This isn't going to be a summary of what's happened. That's been covered elsewhere with more thoroughness than I could ever accomplish on my own. If you haven't been following the latest fuel added to NaNoWriMo's dumpster fire, here are some videos, posts, and documents I recommend perusing:
TikTok - summary of events by EmmaSkies
Youtube - "criticizing AI is racism," says AI-backed writers group 😐 by D'Angelo
Article - National Novel Writing Month defended the use of AI. Now authors are stepping down from its board by Kevin Maimann
Youtube - The End of an Era: NaNoWriMo's AI mess by Meredith E Phillips
Reddit - In an official statement, NaNoWriMo calls critics of AI ableist and classist. by Janukanu (leader of roguewriters.net)
Document - nanoscandal.com - Megadoc detailing everything NaNo has done to lose its community
So, while I'm not going to rehash what's happened, I do want to talk about some of the surrounding issues and the fallout.
AI and Creativity
As I am writing this blog post, there is a floating blue star symbol just to my left--a friendly invitation to write this entire blog post using AI. Ironic, given the topic obviously, but let's pretend that I'm writing about something else.
Maybe I had the idea for a post: journaling supplies to take with you while you're camping. I could have AI write the entire post for me, just based on that scant sentence fragment. I don't need to have ever journaled while camping before in order to post about it. I don't need to ever have even gone near the woods in order to post opinions and recommendations about what tools you should take with you. In other words, I don't need to know what I'm talking about at all. And honestly, it would be so easy to just hit that little blue button and let it write for me. I'd be able to post so much more frequently, I'd be SEO optimized to hell and back. I'd have a "better" blog and look ma, no hands!
But I have to ask: If I can't be bothered to write the post, why should you be bothered to read it?
I don't have this blog with the sole intent of flooding the internet with my content. I write because I have things to say. And that, I think is the main difference between genAI supporters and the people who are pissed off at NaNo--people who support using generative AI in their writing are people who view their writing as a product. These are the people who flood e-books onto storefronts, never having actually read the book with their name on it. They're hoping to fool us into purchasing their course, or maybe they just want to claim they've published a book. Who knows. But let me be clear: they are not writing. They are putting their name on words they did not write.
And we haven't even touched on the ethical dilemma of where those AI words came from in the first place. At the time of writing, to my knowledge, there is no ethically sourced AI. The data sets--text that the AI uses as a model for what it spits out--all include content that was taken without consent from authors who actually wrote it.
Then, the AI is further trained on content produced by other AI like
As if it's somehow going to improve by teach itself. If you've ever played with google translate by running the same text through the machine over and over, you know what's going to happen with AI training itself.
Generative AI has no place in creative spaces. Full Stop.
Ethical Non-Generative AI
Yes, it exists! No, no one's trying to take away spell check or dictation!
The thing is, the term "AI" has gotten so buzzwordy, that's it's effectively useless. We've got people out here trying to say that if you're against AI, you can't use spell check. This is a red herring argument that takes advantage of the complexity of the issue at hand. We've had spell check for decades and it's been fine because we used to call it an algorithm, not AI. Now that everybody wants to call everything AI, suddenly it's confusing.
The term "AI" has now become synonymous with "algorithmic" and even "automatic." This makes it almost impossible to argue about generative AI on a good-faith basis. This conflation of issues is the crux of NaNo's original statement: If you're against genAI, you're against disabled people using dictation instead of typing, you're against alt-text readers for visually impaired people, you're against Representative Jennifer Wexton, who uses an AI model of her own voice due to a neurological disorder that affected her ability to speak.
No. No one is against these tools.
Kilby Blades, director of NaNo and presumptive author of the original AI post, is conflating these tools with generative AI in a purposeful or ignorant attempt to deflect criticism.
They are using unconsenting disabled people as human shields to block arguments against genAI.
And for what?
This is the part I really wanted to get down to. Why is NaNo taking this stance?
The very quick and easy answer is that they have an AI writing company, ProWriting Aid, as a sponsor. Of course, NaNo wants to soften the public's view on AI because they stand to keep their sponsor happy if NaNo's users purchase the software. Easy. Done.
But, it appears that even PWA came out against NaNo's insane statement. CNET's coverage of the issue says this:
The sponsor (ProWritingAid) didn't know about the statement before it was issued, a representative said. "ProWritingAid has supported NaNoWriMo for many years, and we were completely unaware they were going to make this statement," company founder and CEO Chris Banks told CNET. "We fundamentally disagree with the sentiment that criticism of AI tools in inherently ableist or classist. We believe that writers' concerns about the role of AI are valid and deserve thoughtful consideration. Writers are the reason ProWritingAid exists, and we are committed to supporting human creativity, not undermining it."
PWA did not pressure NaNoWriMo to make this statement. Kilby wrote it herself, unprompted.
Why?
The original post claimed that it was because comments on NaNo's PWA-related social media posts had become vitriolic. So was the "classist and ableist" comment purely spiteful deflection to make those nasty people on facebook feel bad about what they'd said? (Which, by the way, those facebook comments were not vitriolic, they were well-reasoned concerns about the PWA sponsorship.)
I have a hard time believing that. I have a hard time believing that Kilby, with all the claims on her resume about leading nonprofits through times of upheaval, cares so little about this organization that she can't be bothered to research the company that's sponsoring them. That she can't be bothered to research the type of AI that their sponsor uses and how that differs from accessibility tools. I have a hard time believing that Kilby would think that posting that little update on Zendesk was going to assuage users' concerns. Or that the catty tone she took in this other unprompted Zendesk post, was an appropriate way to communicate with the community. Or that she thinks posting there at all constitutes "transparency" when users need to actively seek out these posts. Try finding these "Notes to the Community" from the nanowrimo.org home page. I dare you.
I have a hard time believing all that. And yet...
Here we are.
Kilby is leading NaNo into the dirt.
It is simultaneously sad to watch and impressive that it is still going.
Comentários