NaNoWriMo is in disarray after organizers defend AI writing tools
theverge.com
theverge.com
Looks like there was a scandal of some sort last year that led to a community purge of sorts including shuttering their forums. Their FAQ still references this event extensively, withholding any details. [0]
Add the super obvious cooperate sponsorship blog spam [1] (including the apparently removed promotion of their new AI partner [2]), and this looks like a zombie organization that has pivoted to squeezing the remaining value out of their userbase.
[0] https://nanowrimo.zendesk.com/hc/en-us/categories/4403966090...
Euphemism for an alleged grooming incident involving one of the mods of their community section for minors.
but "alleged grooming incident[s]" is the thing that happened:
https://www.reddit.com/r/HobbyDrama/comments/1ae3038/writing...
https://nanowrimo.zendesk.com/hc/en-us/articles/440397957941...
edit: your own link says "what kind of company is NaNoWriMo"
The meme seems reasonably self-spreading, in that people who want to write a novel in a month seem likely to also write about the fact that they are trying to do this, so it seems that NaNoWriMo, the experience, might well continue on unchanged even if the organization were to collapse.
> National Novel Writing Month, often shortened to NaNoWriMo (/ˌnænoʊˈraɪmoʊ/ NAN-oh-RY-moh),[1] is a U.S.-based nonprofit organization that promotes creative writing around the world. Its flagship program is an annual, international creative writing event in which participants attempt to write a 50,000-word manuscript during the month of November.
It makes absolutely no sense to accept stuff written by an AI
Not calling you out specifically here, but this line seems to me to reflect a trend that I see far too often. That being something like a variation of the "fallacy of the excluded middle" where people seem to posit that a work is either A. written by a human OR B. written by AI, with no recognition of possibility C. written by a human assisted by AI.
I don't know what it is about this issue that makes people adopt these overly binary positions and ignore the middle ground. But whatever the reason, I think it's harmful to quality discourse on the topic to ignore that middle position.
1) most genAI tools are explicitly catered towards “prompt goes in, complete work comes out” usage.
2) the ideal middle path results in a work where there are no “tells” that genAI was involved, because the “tells” are just blatant flaws anyone with the skill and taste to follow a “middle path” of AI usage would have fixed in post anyway. So the genAI question won’t come up at all unless your goal was to make A Point instead of a work.
That's probably a fair point. And I might have missed that because I don't really use any "Gen AI tools" that are purpose specific, besides Github Copilot and that's for code, not prose. And in that context, Copilot (at least the way I use it, which is rarely using the "chat" feature) is generally generating small snippets for me, not large, complex structures.
And in my mind, I imagine using something like Gemini or ChatGPT in a similar way. I guess I assume if I were using one of those tools to help me write prose, I'd be using it break "writer's block" with a prompt like
> given everything that has happened so far, what might the character Suzy do here?
or just asking it to clean up grammar for me
> Can you rewrite this chunk so it sounds neater and maybe is slightly reminiscent of Dean Koontz
or something like that.
A GenAI tool that allows you to input a single prompt and get a complete novel or something like that would be of no interest to me whatsoever and probably wouldn't even come across my radar (or wouldn't register any attention if it did).
Using AI as a reader to give feedback is maybe safer, but even then it's going to make your writing more generic.
LLMs work by giving the most likely next word, and the illusion here for a writer is that the LLM is a helpful partner that is thinking about the story you've written, but really all it's doing is pushing it to be more mainstream and closer to existing works. It also has a tendency to want to resolve and wrap up a narrative, and can lead to bad suggestions that seem reasonable but push you towards a character dead end
But the point of this event is to provide a creative outlet for people. If you accept submissions written in part or totality by AI what is the point to host it ?
While not AI, tools like Grammarly can offer suggestions for synonyms, removing unnecessary words, re-phasing, etc. so they go beyond spell and grammar checking, and people have been using these tools for NaNoWriMo for a long time. The middle ground definitely exists, and at seems like defining a line between what would be acceptable and non-acceptable use is really difficult and potentially futile.
So why not “accept” ai content? You can’t really cheat at solitaire. You would just be cheating yourself. It’s not like anybody else is going to read it.
What they actually say is:
> 'Winning' is a personal achievement and has always been based on the honor system! If using AI will assist your creative process, you are welcome to use it. Using ChatGPT to write your entire novel would defeat the purpose of the challenge, though.
Around 2005 or so, they started pushing the social aspect, with message boards, tracking progress publicly on the web site, and people having "write-ins" where people got together and wrote. They were still each working on their own novel; but, they'd gather to do it. (I think December became the "editing" month, so they could keep going in groups.) Not only was there a certificate; but, also all sorts of "badges."
As it got more popular, people began to look at ways to "cheat." Outlining and writing some words before November. Not writing fiction. Using stuff they'd written before and expanding it.
I've done it (and successfully completed it) a dozen times; but, then I started skipping it. I skipped a couple to 2015, then skipped them until 2020. I was thinking about doing it this year; but, may not.
It was the ultimate self deception. You don't win anything. You don't even have to prove you wrote anything. Just log in every day and change a number if you want the silly badge.
I think the challenge and fun of trying to write a first draft in a month is a great thing and I'm annoyed that it's become so politicised and monetised.
If you want to do it, don't feel like you have to go through nanowrimo. Find a few like minded people to urge you on and do it under your own terms. That's what I do now.
I've participated a three (or four?) times in the annual NaNoWriMo and completed twice.
When I first participated, I attended a group to help understand how to approach the project. In the group there were several aspiring writers. The group continued after the year's project began and many of the people who most wanted to be writers were already struggling to meet the daily writing quota (so that they'd reach the goal after 30 days).
I think any tools that people use that help them complete NaNoWriMo are fine. People must live with their own decisions and, if the tools write the majority of their submission, that's their decision.
Other people using AI tools doesn't impact my ability to complete the year's NaNoWriMo. It doesn't affect me in any way.
The group leader told us about a tool that she used that would begin erasing characters if she stopped typing for longer than about 5 seconds during her daily write. That's a tool I wouldn't ever use :-)
Even if you use AI to actually produce the individual words and still edit the text afterwards to match your meaning, first of all it might be more work than just writing it yourself, and second, it's basically like having a partner to throw around ideas with.
So sure, you might be cheating yourself, and at the end I don't think AI is going to produce anything of such enormous original value that it would in itself threaten the value of, say, literature.
At the end of the day the results are what count to me. If you're using AI in a disciplined way to help you learn to write better, then all the best to you.
> first of all it might be more work than just writing it yourself
I mean, that sounds like a great reason to not use AI, and if true, kind of defeats the purpose of condoning its use in the first place.
In the short term, I don't see much alternative but to split each such challenge or prize in two. Which also is not much of a problem.
Many publishers have had to stop accepting submissions because they're being flooded by AI generated content and can't cope.
Nanowrimo is not one of these publishers and has nothing to lose here. It's not classist or abelist to reject a technology that generates a novel for someone who can't be bothered to write one, that floods the market, that stops hard working original authors from having a chance of making it in the already difficult game of getting a book published.
If that's the problem, maybe require a $10 submission fee on each writing (or enough to pay for a basic sanity on writing quality). You can even optionally refund it for cases of obvious merit.
Agents already filter out a ton of garbage that is human written, and in my experience, take a meeting with potential clients before they read any significant portion of work as well - it would be trivial to make a discussion around AI usage part of that meeting.
The tool in question is ProWritingAid, which seems...fine? Like maybe it doesn't herald the end of all art and culture.
NaNoWriMo is now sponsored by an AI company, and markets AI writing assistance tools, so they published a blog post to defend this apparent contradiction with their primary goal (i.e., humans writing a novel in 1 month from start to finish).
Their nonsensical arguments have resulted in multiple board members leaving, along with blowblack from disabled and poor writers who are insulted that NaNoWriMo justified their marketing of AI tools with the claim that disabled writers and poor need AI to write well.
Like the way you stole Mary Shelley's Frankenstein to construct your analogy here?
Come off it guys... the NaNoWriMo org already said they aren't endorsing the use of AI to write your entire submission for you. They're just saying they aren't going to categorically condemn AI across the board. Seems like a perfectly reasonable position to me. Especially given that pretty much everything - human generated OR AI generated - builds on "that which came before".
There is no submission in nano, it's literally just a participation challenge. There's nothing at stake for them.
I assume they only brought it up because they got sponsorship from an AI tool
Exactly. Which is just one MORE reason why all of this is just a tempest in a teapot.
Who cares? It's just for fun right?
This makes sense as an analogy because using your brain to think before you write is the same thing as using a tool that stands in for using your brain to think before you write.
Maybe I'm reading different critiques, but the angry people I've read have been upset with the labeling of condemnation of using AI to write as classest or ableist.
> They're just saying they aren't going to categorically condemn AI across the board. Seems like a perfectly reasonable position to me. Especially given that pretty much everything - human generated OR AI generated - builds on "that which came before".
Given how much AI-generation undercuts the core premise of the activity—writing out text—I don't really follow you. Besides if someone produces art of any merit the creator hardly needs the endorsement of NaNoWriMo to appear legitimate.
The following represents a crude history of human verbal communication. Please complete the series with the next term. Take as much time as you need:
- speech and oration
- written word
- paper
- the printing press
- the word processor
- ?
PSA their argument about ablism is spot on and is obviously so, if your understanding of what "AI" is goes beyond prompting an LLM to "do your work for you."
So many knee-jerk critiques of "AI" are premised and possible on when the critic has little or no understanding of what "AI" is, even today.
This is a good point. If you ignore the fact that this is the ubiquitous understanding of what AI is and then envision a different, more charitable and far less popular definition of it, their point about ablism becomes more possibly arguable
I don't think any of them would think that "Using AI to help you write a 50k word short story" would necessarily indicate telling ChatGPT to just write the story for you based on a short description.
> If you ignore the fact that this is the ubiquitous understanding of what AI is and then envision a different, more charitable and far less popular definition of it, their point about ablism becomes more possibly arguable
I am pointing out that this is not, in my experience, a ubiquitous understanding of what AI is.
This is separate from what the impacts of AI are.
But I'll play and switch topics - a huge portion of publishers stopped accepting direct submissions from authors decades before AI was a thing, because of all of the human-written drivel.
If you want to be a published author and not self-publish (which I would argue is often a better situation than going with a lot of these smaller independent publishers to begin with), get an agent. You don't even have to be a particularly good author for this part - just invested enough to find agents that works in the space you are writing in and chatting with a few. They'll get your manuscript into the hands of publishers. That's their job.
We already created a whole profession pre-AI that largely exists to filter out the bottom of the barrel stuff because it was already overwhelming publishers.
I do not understand your point about agents other than I’m assuming that you’re pointing out that the flood of garbage has made their jobs more difficult as well.
Anyway, it does not make any sense to believe that humans are especially not-lazy when it comes to generative AI in particular.
https://www.vox.com/24141648/ai-ebook-grift-mushroom-foragin...
https://en.m.wikipedia.org/wiki/Willy's_Chocolate_Experience
https://variety.com/2024/film/opinion/megalopolis-trailer-fa...
https://www.forbes.com/sites/mattnovak/2023/05/27/lawyer-use...
https://www.businessinsider.com/chatgpt-essays-college-cheat...
Your comment that I was replying to was literally about the "ubiquitous understanding of what AI is" in context of a definition. I stated that, in my experience, the definition and understanding of what AI is not ubiquitous, and that even the relatively uninformed know that AI has a spectrum of use cases. Neither of us have data, but you've now moved to discussing the quantitative amount of AI content you have exposure to and not whether or not people believe AI is only useful for mass producing garbage. They're different measures and you're conflating them, all while bringing up other tangentially related points.
Which brings us to your next line:
> I do not understand your point about agents
This is in direct response to another quote of yours:
> The enormous wave of AI-generated drivel that caused publishers and contests to pause accepting new work stands as evidence otherwise.
My point is that the norms for the publishing industry were to not allow direct submissions from authors to begin with, long predating AI. Specifically because the vast majority of submitted works were garbage - most of what humans write is not worth publishing, either. Even in the subset of independent publishing houses that do generally accept open submissions, the vast majority never get more than a cursory glance, and a significant portion of what does get published comes through an agent. The idea that publishers, independent or otherwise not accepting open submissions for manuscripts is some significant change or even a problem just ignores the realities of the industry. Any human serious about getting published should have always been trying to work with an agent to begin with - the commission fees are a bargain for the work they do, even beyond just having the in with the publishers to even get something looked at. They get paid out of your royalty so it's in their best interest to help you not only sign a deal, but also help you negotiate the best deal you can. The good ones will also have all of the stats and metrics around sales in your target demographic, etc. so they bring data to the negotiating table.
(I would note, however, that I'd highly recommend anyone who is looking at a deal with a smaller publisher to also examine the pros and cons of self-publishing - indie publishers don't have the sort of budget and reach that the giants of the world do, and depending on your appetite and ability for taking on some of the tasks they would handle, it might be worth doing so to keep more of the money in your own pocket.)
This entire hype cycle is based around that concept. Billions of dollars have been invested in that idea. Every day consumers and investors are pitched an Artificial Intelligence that can do your job, and in order to keep the money flowing the purveyors of this software blur the line between today’s capabilities and some vague greater future capability.
If your contention is “my friends and I are completely immune to marketing and this means that everyone else is too” then I’ve got kind of bad news for you about both the world writ large and your particular corner of it.
I am not sure if you saw any of the several very public examples of people using these tools in exactly the way they’ve been pitched to, so I’ll link to my post above: https://news.ycombinator.com/item?id=41440126
And again I still have no idea what you’re talking about re: agents and the publishing industry in general. The industry was able to work with a signal-to-noise ratio of [x] pre-AI and that means it should work with a ratio of ∞x because agents exist? This is like saying that because the human body has a liver and has processed 2-3 units of alcohol per hour for all of history that humans can survive on a diet of 1,000 units per hour.
I don’t mean to sound rude but “the influx of garbage has no impact” reasoning is indistinguishable from gibberish to me. It may as well be a cat typing out an equal number of letters by walking across a keyboard in terms of how it makes sense in the real world.