I think there's probably some discussion about using AI to write fiction that should be had.
But.
Passing off your AI output as "human written" should be punished somehow. As a fiction reader, I don't care what ChatGPT has to say, even if you think you prompted it into a publishable story. I want to read stories written by humans, and I want to reliably be able to tell which are and which aren't, so I don't spend half my attention while reading trying to work out if this is an AI or not.
I have no clue how to make that come true. Part of me is sitting here thinking "I'm glad I'm old enough that I can spend the rest of my life reading fiction written before about 2020 and never run out of genuinely great human written stories."
It may come down to "showing your work". The post talks about how they investigated the two suspect submissions by asking them for evidence of their process, but none was forthcoming.
It may just start to become the norm for publishers to do this for all submissions, like checking references for a prospective employee.
I have very little physical evidence of any book I've ever worked on, but I'm sure I could find heaps of digital versions and notes. I hope this would be good enough although obviously it could be falsified by someone determined enough I guess (note to self, start writing notes by hand)
While watching it, I thought about AI generated content.
I have never personally met anyone who worked on any of the Toy Story movies. I know, from documentaries etc, who Brad Bird and John Lasseter are. I've also watched the video [0] about how Toy Story 2 almost got deleted (which I highly recommend if you are in storage, DevOps or SRE).
There are other movies like the Wild Robot that:
- had big emotional impact on me (b/c I'm a parent)
- are 100% animated
- I have zero idea who made them
I say all of this b/c at some point, most people don't care if the movie was hand painted cels, CGI done by humans or fully AI generated with human text prompting. If the feeling is strong, people will have a "bigger" reaction which in turn will make it more memorable. It's all basically on a spectrum of "humans using tools". They care about how the movie makes them feel. It started with humans spitting charcoal at a wall [1] and now it's linear algebra.
>>I say all of this b/c at some point, most people don't care if the movie was hand painted cels, CGI done by humans or fully AI generated with human text prompting. If the feeling is strong, people will have a "bigger" reaction which in turn will make it more memorable.
I think in the near future, people will start filtering out movies that have a non-trivial share created by AI. (It remains to be seen what "non-trivial" means in this context.) I think a movie 100% generated by AI won't succeed. The story, the emotional impact, etc. all may be good, but people won't give it a chance.
> I think in the near future, people will start filtering out movies that have a non-trivial share created by AI.
I'm not _sure_ that's true? "Most people" know that McDonalds don't make the best burgers in their city - probably not even the best burgers in their neighbourhood. But they do make burgers that are reliably the same and for heaps of people "good enough" when things like price and availability and consistency are taken into account. And I don't blame then for that, for those people burger quality is a think that value less then me, and convenience is something they value more than me. They're not "wrong" just different.
Some of us will. I'm more concerned personally with books and music, and go well out of my way to make sure I'm buying then from "real people"(tm). The last few years most of my in cinema movie watching has been at a local cinema that screens classic movies. Recently I've seen a 70mm print of 2001 A Space Odyssey, 35mm prints of Blade Runner and Godfather. Last week we went to see the original Matrix I _know_ I'm not watching much AI generates film.
If we ever reach a point where we have that equivalence in media it'll be a sad day - when you can buy a ticket to a cheap but perfectly good movie written and directed by AI or you can pay a premium for something written by a human.
What does "100% generated" mean? Do you mean a movie executive told Claude to "make me a 100 million dollar blockbuster, make no mistakes" or do you mean that folks use generative tools to help write the screenplay, generate scenes, etc, or something else?
It's been weird finding out how many people don't see media past its finished surface text. People for whom the subtext of its creation is so unimportant that the concepts of coming up with a prompt and creating a work are mutually fungible.
Earlier today I pointed someone toward Sade's song Young Lion.
It's a really good song all on it's own.
But if you take time to not only listen and understand the lyrics, but to find out the backstory of who she wrote that song for and why, it's a much more powerful story and a way more meaningful song.
AI can't make me feel that way. You can't have empathy with the author if there is no author. Whoever prompted some AI song/story/movie is more like the editor or producer than an author, and Sade's song can't make me feel empathy for her band or recording engineer.
Well, Death of the Author has been a concept for quite a while. How it's made and what the author intended is explicitly not a concern for a lot of criticism and analysis.
I disagree with your assertion that "most people don't care if the movie was…fully AI generated with human text prompting". And more than disagree, I believe conclusive proof is out there that people absolutely do care.
So interesting to read about AI perceptions from the other side. In my mind, the problems with these drafts were moreso that the writers using AI could not (or would not) actually engage and improve them when given feedback, not necessarily that their writing process did not match the traditional one.
I'm not sure I see any inherent problem with publishing books written with the help of AI. As with software, I don't really care much how it's made, I care bout my experience with the finished product.
"Is it worth the $?" is ultimately the question that will be asked of anything one pays for, regardless of how exactly it was produced.
I think there is an interesting use case for AI in fiction, but it's not spamming publishers and readers with slop books. The thing about fiction is we already have SO MUCH and the good stuff is evergreen, so the number one metric that matters is quality not quantity.
I think if people want to use AI in fiction, the most promising area would be dynamically reactive story telling. Kind of a "choose your own adventure" but where the choices aren't fixed. Of course, that's a _hard_ problem, and I don't think inference is nearly cheap enough yet to broadly commercialize it, but it's a way better use of AI then "publish 300 books to try to skim off the long tail"
> The community seemed to agree because stories flooded in: 606 wild, furious tales from queer writers around the world. We had magical punk rockers, trans weredingoes eating cops, zombie gays refusing to be buried. Cutting it to just twenty stories was a challenge, but we were punching the air.
I can't help but feel that what they're asking to be submitted is just another kind of slop. Unsurprising that they get ai submissions as a result
I think there's probably some discussion about using AI to write fiction that should be had.
But.
Passing off your AI output as "human written" should be punished somehow. As a fiction reader, I don't care what ChatGPT has to say, even if you think you prompted it into a publishable story. I want to read stories written by humans, and I want to reliably be able to tell which are and which aren't, so I don't spend half my attention while reading trying to work out if this is an AI or not.
I have no clue how to make that come true. Part of me is sitting here thinking "I'm glad I'm old enough that I can spend the rest of my life reading fiction written before about 2020 and never run out of genuinely great human written stories."
It may come down to "showing your work". The post talks about how they investigated the two suspect submissions by asking them for evidence of their process, but none was forthcoming.
It may just start to become the norm for publishers to do this for all submissions, like checking references for a prospective employee.
I have very little physical evidence of any book I've ever worked on, but I'm sure I could find heaps of digital versions and notes. I hope this would be good enough although obviously it could be falsified by someone determined enough I guess (note to self, start writing notes by hand)
Technicality everyone can do that in perpetuity :)
Took my kids to see Toy Story 5.
While watching it, I thought about AI generated content.
I have never personally met anyone who worked on any of the Toy Story movies. I know, from documentaries etc, who Brad Bird and John Lasseter are. I've also watched the video [0] about how Toy Story 2 almost got deleted (which I highly recommend if you are in storage, DevOps or SRE).
There are other movies like the Wild Robot that:
- had big emotional impact on me (b/c I'm a parent)
- are 100% animated
- I have zero idea who made them
I say all of this b/c at some point, most people don't care if the movie was hand painted cels, CGI done by humans or fully AI generated with human text prompting. If the feeling is strong, people will have a "bigger" reaction which in turn will make it more memorable. It's all basically on a spectrum of "humans using tools". They care about how the movie makes them feel. It started with humans spitting charcoal at a wall [1] and now it's linear algebra.
0 - https://www.youtube.com/watch?v=8dhp_20j0Ys (Toy Story 2 almost lost)
1 - https://youtu.be/6tn3bMbm5uw (humans painting on walls0
>>I say all of this b/c at some point, most people don't care if the movie was hand painted cels, CGI done by humans or fully AI generated with human text prompting. If the feeling is strong, people will have a "bigger" reaction which in turn will make it more memorable.
I think in the near future, people will start filtering out movies that have a non-trivial share created by AI. (It remains to be seen what "non-trivial" means in this context.) I think a movie 100% generated by AI won't succeed. The story, the emotional impact, etc. all may be good, but people won't give it a chance.
> I think in the near future, people will start filtering out movies that have a non-trivial share created by AI.
I'm not _sure_ that's true? "Most people" know that McDonalds don't make the best burgers in their city - probably not even the best burgers in their neighbourhood. But they do make burgers that are reliably the same and for heaps of people "good enough" when things like price and availability and consistency are taken into account. And I don't blame then for that, for those people burger quality is a think that value less then me, and convenience is something they value more than me. They're not "wrong" just different.
Some of us will. I'm more concerned personally with books and music, and go well out of my way to make sure I'm buying then from "real people"(tm). The last few years most of my in cinema movie watching has been at a local cinema that screens classic movies. Recently I've seen a 70mm print of 2001 A Space Odyssey, 35mm prints of Blade Runner and Godfather. Last week we went to see the original Matrix I _know_ I'm not watching much AI generates film.
If we ever reach a point where we have that equivalence in media it'll be a sad day - when you can buy a ticket to a cheap but perfectly good movie written and directed by AI or you can pay a premium for something written by a human.
When the human written books cost more.
What does "100% generated" mean? Do you mean a movie executive told Claude to "make me a 100 million dollar blockbuster, make no mistakes" or do you mean that folks use generative tools to help write the screenplay, generate scenes, etc, or something else?
Clearly there are a lot of humans who derive specific feeling knowing art was created by another human, and not a calculator.
It's been weird finding out how many people don't see media past its finished surface text. People for whom the subtext of its creation is so unimportant that the concepts of coming up with a prompt and creating a work are mutually fungible.
Yep.
Earlier today I pointed someone toward Sade's song Young Lion.
It's a really good song all on it's own.
But if you take time to not only listen and understand the lyrics, but to find out the backstory of who she wrote that song for and why, it's a much more powerful story and a way more meaningful song.
AI can't make me feel that way. You can't have empathy with the author if there is no author. Whoever prompted some AI song/story/movie is more like the editor or producer than an author, and Sade's song can't make me feel empathy for her band or recording engineer.
Well, Death of the Author has been a concept for quite a while. How it's made and what the author intended is explicitly not a concern for a lot of criticism and analysis.
I disagree with your assertion that "most people don't care if the movie was…fully AI generated with human text prompting". And more than disagree, I believe conclusive proof is out there that people absolutely do care.
I got the feeling watching that last season of Stranger Things that the storyline and some of the dialogue felt AI generated. It sucked by the way.
[dead]
So interesting to read about AI perceptions from the other side. In my mind, the problems with these drafts were moreso that the writers using AI could not (or would not) actually engage and improve them when given feedback, not necessarily that their writing process did not match the traditional one.
I'm not sure I see any inherent problem with publishing books written with the help of AI. As with software, I don't really care much how it's made, I care bout my experience with the finished product.
"Is it worth the $?" is ultimately the question that will be asked of anything one pays for, regardless of how exactly it was produced.
I think there is an interesting use case for AI in fiction, but it's not spamming publishers and readers with slop books. The thing about fiction is we already have SO MUCH and the good stuff is evergreen, so the number one metric that matters is quality not quantity.
I think if people want to use AI in fiction, the most promising area would be dynamically reactive story telling. Kind of a "choose your own adventure" but where the choices aren't fixed. Of course, that's a _hard_ problem, and I don't think inference is nearly cheap enough yet to broadly commercialize it, but it's a way better use of AI then "publish 300 books to try to skim off the long tail"
> The community seemed to agree because stories flooded in: 606 wild, furious tales from queer writers around the world. We had magical punk rockers, trans weredingoes eating cops, zombie gays refusing to be buried. Cutting it to just twenty stories was a challenge, but we were punching the air.
I can't help but feel that what they're asking to be submitted is just another kind of slop. Unsurprising that they get ai submissions as a result