not going into the bag of holding because being an animal in a bag of holding that unexpectedly enters an anti-magic zone is scary, uncomfortable, or both

there might be some sort of Twist this is foreshadowing, or some way to connect it with Eugene, but I'm not seeing one at the moment, but I am seeing something that makes sense as someone who's been around a lot of pets

as to why it's being called out, making sure we're aware bloodfeast is permanently on the board now seems enough for me

[Kish]

honestly I figured bloodfeast is adamant about not going into the bag of holding because being an animal in a bag of holding that unexpectedly enters an anti-magic zone is scary, uncomfortable, or both

there might be some sort of Twist this is foreshadowing, or some way to connect it with Eugene, but I'm not seeing one at the moment, but I am seeing something that makes sense as someone who's been around a lot of pets

as to why it's being called out, making sure we're aware bloodfeast is permanently on the board now seems enough for me

I'm pretty sure the idea is that Bloodfeast is desperately trying to convey that he saw something relevant, and Belkar is misunderstanding it as "I don't want to go back in the bag."

Since this Bag of Holding, unlike official D&D ones, apparently has air, there's no reason for Bloodfeast to have noticed that the Prime Material Plane end of the bag was in an anti-magic zone; the end he was in remained on a different plane of existence from the anti-magic zone.

[brian 333]

That's not Frodo speaking. Sam sees Frodo as white-clad grim figure wearing a wheel of fire upon his breast, and it is from the fire that the Doom is spoken. The Ring is forbidding Gollum from intervening further. The Ring knows that Frodo cannot resist him once he is at the Cracks of Doom and will put It on rather than casting it in the Fire. This will alert Sauron to his position and allow the Ring to come back to his Master at last. This has been the Ring's intention ever since it made Frodo volunteer for the Quest all the way in the House of Elrond. However, Gollum would not take the Ring to the Cracks, he would attempt to flee Mordor again so the Ring orders him away.

Sauron is undone not because of the valour of Men and Hobbits (well, not just) but because Bilbo, Frodo and Sam all chose to spare Gollum when they had the chance and this came back to them in mysterious ways, mercy was rewarded, but Evil only knew to curse and threaten Gollum and in doing so, unwittingly caused its own demise too.

Gollum was the only one stronger than the ring!

(Bilbo was frightened into giving it up, but he did give it up. Gollum disobeyed it.)

[H_H_F_F]

I wonder if it isn't more accurate to think of the AI in AI generated art not as equivalent to the painter of a painting but as the brush, paints, etc. The AI image generators don't decide they want to make and image of a cat, for example, but they are prompted to generate it by a user, as far as I know. So generating an image with AI is more equivalent to choosing paint vs. pencil or some other medium of expression. Where a painter might demonstrate skill through their use of different strokes etc, perhaps someone using AI might demonstrate it by their ability to choose the right prompts to generate the outcome they desired. (I don't have any experience with AI image generators to know how much of a role this can play, but I would suspect that it might be possible.) Under this view the images generated could be viewed as art, but the artist is the person employing the AI, not the AI itself.

My point exactly is that the AI isn't equivalent to a painter, because it's not a person and not anything approaching one. However, it's also not a brush. It has no equivalent - which in my view is why it makes things that we tend to equate with art. We've never needed the heuristics it takes to differentiate art from whatever that is before.

As I've said, there'll obviously be edge cases, but saying "paradox of the heap!" and running away doesn't make a convincing argument, in my view.

Let's take this one step further to the extreme.

It's 2027, in the Smith household. The Smiths are a wealthy family, with a smart home. The smiths have a two year old child. Before bed, he says "mom, tell me a fairytale", and mom does. Usually, she reads to him from a book of fairytales she got from her great grandmother.

One night, the toddler asks mom to tell him a story. Mom says "I'm too busy, ask ChatGPT."

The Kid doesn't know what ChatGPT is, but he's familiar with it as a voice coming from the walls. He doesn't differentiate between the way his Mom talks to ChatGPT, Alexa, or her friends on the speaker phone. They're all people his mom sometimes talks to, and their voice comes from the walls. He's met some, but not others. No one ever told him ChatGPT wasn't just a weird name.

The toddler is a bit shy. He doesn't know ChatGPT very well. It's his mom's friend, and a stranger adult. But he can't fall asleep without a bedtime story.

So the kid says "ChatGPT, tell me a fairytale".

ChatGPT makes a string of coherent sentences that no one has ever heard before, with a recognizable plot and some classical themes. As far as the kid can tell, it's just another story he's never heard before. He falls asleep.

Is that story art made by the toddler? Did he write the story, using ChatGPT as a tool, equivalent to a brush?

I think not, and I think if you look at it honestly, you'll have a hard time saying otherwise.

You might not think this way on pieces of "art" that I would. You might even say that we're not there yet, and that nothing even approaching this has ever happened.

But I think you'd simply have to recognize that we're living in a world were AI can generate things that would look to us like art - but aren't.

[Kish]

I'm reminded of when someone asked ChatGPT about the creature in the darkness.

As an AI language model, I cannot provide opinions or preferences, but I can provide you with information about D&D creatures that could potentially fit the description of the "Monster in the Darkness" character from the "Order of the Stick" webcomic.

Based on the character's appearance, behavior, and abilities, there are several D&D species that could potentially fit:

Beholder: Beholders are floating, spherical creatures with numerous eyes and powerful magical abilities. They are notorious for their paranoid and manipulative nature, which could fit the "Monster in the Darkness" character's behavior. The character's ability to cast a wide variety of spells and control powerful creatures could also be reminiscent of a beholder's magical abilities.

Mind Flayer: Mind flayers, also known as illithids, are intelligent, squid-like creatures with psionic abilities. They are known for their mastery of mind control and manipulation, which could be a good fit for the "Monster in the Darkness" character's behavior. Mind flayers are also known for their ability to control and manipulate other creatures, which could explain the character's control over powerful monsters.

Aboleth: Aboleths are ancient, underwater creatures with powerful psychic abilities. They are known for their manipulation of memory and ability to control others, which could fit the "Monster in the Darkness" character's behavior. Aboleths are also known for their connection to powerful, underwater creatures, which could explain the character's control over powerful monsters.

Slaad: Slaads are chaotic, frog-like creatures with the ability to transform other creatures into slaads. They are known for their unpredictability and their chaotic nature, which could fit the "Monster in the Darkness" character's behavior. Slaads are also known for their ability to control and manipulate other creatures, which could explain the character's control over powerful monsters.

These are just a few examples of D&D creatures that could potentially fit the description of the "Monster in the Darkness" character. Ultimately, the identity of the character's species is a mystery and may be left up to the interpretation of the reader.

Based on my knowledge of things ChatGPT specifically has done and of currently functional "AI," if Kid Smith said "ChatGPT, tell me a fairy tale," ChatGPT would tell a common version of a public domain fairy tale.

If Kid Smith phrased it in a way that clearly wanted a new fairy tale, ChatGPT would combine elements of existing fairy tales in a way that might look like a coherent story but probably wouldn't. Characters would appear and disappear, unheralded.

If someone asked ChatGPT to produce a painting in the style of Leonardo da Vinci, ChatGPT would trace one of his existing works.

If someone asked ChatGPT to produce a painting of Rich Burlew, well. This one's been tried, not to my knowledge of Rich, but with lots of people. It might produce a decent though amateurish mockup up the specific person named; it might accidentally produce something grotesque and disturbing, probably in a relatively subtle way...like what the current strip references actually. And whether a human viewer was going, "Yes, I can see that's meant to look like him" or "gack, why is his neck that long and that thin?" ChatGPT would be wholly unable to tell the difference. As far as it would be concerned it would have done what was asked for.

AI isn't I, and so AI art isn't art. We're still hundreds of years out from producing anything that could pass for Lieutenant Commander Data, even in dim light, without a huge dollop of wishful thinking.

[Jasdoif]

I wonder if it isn't more accurate to think of the AI in AI generated art not as equivalent to the painter of a painting but as the brush, paints, etc. The AI image generators don't decide they want to make and image of a cat, for example, but they are prompted to generate it by a user, as far as I know. So generating an image with AI is more equivalent to choosing paint vs. pencil or some other medium of expression. Where a painter might demonstrate skill through their use of different strokes etc, perhaps someone using AI might demonstrate it by their ability to choose the right prompts to generate the outcome they desired. (I don't have any experience with AI image generators to know how much of a role this can play, but I would suspect that it might be possible.) Under this view the images generated could be viewed as art, but the artist is the person employing the AI, not the AI itself.

There's creativity in arrangement, in visions being sought, in tailoring output to fit those visions. That's why 3D artists are artists; even though the final images are produced wholly by a series of calculations, it's an artist who has the vision and adjusts the inputs of those calculations to conform to it. That's why movie directors are artists; even though it's a lot of other artists who produce what ultimately goes on screen, it's another artist who guides and arranges them together to advance a vision. That's why procedural generation is art; even though it's literally a bunch of random numbers filtered through a set of rules, it's an artist who defined those rules to produce results in line with their vision.

Ultimately, art results from arranging objective components in a way that implies a human perspective. It is entirely true that a single roll on a fully-randomly-chosen random table does not make a campaign; and similarly an image rendered without input does not make art. But this is not an inherent fact of the roll's result, or the image's nature. Rather it's the absence of the human element over it, and that can be added on...just like how adding a second random roll and tying its result together with the first can form the foundation of an entire session, or adventure, or campaign.

It's one thing to say an image isn't art. It's quite another to say that same image can't be art.

[nleseul]

If anyone should wish to put some real pretend money down on the question of the Order visiting Xykon's astral fortress, I've created a market for it over on Manifold Markets! 😉

(Rumor has it that there's an awful lot of talk about AI art over there as well...)

[Psychronia]

The Giant is just flexing with this one.

Sure, let's draw the notoriously difficulty part for the joke of calling it easy.

[pearl jam]

There's creativity in arrangement, in visions being sought, in tailoring output to fit those visions. That's why 3D artists are artists; even though the final images are produced wholly by a series of calculations, it's an artist who has the vision and adjusts the inputs of those calculations to conform to it. That's why movie directors are artists; even though it's a lot of other artists who produce what ultimately goes on screen, it's another artist who guides and arranges them together to advance a vision. That's why procedural generation is art; even though it's literally a bunch of random numbers filtered through a set of rules, it's an artist who defined those rules to produce results in line with their vision.

Ultimately, art results from arranging objective components in a way that implies a human perspective. It is entirely true that a single roll on a fully-randomly-chosen random table does not make a campaign; and similarly an image rendered without input does not make art. But this is not an inherent fact of the roll's result, or the image's nature. Rather it's the absence of the human element over it, and that can be added on...just like how adding a second random roll and tying its result together with the first can form the foundation of an entire session, or adventure, or campaign.

It's one thing to say an image isn't art. It's quite another to say that same image can't be art.

Yeah, I think the last bit sums up my feelings quite well.

As for the toddler, you could say the same about anything produced by a toddler with instruments they are capable of wielding, whether that be a brush, a pencil or just a hand or foot dipped into paint, etc. Is what they produce art? As it is, perhaps not, but if an adult then arranges it for display in some manner is it art then? Perhaps? I think like many things there are different standards based on context. I'm not necessarily convinced that AI generated images are art, but I hesitate to unequivocally say they can't be.

[RatElemental]

I know it's a pretty generic design by this comic's standards but I can't help but find Mimi's humanoid form adorable. Might be the expression in the initial panel.

[danielxcutter]

So far, I believe the "put request into AI tool and literally post the unaltered result" isn't really art, and it definitely isn't drawing. But photographs and digital formats took a while to become art too... and frankly, I don't have an issue with them being used in general, especially non-profit uses such as generating initial images for your newest character. (Especially if you don't immediately have the money to afford commissions.)

[Kantaki]

Normally they can't in 3.x:

So either a "houserule"/Mimi's a special case or maybe she's cheating by imitating a doll, marionette, mannequin or something like that.

Or the dead postman.
Corpses count as objects, right?

[otakuryoga]

how would you even use that many fingers?

[The MunchKING]

Watch out, Mimi!! That's how entire species get stereotyped as bad at shapeshifting. :p (And yes, I realize that the comic says that explicitly, but I wanted to rant about Ditto.)

Like Ditto from Pokémon! Originally it was one specific Ditto that couldn't transform its face right, and it got over its shortcomings by the end of the episode. But then "Ditto face" became a thing, and then ALL Dittos in spin-off games like Snap and merchandising couldn't transform their faces! And then it just became a thing everywhere except the main games (Where it would be extra spritework) for Dittos to not do faces.

SO your bad at hands now, but soon people will be saying ALL shapeshifters can't do their hands right, and then all of you will be FORCED to do hands wrong, because it's merchandisable!!

[brian 333]

So either a "houserule"/Mimi's a special case or maybe she's cheating by imitating a doll, marionette, mannequin or something like that.

Or the dead postman.
Corpses count as objects, right?

Or Mimi's leveled up enough to get an extra feat and she took Humanoid Shape, which enables her to use humanoid movement, tools, and communication.

[KorvinStarmast]

There is also the endgame scenario of "Gee, this would be a great place to store a Gate that would destroy the world. Can we move them to this plane and call it a day?"

I mean, what better place, right?

The gate ends up in the demi plane, and the phylactery doesn't? Is that what your post suggests?
Hmm, how does one convince Redcloak to get on board with that plan?

I'm pretty sure the idea is that Bloodfeast is desperately trying to convey that he saw something relevant, and Belkar is misunderstanding it as "I don't want to go back in the bag."

Yes, which means that Belkar's speak with animals ranger skill (which he used in the Arena¹with an Allosaurus) may be on display again soon.

¹_{Or, was that him faking it, and not actually speaking with animals?}

[Tzardok]

The gate ends up in the demi plane, and the phylactery doesn't? Is that what your post suggests?
Hmm, how does one convince Redcloak to get on board with that plan?

I don't think Xykon's fortress is a demiplane. It seems to be a physical place on the Astral.

Yes, which means that Belkar's speak with animals ranger skill (which he used in the Arena¹with an Allosaurus) may be on display again soon.

¹_{Or, was that him faking it, and not actually speaking with animals?}

You are mixing up the Handle Animal skill with the speak with animal spell.

[theNater]

My point exactly is that the AI isn't equivalent to a painter, because it's not a person and not anything approaching one. However, it's also not a brush. It has no equivalent - which in my view is why it makes things that we tend to equate with art. We've never needed the heuristics it takes to differentiate art from whatever that is before.

AI is more like a camera than a brush. Most of what it does is akin to taking photographs of one or more existing pieces of art, and laying them on top of each other.

Is that story art made by the toddler? Did he write the story, using ChatGPT as a tool, equivalent to a brush?

Yes and yes, in the same way that a toddler who opened all his favorite storybooks to random pages, photographed them, and had a parent read the photographs would have created a new story. As Kish notes, with modern AI it's likely to either be mostly one pre-existing story or largely incoherent.

As for the toddler, you could say the same about anything produced by a toddler with instruments they are capable of wielding, whether that be a brush, a pencil or just a hand or foot dipped into paint, etc. Is what they produce art?

Yes, though it may not be particularly high quality art.

[Fish]

The gate ends up in the demi plane, and the phylactery doesn't? Is that what your post suggests?
Hmm, how does one convince Redcloak to get on board with that plan?

The Order doesn’t know the fortress exists (except maybe Durkon, depending whether Thor’s joke was meant to be informative). I’m unclear on the Astral Plane and what, exactly, Xykon did to build that place, and whether it will long outlast him if he were to be completely destroyed. Is it based on Xykon’s thoughts? Is it real? Are the spells permanent? I don’t know.

In any case, my post presumes that Xykon is at least defeated, and Redcloak is converted, so he can assist in moving the gate there. The phylactery is more or less a bonus.

That is to say, I don’t think Xykon’s fortress necessarily appears as a Final Dungeon; but it could appear briefly as a location for other reasons.

[Aquillion]

I wonder if it isn't more accurate to think of the AI in AI generated art not as equivalent to the painter of a painting but as the brush, paints, etc. The AI image generators don't decide they want to make and image of a cat, for example, but they are prompted to generate it by a user, as far as I know. So generating an image with AI is more equivalent to choosing paint vs. pencil or some other medium of expression. Where a painter might demonstrate skill through their use of different strokes etc, perhaps someone using AI might demonstrate it by their ability to choose the right prompts to generate the outcome they desired. (I don't have any experience with AI image generators to know how much of a role this can play, but I would suspect that it might be possible.) Under this view the images generated could be viewed as art, but the artist is the person employing the AI, not the AI itself.

One thing which is often left out of this is that there's another, more important step.

The person who designs and trains the AI is really contributing more to it, because they're the one who defines "art" (and, indirectly, everything else) to it by choosing what to put in its training set and deciding how it is used, how it's given emphasis, and so on by the rules they set for training. This is where the human / creative aspect unique to a particular model comes from.

In that respect creating an AI model is like creating a weird multidimensional holistic collage, and the question of how independent it is from the individual items in its training set becomes complicated (and can, like a collage, vary depending on the exact process they use.)

[Provengreil]

Or Mimi's leveled up enough to get an extra feat and she took Humanoid Shape, which enables her to use humanoid movement, tools, and communication.

Instruction from Serini, perhaps? She clearly has a talent for getting her monster friends to work with her on stuff, maybe other mimics just never get the help.

[H_H_F_F]

I'm reminded of when someone asked ChatGPT about the creature in the darkness.

Based on my knowledge of things ChatGPT specifically has done and of currently functional "AI," if Kid Smith said "ChatGPT, tell me a fairy tale," ChatGPT would tell a common version of a public domain fairy tale.

If Kid Smith phrased it in a way that clearly wanted a new fairy tale, ChatGPT would combine elements of existing fairy tales in a way that might look like a coherent story but probably wouldn't. Characters would appear and disappear, unheralded.

If someone asked ChatGPT to produce a painting in the style of Leonardo da Vinci, ChatGPT would trace one of his existing works.

If someone asked ChatGPT to produce a painting of Rich Burlew, well. This one's been tried, not to my knowledge of Rich, but with lots of people. It might produce a decent though amateurish mockup up the specific person named; it might accidentally produce something grotesque and disturbing, probably in a relatively subtle way...like what the current strip references actually. And whether a human viewer was going, "Yes, I can see that's meant to look like him" or "gack, why is his neck that long and that thin?" ChatGPT would be wholly unable to tell the difference. As far as it would be concerned it would have done what was asked for.

Note that I said 2027. You're referencing limitations that aren't the point. I think it's completely feasible that by 2027, ChatGPT could respond to that kind of request by producing an amalgamation of fairytales that'd pass as a story to a casual listener. To me that doesn't change the main point:

AI isn't I, and so AI art isn't art.

Yeah, and that's true regardless of whether or not it can "pass" as art.

There's creativity in arrangement, in visions being sought, in tailoring output to fit those visions. That's why 3D artists are artists; even though the final images are produced wholly by a series of calculations, it's an artist who has the vision and adjusts the inputs of those calculations to conform to it. That's why movie directors are artists; even though it's a lot of other artists who produce what ultimately goes on screen, it's another artist who guides and arranges them together to advance a vision. That's why procedural generation is art; even though it's literally a bunch of random numbers filtered through a set of rules, it's an artist who defined those rules to produce results in line with their vision.

Ultimately, art results from arranging objective components in a way that implies a human perspective. It is entirely true that a single roll on a fully-randomly-chosen random table does not make a campaign; and similarly an image rendered without input does not make art. But this is not an inherent fact of the roll's result, or the image's nature. Rather it's the absence of the human element over it, and that can be added on...just like how adding a second random roll and tying its result together with the first can form the foundation of an entire session, or adventure, or campaign.

It's one thing to say an image isn't art. It's quite another to say that same image can't be art.

So, I mentioned edge cases a lot here, so it's time to make it clear that I'm definitely not saying AI can't be used while making art. I'm not saying AI generated imagery cannot be used in an art-piece. I'm saying "AI art" isn't a thing, and that we're currently comfortable calling plenty of things that aren't art art, and provide categorical defenses of AI "art" that don't stand to reason.

As for the toddler, you could say the same about anything produced by a toddler with instruments they are capable of wielding, whether that be a brush, a pencil or just a hand or foot dipped into paint, etc. Is what they produce art? As it is, perhaps not, but if an adult then arranges it for display in some manner is it art then? Perhaps? I think like many things there are different standards based on context. I'm not necessarily convinced that AI generated images are art, but I hesitate to unequivocally say they can't be.

I honestly can't grasp the comparison. At all. In my example, the toddler didn't grab a tool and randomly swing it. There is no discussion to be made on whether or not something was intentional. He just asked an adult to tell him a story. There is no differentiation in his intention, nor in the actuality of his action, to when he's asking his mother to do so. If that's "debatebly art, toddlers are an edge case", than him asking his mother to read him a story is also "debatebly art".

Change ChatGPT to a competing future system called "Allen", make the smith kid a 60 year old man that wasn't properly given context for what was going on and thought he was on a voice-call with one of his daughter's friends, whatever. I just don't think your argument works, or that the comparison is apt.

I just looked up "beautiful art" on google. Clicked the first option that came up:



Did I just do art? I wanted to have "beautiful art" to post here. Do I have some claim to being an artist involved in creating this piece?

If not, then I can hardly see how going to Dall-E 2 for the exact same purpose, and doing the exact same thing, changes things.

To give an actual example: this is an Iron Chef entry of mine, from way back. I'd say the story I wrote there qualifies as an artistic endeavor, even if not one I'm particularly fond of.

You could also perhaps claim that I "did art" while deciding on the arrangement of my entry - at least, when I busied myself with parts of that task for purely aesthetic purposes. One of the things I wanted for the entry was a picture. I thought (and think) that a picture can greatly complement an optimization competition entry. It makes the character pop-out, it helps imbue a certain vibe, and it also makes the arrangement itself more easy for digestion - it provides sort of a break point, an opportunity to reflect on what you've read so far, and to prepare yourself for the lion's share of the entry ahead.

So I googled "bald woman with a sword" or something like that, and picked the photo that best suited me (despite being way too horny, as I mention in the entry). I picked out a drawing that suited what I was doing best, and combined it into the work.

Now, that drawing is definitively a part of an art-piece by me. No question. But if you right-click the image, and select "open image in a new tab", and sit there looking at this - you're not looking at a work by me, in any sense. Whatever intention I may have had, what you're looking at is not an art-piece I made, or took any part in making. Me spending time and effort on google, with artistic intention, isn't relevant.

That doesn't change if instead of google, I would've used an AI. There'd be an image there. The image would be a part of an art piece by me. And if you were to right click it, and select "open image in a new tab", you'd no longer be looking at an art-piece by me. You'd no longer be looking at an art-piece by anyone.

You'd no longer be looking at art.

[Mic_128]

Or Mimi's leveled up enough to get an extra feat

No, it's extra fingers that she has.

[Fyraltari]

No, it's extra fingers that she has.

You don't know how tall she used to be!

[Quizatzhaderac]

Epic-level Wizard: It appears to be a common Holy Symbol of the clergy of the Dark One. Not one particularly fancy or anything. The kind an acolyte would get the day they receive their White Cloak.
Epic-level Paladin: So this is...
Epic-level Wizard: ... completely useless to anyone who isn't a Cleric of the Dark One, yes. And those who are probably already have their own.
Epic-level Rogue: You're telling me this is a masterly crafted replica of a random piece of junk gear?
Epic-level Wizard: Of a very specific piece of junk gear. The divination tells me that most of effort went into simulating the wear and tear.
Epic-level Rogue: Who the hell came up with this adventure!?

Reminds me of the Dwarf Fortress Let's play Boatmurdered:

We all know our people like to engrave historical events, so I've included a few charcoal rubbings of some typical engravings here:

Apparently the 2 most significant historical events here in Boatmurdered are elephants and cheese. Take a close look at the cheese ones actually, they aren't even carvings of cheese, but renditions of some other image of a cheese. They're freaking homages!

Intrigued, I investigated the art history of the settlement further. I discovered this artifact which I can only presume was the inspiration for all the imitators.

Having viewed it for myself, I must agree that this image of a cheese speaks to the dwarven spirit, and will be a cultural treasure for generations to come.

[Precure]

Telling AI to draw something is like Stan Lee telling Jack Kirby to draw him a comic book, except, in this case, Kirby is tracing from other comic books.

[Peelee]

I just looked up "beautiful art" on google. Clicked the first option that came up:



Did I just do art? I wanted to have "beautiful art" to post here. Do I have some claim to being an artist involved in creating this piece?

No, because you just shared something. If you altered it in any way, then yes, you'd have just done art, as you put it.

I am firmly against gatekeeping art and music. Not to panty with too wide a brush here, but that has historically been bursting with elitism. Look at avant-garde art and music. Pollock tossing paint on canvas. Cage sitting at a piano for four and a half minutes. Once art and music becomes easily accessible to the common people, reproductions of works readily available, elitism demands that there be some form of it that is above the reach of those plebians. Nobody can fling paint like Pollock. Every performance of 4:33 is different because of the ambient sounds, you may have a recording of it but that is a poor substitution for experiencing it live in concert. The plebians would not understand the music of laying your forearm down on the piano to create what would otherwise be discordant noise, having a composer changes it and people purchase tickets to hear it.

So yeah. If all that bull**** is art, then so is AI art. Even not being actual intelligence, even just taking existing inputs and altering them, screw it, it's all art. Open the gates.