R A T B O Y 🐀🐀 ♘ ♗ ♖

book cover reveal (publisher reveal pending)

@BarneysRubble0 @BDanielBlatt @JoyceCarolOates oh man really sweet burn big glasses. almost as good as four eyes! wow so funny

@BDanielBlatt @JoyceCarolOates It referred to the heinously large glasses she wore.

Did you know Cormac McCarthy stopped reading new novels after Infinite Jest and referred to @JoyceCarolOates as "the goggle lady?" My definitive takedown of Jest, MFAs, and the coinage of a new term "publishing-inudstrial complex" to explain it all :) unherd.com/2026/03/why-co…

@theauthorbrian @Lyons_Pen ok, you still don't understand how evidence works and seem to operate in a fantasy land of 100% certainty (seriously, spend time thinking about that). cheers!

@charmreduction @Lyons_Pen Accusations about AI without proof are causing real world consequences for authors who may have not actually used AI and they deserve to be protected. Anyway, have a good day!

I am very curious what the evidence the NYT sent is, as all I can find online is allegations based on bad writing. I don’t feel too bad for Ballard though, given that she straight up stole art off pinterest to use as her initial selfpub cover. My questions to Hachette would🧵

@theauthorbrian @Lyons_Pen you are never getting a world where there is regularly undeniable *proof*. we get the facts, and we make a decision. analysis can help, and it will get better. this "proof" idea is oversimplified.

@charmreduction @Lyons_Pen With written text, however, nothing on the market comes close to accurately detecting whether it was an AI generated or reasonable doubt that it wasn’t. So there is no proof; only suspicion.

@theauthorbrian @Lyons_Pen but really, the thing here is that we are not looking for certainty. how many court cases are solved with certainty? nowhere near 100%. we need some approximation of beyond a reasonable doubt. and we can already get pretty close

@theauthorbrian @Lyons_Pen it is more difficult, but unfortunately, you show a lack of understanding of chess. there is, in fact, not a perfect way to play chess, as in many positions there are often several more or less equal moves based on computer eval

@AustinWyckes how was it "baseless," lol. are you familiar with the backstory? her previous AI use? her theft of content?

I genuinely do hope she did use AI, and that she isn't being accused of it due to baseless accusations. Many authors/writers still use em dashes, semi colons, and other things that are being flagged as AI. I have been using em dashes since elementary school.

@theauthorbrian @Lyons_Pen it is not a "feeling" we are talking about analyzing LLM here. you really don't seem to get that it is not either "feeling" or "certain proof" and that there is a lot of room in between. like i said, in the same way i (or chess dot com) use analysis to root out cheaters for years

@charmreduction @Lyons_Pen But you can’t quantify it, so it is only a feeling, and that turns into a slippery slope fast.

@Lyons_Pen @theauthorbrian it is not perfect, but it is silly to entirely disregard as marketing. everybody has an interest when quoted in the media, lol. is every comment marketing?

@theauthorbrian @charmreduction 💯 this is marketing, not evidence.

@Lyons_Pen @theauthorbrian there is no certainty, but it is like chess. we can detect cheaters who are using an engine to play chess with a vary high rate of success. also, if i am playing a cheater, i can almost always tell. with language, it is similar.

@charmreduction @theauthorbrian I get called a witch hunter with some frequency, and I have a public list of AI authors, but determining AI writing with certainty is just beyond our ability right now. The thing is, the standard of writing for a tradpub company should be high enough that AI writing wouldn’t🧵

@HughBlanton5 "As a literary critic I expect my opinions to often be overruled by the public's reading tastes and by awards juries." so self-important lol. ohhhhhhh the critic doesn't have mainstream taste

And then there's those books that would have been better if they'd been written by AI: athinsliceofanxiety.com/2026/02/review…

@Ekaylasimpson it was not good enough to pass. there were early reviews about the original self-published version flagging it for AI. this was a lack of due diligence (or worse) on the publisher's side

Here’s the uncomfortable truth: this Ai book was good enough to pass. It was a successful product. If this upsets you (it should), why? Because the ends don’t justify the means. It isn’t *just* about the end product, it’s about what the process of creating creates *in us*🧵

@thefancymike dennis cooper blog zac smith blog ... you probably know those

What are the websites you recommend that do good reviews of indie books? Who are the people you recommend that do good reviews of indie books? Throw it all at me!

@FifthCrichton @LukeEpplin greg you sound like a massive loser

@LukeEpplin Nice to know I can write you off as a "writer" who doesn't care about the quality of his "work"

It never fails: you publish a book and within weeks the emails come in: "Hey, I liked your book and all, but on page 257, there's a typo that I want to call your attention to...." I don't understand this impulse to contact authors in this way.

@ElonsTopJeet typical humourless soylionath follower response. good one.

@charmreduction @Anarseldain Cool, unfortunately you’re Canadian and therefore less than human

The foid cries out as she strikes you.

@roo_richardson it mostly seems to be genre writers, who often suck at writing, who are worried about this lol

re: the AI book being pulled at Hachette - though I agree we shouldn’t let AI ‘books’ be published, how are they making sure to be absolutely 100% certain it was indeed written with AI? Because if there isn’t a definitive system in place this could become very messy for authors

@birchlse they dropped the ball on that one, tho. many months ago there was a thread on reddit about the book. and before it was even purchased by this particular publisher there was lots of talk. they did zero due diligence

Reading into this "AI horror novel" scandal, I fear people in the publishing world may be massively, massively overestimating their ability to detect AI prose by eyeballing.

@theauthorbrian @Lyons_Pen what, exactly, would "evidence" look like to you? nothing short of an outright admission? a reasonable case is made, in the longer article, for why this is likely bad, AI-generated writing.

@charmreduction @Lyons_Pen So those detector programs aren’t reliable enough to be considered evidence.

@theauthorbrian @Lyons_Pen for example:

@theauthorbrian @Lyons_Pen expanded reporting, same nyt reporter: nytimes.com/2026/03/19/boo…

@Anarseldain lol are you like 86 years old? no, i know you're broke, 23, and, from the sound of your voice on twitter spaces, overweight. but you sure tweet like somebody's geriatric grandparents!

If every club and all the people inside them instantly ceased to exist, the world would be massively improved.