SArpnt

come on people

LLMs ("large language models") need a lot of input to generate their statistical text models. Orders of magnitude more than is contained in a single webcomic, even one as large as Achewood. The way this works is, roughly, you train a "base model" on approximately the entire internet. Then you take this "base model" and you add additional training on top to give it a specific flavor, such as "Ray Smuckles". The article alludes to this:

The first challenge was formatting the data in a way the language model could use. Then, there was the matter of picking an underlying language model to train with Ray’s voice. OpenAI’s ChatGPT was a little too sanctimonious for Ray, who likes to color outside of the lines, Hall says. They wound up using a fine-tuned version of OpenAI’s Davinci, which Hall estimates is about 60 times more expensive than ChatGPT.

So, this is not just a matter of "he's only using his own writing so it's fine". The model Onstad is working with is exactly as plagiaristic as anything else OpenAI has put out, it just adds a layer of Smucklesiness on top of that. Whether you think "training a statistical model on the entire internet without authors' consent" is specifically plagiarism, otherwise exploitative, or totally fine is up to you. But you can't draw a clean line and say "training AI nonconsensually is bad but what Onstad is doing is okay."

training a model on the entire internet and using a model already trained on the entire internet are two different things, especially when the model is already available to everyone

also a good factor is however the person using ai involved handles copyright on what they create with ai (directly or indirectly), especially however they give permission to others to use it for ai stuff

i don't really like the way copyright law works currently, but i generally don't dislike as much when regular people use it in the ways i don't like because they're already influenced by how the corporations use it, same with capitalism and ai and other things. if someone uses one of the plagarism bots for a shitpost that's fine by me because it doesn't make sense to allow ONLY evil people to use it for evil things, and i only have a problem with ai in the first place because corporations are using it to get around their own copyright laws and to smush and scramble people in other ways

id probably put an indepth take on what onstad is doing but im too lazy and i don't actually know him or achewood well (ill read the comic someday)

i don't think there is any overlap between the kind of people that would know f1 is supposed to mean help and the kind of people that would benefit from it in any situation ever

the kind of stuff people say praising rust feels very wrong, smug, and annoying and i don't want to agree with them but i've genuinely never seen an actual situation where something like c, c++, c#, go, or whatever other general purpose compiled language would do better than rust unless it's something about compatibility

any time i see anyone that doesn't like rust they either never actually explain why (maybe giving vague hints about the borrow checker or whatever) or demonstrate they just don't know how to write rust code (as in they tried and failed to do a bad thing, and the way they would do it in their other language isn't good either, probably for the same reason)

please tell me why rust bad go ahead i want good reasons

(also if you mention anything about the rust foundation or trademarks or whatever i don't care about them i'm talking about the actual language and tools)

you can easily store gender as a boolean

false: nonbinary
true: binary

there's two values and one is a thing while the other isn't that

EDIT: im probably going to get kneaded and flattened for this post