For the past 5-6 years I've been writing a book in my spare time. The outline of it is how reason emerges in past societies from the needs of social complexity, how it's lessons get converted into rules and rituals, which in turn remove any competitive advantage of aquiring reason, ending it to setup a new cycle. And in the meanwhile LLMs became the ultimate heuristic of humanity.
I've gotten it 60-70% ready, and I really don't know if it'll have an audience in a post-AI world. I never meant to strike big with it, but I'm now wondering if thousands of hours of research and writing can amount to more than a novelty gift I'd give to friends.
I really hope you don't give up. I've built and shipped systems with 80000 to 90,000 lines of code, only to see most of the companies that bought them go out of business. But there was still immense value in the act of delivering them. By putting our creations out into the world, we connect with each other.
The act of writing and building is, in itself, humanity's grand narrative for trying to understand the world. The journey itself is inherently valuable. Isn't the ability to organize our thoughts, pass them down to the next generation, and continue that narrative exactly what makes us uniquely human?
Even if only a few people around you end up reading it, those few could be deeply inspired to go on and build an even greater world. Please don't stop. I'm rooting for you.
Why wouldn’t people read it? The problem is not writing - the problem is finding an audience. Yes, AI means there are more poor works competing. But if it’s actually good writing, you will find an audience if you market the book. Pay an editor. Publish to kindle. Pay for marketing. Get people to sign up for an email list.
> Why wouldn’t people read it? The problem is not writing - the problem is finding an audience.
And that's precisely the issue here. For a while, the internet allowed you to find an audience, just like that. Start a blog / podcast / YT channel, keep going, get enough attention. You could then approach a traditional publisher and tell them "hey, I'm kind of a big deal", or you could self-publish and rely on the word-of-mouth from your followers.
Now, how would that work? If you have a blog, AI answers will summarize it without attribution and not send anyone your way. Even the "references" cited in AI answers often point to AI-slop blogs, not the original source. The articles we discuss on HN are often AI-written too. So yeah, it's about reaching the audience, but you're now competing with machines that produce an endless stream of human-like text, good enough for most consumers, practically for free.
I wondering about other thing - I'm assuming that publishers are now using LLM's en-masse to "proofread", do initial evaluation, do editorial work etc.
So if an author abstained from using LLM in the writing process - isn't then new, original, not yet on the market book ending up in the LLM training data corpus even before it hits the market?
It's a bad short story. Whether it was generated at the hands of humans or a computer isn't really relevant. Speaking in riddles is cheap now, so authors better learn how to surprise the reader in simpler, more readable words. It seems to be something that LLMs are quite bad at.
Mixed metaphors which sound nice at first glance, but slip away from meaning like an echo chasing itself off a cliff.
Similes that catch in your mind like river trouts tangled in the roots of a redwood tree.
Also mentions some interesting AI tells, for AI generated stories.
It can be "playful" once or twice, especially if the text is playing with how nonsensical the metaphor is. By the twentieth time something is as tired as a willow tree on a Tuesday in May, as persnickety as an ant with a fever, or as rambunctious as a horny lobster, it's just nonsensical bad writing.
Metaphors are generally used to transfer the qualia of one experience into another. When the referent has no qualia, that is, you've never in your life experienced "river trouts tangled in the roots of a redwood tree", it's a failure of a metaphor.
And as much as there are still people on HN who insist that AI text can't be detected algorithmically, it's worth noting that the original story is marked 100% AI by Pangram. So it's not just this person seeing things.
I recently built and delivered an AI-driven novel-writing program. The architecture involved chaining the Claude, GPT, and Gemini APIs together so they could cross-critique and iteratively revise the text, while systematically saving key plot points and lore chapter-by-chapter. (Serialized 'web novels' published on a per-chapter basis are a massive industry here in Korea). AI has already heavily infiltrated the fiction space in Korea, and it looks like the exact same trend is hitting the US.
Personally, I find it incredibly easy to spot AI-generated text in Korean, but catching it in English is much harder for me. That being said, they still have very distinct, overused patterns. You constantly see words like 'ultimately' or 'structural,' and they rely heavily on highly formulaic 'X is Y' sentence structures.
The worst thing about this is not really that somebody might have had AI help writing a story, but that an editor thought they could get any kind of useful information from asking an AI whether the story was written by AI. If there's any hope of editors staying ahead of this phenomenon, they will need to educate themselves a lot better about how it actually works.
For the past 5-6 years I've been writing a book in my spare time. The outline of it is how reason emerges in past societies from the needs of social complexity, how it's lessons get converted into rules and rituals, which in turn remove any competitive advantage of aquiring reason, ending it to setup a new cycle. And in the meanwhile LLMs became the ultimate heuristic of humanity.
I've gotten it 60-70% ready, and I really don't know if it'll have an audience in a post-AI world. I never meant to strike big with it, but I'm now wondering if thousands of hours of research and writing can amount to more than a novelty gift I'd give to friends.
I really hope you don't give up. I've built and shipped systems with 80000 to 90,000 lines of code, only to see most of the companies that bought them go out of business. But there was still immense value in the act of delivering them. By putting our creations out into the world, we connect with each other.
The act of writing and building is, in itself, humanity's grand narrative for trying to understand the world. The journey itself is inherently valuable. Isn't the ability to organize our thoughts, pass them down to the next generation, and continue that narrative exactly what makes us uniquely human?
Even if only a few people around you end up reading it, those few could be deeply inspired to go on and build an even greater world. Please don't stop. I'm rooting for you.
Why wouldn’t people read it? The problem is not writing - the problem is finding an audience. Yes, AI means there are more poor works competing. But if it’s actually good writing, you will find an audience if you market the book. Pay an editor. Publish to kindle. Pay for marketing. Get people to sign up for an email list.
> Why wouldn’t people read it? The problem is not writing - the problem is finding an audience.
And that's precisely the issue here. For a while, the internet allowed you to find an audience, just like that. Start a blog / podcast / YT channel, keep going, get enough attention. You could then approach a traditional publisher and tell them "hey, I'm kind of a big deal", or you could self-publish and rely on the word-of-mouth from your followers.
Now, how would that work? If you have a blog, AI answers will summarize it without attribution and not send anyone your way. Even the "references" cited in AI answers often point to AI-slop blogs, not the original source. The articles we discuss on HN are often AI-written too. So yeah, it's about reaching the audience, but you're now competing with machines that produce an endless stream of human-like text, good enough for most consumers, practically for free.
I wondering about other thing - I'm assuming that publishers are now using LLM's en-masse to "proofread", do initial evaluation, do editorial work etc.
So if an author abstained from using LLM in the writing process - isn't then new, original, not yet on the market book ending up in the LLM training data corpus even before it hits the market?
It's a bad short story. Whether it was generated at the hands of humans or a computer isn't really relevant. Speaking in riddles is cheap now, so authors better learn how to surprise the reader in simpler, more readable words. It seems to be something that LLMs are quite bad at.
I liked the playfulness of:
Also mentions some interesting AI tells, for AI generated stories.It can be "playful" once or twice, especially if the text is playing with how nonsensical the metaphor is. By the twentieth time something is as tired as a willow tree on a Tuesday in May, as persnickety as an ant with a fever, or as rambunctious as a horny lobster, it's just nonsensical bad writing.
Metaphors are generally used to transfer the qualia of one experience into another. When the referent has no qualia, that is, you've never in your life experienced "river trouts tangled in the roots of a redwood tree", it's a failure of a metaphor.
https://nostalgebraist.tumblr.com/post/778041178124926976/hy...
And as much as there are still people on HN who insist that AI text can't be detected algorithmically, it's worth noting that the original story is marked 100% AI by Pangram. So it's not just this person seeing things.
I recently built and delivered an AI-driven novel-writing program. The architecture involved chaining the Claude, GPT, and Gemini APIs together so they could cross-critique and iteratively revise the text, while systematically saving key plot points and lore chapter-by-chapter. (Serialized 'web novels' published on a per-chapter basis are a massive industry here in Korea). AI has already heavily infiltrated the fiction space in Korea, and it looks like the exact same trend is hitting the US.
Personally, I find it incredibly easy to spot AI-generated text in Korean, but catching it in English is much harder for me. That being said, they still have very distinct, overused patterns. You constantly see words like 'ultimately' or 'structural,' and they rely heavily on highly formulaic 'X is Y' sentence structures.
The worst thing about this is not really that somebody might have had AI help writing a story, but that an editor thought they could get any kind of useful information from asking an AI whether the story was written by AI. If there's any hope of editors staying ahead of this phenomenon, they will need to educate themselves a lot better about how it actually works.