Pre-2022 Books
notes.lorenzogravina.com
notes.lorenzogravina.com
All I know is that when I see a new self-published book on Amazon, I go and check the author's page to see how many books he published and at what intervals. If I see 10 books published in the same year on various subjects I avoid it like the plague. Life is too short to waste my time reading slop.
For example, there are multiple evidently AI-generated titles that come up on the front page if you search for "Rust programming", "cybersecurity book", etc. I guess I can't rule out that "Winston Knowles" is a real person, but I'm not gonna bet money on that: https://www.amazon.com/Cybersecurity-Career-Manual-Interview...
Yet, from what I can see, AI writing is mostly used by people who don't know a thing about writing, and because they have bad taste, they do not see what's wrong with AI writing and put it out there.
At the end, you write for a purpose: for marketing copy, etc., you would require a different type of writing talent than something like writing a fiction book. But AI doesn't understand this nuance; it has only a default type of communication, which is highly optimized for being a chatbot. It is possible to write a very good text using AI if you have taste and you know what you're doing, but most people don't.
Similarly, a lot of vibe-coded apps are garbage, but because the people creating them lack software domain knowledge and don't even know what they don't know, they think it's good and put it out.
We have a massive problem here that's not just limited to writing - the promise of AI for the mainstream market is that you can replace domain-specific knowledge and have world-class execution in any vertical with just AI, but that's very overhyped imo and doesn't stop the people who don't have domain experience to try out stuff with AI and not realize what they made is a steaming pile of shit in reality.
If you can't take the time to write it, why should I take the time to read it?
I'm thinking about just leaving the online community entirely.
I highly doubt there is any mainstream search engine that doesn't use any LLM like technology in the backend to provide search results.
> I checked it with a couple of AI detection tools and they confidently said that 60% of the article was written by AI.
These tools are as reliable as the polygraph. I have seen passages from books published before 1900 getting a 90% likelihood of being AI generated from these tools. At the end of the day there are only so many ways to write a grammatically correct sentence and if an AI can write it, chances are some human somewhere already did.
It really put a sour taste in my mouth. At that volume (2 books per week) you’d think Amazon could identify and label that it’s AI generated.
Of course there are no doubt people out there realising that a fair few of us do this, and are starting to edit posts to pre-date them as a sort of SEO trick…
I think near term this feels like something bad for individuals because we can no longer find truthy information. But longer term this is going to be an issue for the same special interests, as people normalize to this and longer view the internet or media as a place they can get reliable information.
I really do not think children born today are going to think “let me just look up reviews for a product online” or even “let me just ask AI to summarize reviews” in 20-some years when they have purchasing power, because it will be ubiquitously known that nothing written online can be trusted, and AI is just ouroborosing itself (and a few more decades of increasingly desperate and short-sighted advertising/over-productization by the controlling corporations).
Have you ever met someone who could say all and do the right things but never made you feel anything, or your gut was sensing an ulterior motive? It's a magic trick we are all bewitched by at some point in our lives. I suppose I filter by published year because I dont want think about if I am being tricked or not.
[1] There are some very talented writers[A] out there who (I assume) cannot do the world building part.
[A] Recent Favorite: https://archiveofourown.org/works/1134255/chapters/2292768
I've spent very little time on ao3, but I imagine it's slightly better than RR where seemingly a mark of value is "huge length (word count and chapters)" with weekly updates.
It was profoundly alienating to see hundreds of people liking and commenting on chapters of obviously AI-generated prose full of LLM-isms and broken metaphors and similes - and these 'authors' have the gall to link a patreon account and ask for money for this tripe!
(The phone-vs-computer element also can't be discounted - it's much, much harder to do in-depth long form discussion on a smartphone. Input is a hassle, and doing a bunch of research is extremely annoying when you can't open multiple windows and an editor.)
After 2023 or so, a new problem popped up: posters would read a chapter, hallucinate a series of events that never happened, get mad about what they thought they had read, and then attack the author! Arguments born from a complete lack of reading comprehension were breaking out in all the busiest story threads, with waves of reports every time a new chapter was published. It would not be an exaggeration to say that the average IQ of the board had dropped 20 points.
There was also the monetization aspect to consider; fanfiction used to be by fans, for fans - the work of amateurs, in every sense of the word (amātōrem, 'lover'). While that certainly didn't mean everything was /good/, it at least meant it was a work born of genuine passion. The new authors treated it like a business, churning out updates every week and advertising to readers that they could read several chapters ahead on their patreon. Without even touching the ethics of the matter, you can imagine how the incentives changed the output.
I tend to buy books from second hand book shops and eBay now and usually older or well used copies. A good sign of their authenticity.
We should probably work on developing standards for what we want in a book instead of clinging to a losing position.
On the other hand, it means the age of self-publishing was short lived and is effectively dead now.
This is... satire, right? right?
Something about the industrialization of thinking makes it uninteresting. similar to frozen pasta from the shelves, heated and served.
In addition to the obvious negative effect on learning curves, I'd argue this dearth of "long-form documentation" has a negative impact on programming language and app development, as well as software usability in general (isn't CONTROL+COMMAND+SHIFT+3 for a MAC screen capture intuitive?)..
AI has become the new "teacher of record" to fill this documentation void, but it's very individualized and narrowly focused. There's no longer a shared mental model, where everyone has read the same books and are working somewhat off the same page. Ironically, it's probably never been easier to write a good tech book, but there's likely zero money in it and the search and AI giants are likely to hover it up and start serving snippets of it up for free before you see a dime, making people very unlikely to buy your book, even if you've written a technical masterpiece!
Each person that put in the effort is one additional review point.
Thinking, Writing, Editing, Proofing, Publishing, Distributing, Selling...
At every step, a human had the opportunity to say "no, this is bad". And the fact that they didn't is a vote of quality and reputation.
These days presumably every step can be automated without a human in the loop at all, and it's up to the buyer to discover "no, this is bad", but at that point their money is already spent.
Romantic image. Doesn’t hold. Traditional publishing has pushed out lots of poorly edited slop for decades; it didn’t start in 2022.
> Crafting Interpreters, by Robert Nystrom
> Re[Coding] America, by Jennifer Pahlka
> Systems Performance, by Brendan Gregg
You will still have gatekeepers and taste makers. Publishing houses will screen fiction for well-written and interesting fiction. Word-of-mouth, personal recommendations, and endorsements from people you respect will continue to outweigh algorithms, if you care.
For cheap reads, how much of a difference is there between James Patterson’s 734th beach read thriller and what an LLM with a 50m token context window can produce? Does it matter that it’s not written by six ghostwriters? Probably not to the median Hudson News buyer.
For non-fiction, it’s easier to gather research and related materials. If you were cherry-picking facts to make a narrative, yeah, that’s easier, but it’s not like we haven’t gotten really good at that anyway. Again, there will be cooling off periods for scholarship to be debated and coälesce.
What will get better is people asking questions and getting well-researched pieces on a specific niche or confluence of topics. AI is just-good-enough-to-be-dangerous now. It will get better. We’ll learn to harness it (literally) to iteratively fact check and cite sources. We will build repositories with heavily sourced facts for it to build upon. It will be pulling together “truths” that can be traced, then incrementally adding inference across those, which can then be verified and are a new fact.
I read a lot. I love, love, love new and original authorship. I deeply value writing as a craft. There will be a lot of garbage. More than there is now, at an incredible rate.
And we’ll figure it out.
Growing critical thought, in my experience, has always been the much harder problem. Not sure we’re in for a good time on that front.
Why would they start now?
Pre-2022, when someone posts a Show HN, even if it's not something you would normally be interested in, there's a baseline understanding that _someone_ cared enough to spend time and effort to build it. So in a hypothetical future scenario if you do find yourself looking for that particular tool, there was value in you seeing that Show HN so you can revisit it.
Now, I just ignore all Show HN posts.
The liberal concept that the everyman should have their own original thoughts that others should consider is a historically a very new concept. And we start getting things that look a lot like C-c C-v quickly after the Renaissance.
See humans have the tendency to romanticize the past, and if this is allowed to compound they elevate really quite dismal people to the realm of literal godhood in some cases. If you asked someone a thousand years ago what they though life was like thousands of years in the past and what it will be like thousands of years in the future most would have said the past was better in all regards including health, strength, morals even technology; while the future would be viewed as the continual circling of the drain. Put yourself in their shoes, you go look at a Roman Colosseum, you can't build that, nobody you know can build that. If you asked Vitruvius during the construction of the Aqueducts he would tell you that he's maintain the knowledge of his ancestors, whom could have build such structures if they needed them or had the manpower, and the technical problems are just a trifle. If you pushed him, he might invoke Providentia and that if the gods stopped blessing you we'd fall even faster.
This kind of discovered then lost fits better narrative within the human psyche better than the unintuative truth is a constructed social conversation, that can be semi-formal and rigorous (the scientific method) or lax (common sense) depending on the setting.
See humans have the tendency to romanticize the past, and if this is allowed to compound they elevate really quite dismal people to the realm of literal godhood in some cases. If you asked someone a thousand years ago what they though life was like thousands of years in the past and what it will be like thousands of years in the future most would have said the past was better in all regards including health, strength, morals even technology; while the future would be viewed as the continual circling of the drain.
I do have one freely downloadble Sci-Fi book from 2022 which took 10 years in total to write so it might safely pass all your standards, but if a fixed cut off date is your criteria for a good book then there really is no hope for writers like me continuing.
If it's any consolation, I'm releasing a technical book on Java this year. In 2026. When AI exists and, at according to the front page most days, programming is "solved." All of it was written by hand because, well, why else would I bother? What's the point of a craft when you're not the one doing it? Half the joy is getting better at it, and that requires doing.
It's demoralizing to know post-2022 books are devalued in people's minds. It's demoralizing that I also think that way. I spend enough time at work reading other people's copy/pasted LLM output. Books should be a relief from that world, but the risk is unavoidable unless you go back in time.
But hopefully good things rise above the noise. If they don't, it's still OK. Honing a craft is its own reward, even if not celebrated. (this is the lie I tell myself as I lay in bed awake realizing I've wasted that last several years of my life)
If you can’t trust a contemporary writer to not use genAI, then find another interest with true zero trust creative verification (like improv) because copying and cheating on writing has existed for decades.
However, my unpopular opinion is extremist in another way: societies should probably not tolerate AI in creative fields like creative writing at all, if Sora can be shut down, so can other useless and toxic parts of LLMs.
For a sanity check, I did run the first page through an online AI checker and it confirmed what I already knew: 100% human.
There are mobile game ads on TV here. My father asked me what actually the players get from paying the game companies money. He still doesn't get it after I tried to explain how it works twice.
Given the quality of 2026's entertainment, looks like they had a point. And likely they had one at 2006 and 1986 and 1966 too.
But the corollary to it is the fact 10% of everything isn't crap, and just as there always has been crap, there always has been excellence. Just less of it.
But different times absolutely can have worse production.
I'm not seeing the same from the translated fiction works I've picked up in the same time period, thankfully.
It has to be absolutely demoralizing to make something on your own, and have it immediately labelled ai by someone who can barely spell it.
Low-background Steel: content without AI contamination (2025)
I never let an LLM write or rewrite a post, or even a paragraph, for me. I want to write it myself and I want it to be in MY voice. I think I'm a pretty good writer and I like my writing. However, I suspect those who may be less confident in their writing use an LLM to "check" their rough draft but then succumb to the temptation of just pasting the LLM's output because it "sounds better", it's already finished and... writing is hard. This is always a mistake and no one should do it in a forum like HN. It's rude and we'd much rather hear your words and ideas as you express them.
The sad part is this ends up in an all or nothing between "Never use an LLM when writing a post" and "Have LLMs write posts for you."
In my experience, most formal business writing fell into two types: 1. Pro forma box-checking that probably didn't need to be written at all (or at least read by me, as a senior leader), and 2. Actually important information or ideas. The majority was Type 1, which I tried to avoid, and if it was Type 2, it was either well-organized and intelligible, or such a mess I'd have my admin schedule a meeting so the author could explain it and answer questions.
But I recently retired so I no longer have to suffer through formal business writing anymore :-)
(as an aside, I also struggle with super long sentences and found a pretty good trick for curing it: rewrite by hand with pen and paper. My cramping hand finds entire passages that the work can survive without)
There will be _more_ shit books now, but that's the only difference.
There will be probably a constant rate of "good" books.
How many shit books can or will you wade through to find a good one? Particularly when some percentage of the shit ones are "good enough" that you won't necessarily know it was shit until after you've read some or all of it? What do you do when the ratio of shit to good is 100:1 or 10,000:1? In the past, you could find a trusted publisher or reviewer without too much trouble. Soon, if not already, the publisher and the reviewers will have the same dross ratio as the books do.
Has anyone? Now I'm curious if it's just my particular bubble.
Kidding aside, I would be surprised if something larger than using it as a thesaurus/corrector is slipping by. Literature is genuinely hard.
Eventually artists will figure out how to use AI to make real art that is actually good, just like photographers did with photography, and that will be its own new thing. I don’t see much of that yet but with photography it took a while.
A few years ago, I finally decided to read Romeo & Juliet. I was disappointed to discover that it was full of cliches and tropes.
I just read things and decide if they're good or not. I did that ten years ago and I do it now and nothing has changed. Most things are bad, a few are good. That also has not changed.
The issue with content >2022 isn't that it's ai-generated per se, it's that it's still slop.
Perhaps we’ll be flooded with slop to the point where nobody can distinguish reality from fiction anymore, because it takes too long to check and reason about it with our outdated tools and thinking processes. Then perhaps people will lose faith in the “truth” and simply accept that one-sided ideology and biased belief is all that exists and matters, and corrupt leaders (at all levels) will make use of that to sway masses of useful idiots.
Wait…
In this dystopian sci-fi novel, E.M. Forster depicts modern life for humans: detached from each other, living in solitary silos above the barren Earth below, which has become an infertile wasteland of itself, and only ever connecting from their rooms by projecting “holograms” of themselves to interact remotely with one another. Spoiler alert: the machine experiences a catastrophic failure event.