I’m not concerned about AI outcompeting competent writers or impacting my career directly. I am deeply concerned about AI swamping submission systems and destroying the ability of editors and readers to find the next generation of writers.
AI is very much a danger to the long term health of the field not because of competition for quality readable fiction but because of its ability to create dreck in previously unimaginable quantities and drown submission systems and indie publishing in shit.
@nxylas @thecharmingcompany @KellyMcC editors you say...
...a dying job description, replaced by software.
@nxylas @KellyMcC Does this mean there's an opportunity to jump ahead of AI submission queues by creating a compelling, personalized & inventive submission package?
Derek Sivers wrote about the Captain T conspiracy theory submission pack:
(And Nick, since I think I recognize your name - I'm imagining a demo CD that comes with a tiny glitter sparkle Porcupine toy, a little cardboard robot, and a postcard story about a Robot that won't Obey and wants to be a Porcupine.)
@thecharmingcompany @KellyMcC one doesn't need a machine to be intelligent, just the ability to emulate such.
Which is precisely the point we have reached.
As technology advances, emulation will far outstrip actual intelligence, of which, humanity (on average) seems to have a hard enough time demonstrating itself.
It was only a few years ago, that people were saying that AI wouldn't be creative, that out of all the jobs threatened, artists and writers would be unaffected.
That didn't last long.
Imagine where things will be in another few years.
@KellyMcC
This is my concern about prose/poetry and art, exactly.
A publisher knows good, viable writing. They can get to know when generic swill is being thrown at them. Same with art directors.
But the submission system is already swamped with legitimate writers and artists. The rate at which A.I. prompters can churn out generic mediocre writing and art and submit it is leagues faster than any human creator can even act.
The system will have to change, and it will likely be worse for us.
@KellyMcC
Its just spam these A.I. prompters are delivering. And you know what happens in computing when spam is detected in a system? The source is blocked.
So this could mean that legitimate talent, both writers and artists, get NO chance at all, and we're back to the system of "you gotta know someone" to even get your work seen.
Which is much, much worse.
@KellyMcC Sorry. At some time in the future perhaps. If by AI you mean the LLM models, sorry no.
LLM like GPT3 (ChatGPT), have a context of around 2K-4K tokens, which is a little less in words.
So yes, LLM are made to generate human-style text, and they are good at that. But they can only take so much context into account when predicting the next word.
So anything beyond short stories or paragraphs will have dramatic continuity issues.
@skry @KellyMcC @clarkesworld Purely out of curiosity, how do they know which submissions are LLM based?
Sounds like these guys have solved a problem that not even OpenAI can solve for ChatGPT.
Would be great if they could share their genius method to detect LLM output, that does not have these serious 2 digit percentages rates of false positives and false negatives.
@yacc143 @skry @clarkesworld My understanding is that it's pretty obvious to a human reader*, but that takes enormous amounts of time to implement.
*I believe the reason for that is that it's good at creating a sentence that logically follows the last, and all right out doing that paragraph to paragraph, but absolute shit at creating whole stories that make sense as stories.
"I am deeply concerned about AI swamping submission systems and destroying the ability of editors and readers to find the next generation of writers."
As someone with severe ADHD, online application processes often take too much executive function to jump through all the stupid little hoops. As systems like these get more and more hoops that try to sort out the humans from the robots, people like me will be silently excluded without anyone noticing because we will never even apply.
@KellyMcC said the cobbler about making shoes at the beginning of the industrial revolution. Yet here we are, most of us wearing factory made shoes rather than better custom made shoes because of the economics of the situation.
The quality could suffer, but the cost and availability will compensate. You have two choices, be one of the very few people who continue to hand make custom shoes for the rich or learn to use AI to create stories in a more productive fashion.
@KellyMcC Don't worry. Soon editors will have an AI to scan their inbox.
Damn, I don't even know if I'm joking or not.
@KellyMcC
"Swamping submission systems" hits the nail dead-on. The threat of #AI / #ChatGPT systems to the writing /creative pipeline/lifecycle isn't (so much) that it "kills the crops" as that, unchecked, it has potential to "kill the soil", the cultivation of new writers/creatives.
If social media taught us anything, we need to up our systems thinking game, AND FAST, because we so rarely think about secondary effects much less stave them off.
@PixelJones @KellyMcC I may be late to this, but isn't this one of those things where we are (or perhaps should) all be going, "always, always, check primary sources. Whether it's Wiki, or Chat-whatevs, or anything.
"What are your sources for that?" should almost always be question #1.
@bytebro @KellyMcC Right. "Check your sources" is always good advice but there's a huge chasm of effort between reading/scanning for information and formally _vetting_ it.
Before #ChatGPT (broadly speaking) you could look at a coherent, literate-sounding article & assume no human would go to the effort to produce coherent rubbish. You'd tacitly trust it & only vet it further if needed.
Now, vetting _all_ content _first_ is an added cognitive burden.
Add to it the fact that editors and scouts are already underpaid and overworked, and the introduction of AI manuscripts (weird word considering there’s no "manual" behind it but we get it) becomes a burden multiplier
Yeah, this is really terrible.
One very present danger from "AI" — well, LLMs — is spambots, not the Terminator.
For editors — really anyone who consciously cares about words & meaningful communication — LLMs make the challenges they were already facing worse.
We've already seen the dangers of misinformation through social media since the pandemic started.
We will find it increasingly challenging to trust *anything* that we read *anywhere*.
@KellyMcC Well said.
"AI is very much a danger to the long term health of the field not because of competition for quality readable fiction but because of its ability to create dreck in previously unimaginable quantities and drown submission systems and indie publishing in shit."
Iirc Clarkesworld and Kaleidoscope have both had issues with this already.
Boomy generated 14 million tracks and dumped it on Spotify. Let that sink.
That 14% of ALL recorded music. In history. Let that sink deeper.
...
It's like the brooms in Fantasia, isn't it.
https://www.musicbusinessworldwide.com/ai-music-app-boomy-spotify-stream-manipulation/
pmac_
Kelly McCullough
John Hayden
Martin Johnstone
Nick Xylas
Allie Potts
Court Cantrell does not comply
Miss Rocks
Cookiefiend
Kohan Ikin
BlackPixelDust: Art 
Andreas K
skry
Alonealastalovedalongthe
Stephen Cerruti
Mark H-B
Leia
Stuart Langridge
Jed 🚲
PixelJones
Bytebro 🇬🇧 🇺🇦 🇬🇱
Clockwork ☃️✒️
Craig Nicol
Dhavide Aruliah
bk
Wampus
Kai 🐈⬛
Sheila the Wonderbink
Brian Kolm
surfie
🌀bedient （ ͜.人 ͜.）
The Frog
Scott Gerhardt