For about a week, I’ve been working on a piece that’s meant to elucidate the ways in which genAI has ruined what I called “my houses,” my writer-house and my “editor-house,” even though I do not use genAI. (Yes, the working title is indeed “A Plague on Both My Houses.”) The magazine I edit for, Diabolical Plots, does not accept genAI work, nor do we use it in our processes.

Well, at 4,500 words, I’m struggling to cut that piece down. And this past week, two genAI-related discourses broke out on my little corners of the internet that made me decide to write this instead.

pictures of a disc and a horse: discourse. Disc photo by Photo by Roberto Sorin on Unsplash and horse photo by Photo by Helena Lopes on Unsplash

Discourse the First: That deal with the HWA Newsletter

On Tuesday, the Horror Writer Association’s newsletter went out with a pro-AI opinion piece by a horror writer who just so happened to also sell her own AI-for-writers software.

She was rightly vilified—I saw posts to that effect on my primary social media, BlueSky, and my secondary social media, Instagram—and the HWA subsequently removed her from her position as HWA trustee. 

Discourse the Second: Dredging up a 4-year-old Google AI project

The second discourse was, as far as I know, on BlueSky only, wherein one writer posted about a four-year-old Google AI writing project in which 13 writers had participated. The first post led with “This just in,” and it was only several posts down in the thread that the original poster mentioned that this was four years ago. 2022, when it was not yet clear how much LLMs had been trained upon stolen work. Further, some of the former participants have spoken out that they did not know the models were used on stolen work, and that the project was originally presented to them as a means of offering their opinion on a product. Nevertheless, there were some pile-ons.[i]

The point of this post: My hot take—Hold it, first, a reminder that I’m anti-genAI

I am anti-genAI and do not use it in my writing at any stage: I don’t use AI “summaries” for my research; I sure as hell don’t use genAI/LLMs to draft, revise, or edit my work; and I don’t use genAI for any of the promo-related things I need to do pre- and post-publication. If I ever get to the point where I commission art, I’m going to work with artists who don’t use genAI—and even with such an attestation, I’m going to vet them beforehand to the best of my ability.

I believe genAI in the arts is a scourge that has stolen artists’ work and is significantly terrible for the planet. I believe that people who use it to generate their “art” are not artists or writers. A metaphor I saw on the interwebz that I quite agree with is that using genAI to write is like driving your car to complete a marathon, and then claiming you’re a runner.

They are not the same.

The point of this post for realsies this time: My hot take

I am a-okay with shaming genAI boosters and proponents, as with the former HWA trustee.

What I dislike is behavior such as the second discourse I mentioned. The project in question was old, had not, to my knowledge, resurfaced for any reason, came from a time before it was widely known the LLMs were/are built on stolen work, and ignored several of the writers’ statements against LLMs in the time since the project took place.

Relatedly, I dislike the various Reddit and other social media threads that get thrown around accusing people of genAI use when the accuser has no direct connection with the writer.

The people best positioned to do a thorough review of whether someone lied about AI use are going to be that person’s editor, publisher, and crit partners. People who work with the accused directly. Not a rando on Reddit who read a book that uses the rule of three, or em dashes, or the word “delve,” or any of the latest supposed “tells” of genAI use. So if a reader comes across a published work in the wild they think is genAI, my recommendation would be to contact that writer’s publisher and let them investigate. 

Aside from that, what I hugely resent about the focus on alleged AI users (again, alleged vs. those outright shouting that they use it, who I agree deserve scorn) is that these callouts take up space in which anti-AI folks could be boosting human writers and artists whose work does not incorporate LLMs. If someone chooses not to read an author who worked with Google AI four years ago, fine! That’s their choice.

But tell me who you do read. Make it easier for me to find work by human artists that you dig and love! Help drown out the slop with quality fiction, and shout about it. If you claim to support human artists, then, you know, support human artists.

And if you already post about human authors, wonderful! Please do that, like, ten or twenty times more than you do outing AI slop, because the way the algorithms and people’s brains work, the negative always takes focus.[ii]

Here are some folks I dig that I think are doing an excellent job of such work.

• John Wiswell is a wonderful writer and all-around awesome human. On both his BlueSky and Instagram profiles, he consistently boosts writers. In the unlikely event my readers here don’t know who he is, give him a follow!
• I don’t have a TikTok, but awesome writer and human L. Huang does! S. L.’s TikTok profile is full of some really fun TikToks plugging other writers! Follow follow follow!
• On her website, third awesome writer and human Emmie Christie has a feature called Writer Spotlight. For this calendar year, she’s been picking a writer of short fiction, reading through their some of their back catalogue, and writing lovely reviews. This is excellent because short fiction writers so often feel our work falls by the wayside. (Disclosure: I had the privilege of being one of her featured writers.)

Then there are the collectives:

• InkFoundry, a short fiction aggregator website. You can use it yourself to find short fiction on certain themes or tags. Super-cool means of discovering your next favorite writer.
• That SFF industry standard, Locus Magazine. Locus has been around for almost sixty years, now. I don’t know how many of the early issues have been digitized, but the site is a treasure trove of reviews for long fiction and short fiction alike.
• While I’m on the topic of industry standards, there’s also Reactor. Also been around a long time, also publishes reviews, and both regular and guest contributors frequently have “5 books about X” style lists.

As for me, yes totally read all the stellar short fiction we’ve published at Diabolical Plots! I personally tend to have difficulty avoiding thrive on longer posts, so I’ve begun doing some book recs in my newsletter. The latest issue is here and another set of recent recs is here.

Well, hey. This clocks in at about 1300 words, which is a big improvement over the 4500-word behemoth. I’ll call this good.

Writers, please please please go forth and write your own human work! Readers, share with us all your favorites. Help drown out the slop.

[i] I’m not providing links for this discourse, because the original BlueSky poster has deleted their thread. Further, in this case, I think posting links would be at odds with the point of my post here. 

[ii] I have experienced this with my own brain with the very well-meaning work of Bona Books and their report explaining how they mistakenly accepted AI work. I deeply appreciate their report and their suggestions for how to help editors avoid erroneously accepting AI slop; I’m thinking about how to incorporate those in my own editorial practices. And yet, to my extreme frustration, I know now more about the named AI grifters than I do the human writers who remain in that anthology.