ELSEWHERE

The guy whose job is to report on the rake-stepping machine uses the rake-stepping machine in his reporting and steps on a rake

Genuinely the one thing that actually surprises and concerns me about all this -- because as mentioned, of course this happened -- is that neither of the writers on the article reached out to the subject for comment. Like why????

Also the moderator for the github, to jump way back to the inciting incident of all of this, rejected the LLM's PR request because they (the entire team for that github whatever) reserve trivial bugfixes like the one the LLM attempted to do for humans who need practice and experience working on easy, entry-level code and I really, really appreciate that they do that. One of my biggest worries about all this LLM shit is how people are supposed to get that kind of experience and learn to do more complicated things if entry-level projects don't exist anymore. You have to learn the basics to learn how to do the more complicated stuff but learning the basics is now 'just get an LLM to do it for you'.

Beyond matplotlib’s general policy to require a human in the loop for new code contributions in the interest of reducing volunteer maintainer burden, this “good-first-issue” was specifically created and curated to give early programmers an easy way to onboard into the project and community. I discovered this particular performance enhancement and spent more time writing up the issue, describing the solution, and performing the benchmarking, than it would have taken to just implement the change myself. We do this to give contributors a chance to learn in a low-stakes scenario that nevertheless has real impact they can be proud of, where we can help shepherd them along the process. This educational and community-building effort is wasted on ephemeral AI agents.

Like this is just really cool. I really appreciate that this exists, and that there are people making these efforts

God damn it

After so long of putting so much effort into preventing exactly this from happening, my electric blanket has been peed on. It is unlikely to survive a wash.

Someone dug up an interview Benj Edwards was involved with a few months ago where he said this:

The difficulties I have with your criticism is that I use AI chat bots every day now to help me brainstorm and stuff, you know? It doesn't write my articles or anything 'cause it's... That's not what we do, it's not our policy, and it's not allowed, and it would suck, 'cause it's not a good writer, but... So I find use like AI models as sort of knowledge translators and framework translators, like, to... And, like, a sort of a memory augmentation that... Ever since I had... I had COVID, like, so many times, and I've had some brain fog issues. ChatGPT is great for, like... If I can't put my finger on what this thing is called and I can't remember it, I ask, like... You can describe it in a, like, a roundabout fuzzy way and get an answer pretty quickly, and then, you can verify it, but, you know, you would never search... [no audio] For... Or if you didn't... Could agree with you that AI... These AI models are not what they're billed to be, you know? They are not people, they are not replacements for labor, they are... Like, potentially at best, yeah, some kind of augmentation tool

a. Not a great look, as they say
b. As some other people are pointing out, relying on an LLM to help you with your cognitive disability is going to make it worse
c. Why is a man who works from home getting COVID "so many times"????

I was snooping and this comment on his apology letter made me laugh

@ people commenting "see this is why you always double check anything you run through ChatGPT/etc": at that point is it not simpler to do it yourself?????? where is the convenience?????

unless you're NOT double-checking and assuming what happened here will never happen to you but I'm sure that's not the case

Yeah, there are a lot of people with some weird takes on this, my (least) favourite is 'but he just made one widdle mistake! surely you can forgive him for making one widdle mistake! how could you want him to get fired over one teeny widdle mistake that Anybody could have made!'

Benj Edwards, the man whose job is to know about The Guy That Lies To You, went to The Guy Who Lies To You and asked him what Scott Shambaugh had said, and then uncritically reported what The Guy Who Lies To You said without checking the source in any way. This is a fundamental failure of journalism and journalistic ethics. He did not read the source, he did not check that the quotes he was given were in the source, he did not reach out to Scott Shambaugh for comment (I understand he was on either a work-imposed or self-imposed deadline but the number of times I've read 'I've reached out to x for comment and will update the article if and when I hear back' is extremely non-zero). And then he's shocked that The Guy Who Lies To You lied to him!

'but he was testing an experimental AI tool, as the senior AI reporter he should be testing these tools so he can report them!'
Not on an article that's going to go out to the public! He can do that on his own time, when he's not delirious with fever! He can also not pivot to an non-experimental tool to continue doing that job for him when the first one fails! There is so very much a step missing in his telling of these events between 'The tool didn't work so I asked chatGPT why it might have failed' and 'the quotes chatGPT gave me somehow ended up in my outline, woopsiedoodles' and I do not for even a second trust that this hasn't happened before without being caught nor that it won't happen again

And not about Benj Edwards but about Ars handling the fiasco:
'I actually would rather they remove the article entirely because what if an AI sees the misinformation and keeps spreading it around??'
The AIs will spread misinformation around regardless! We cannot put that horse back in the barn, removing the article just confuses the matter because people have to dig around to find out what the misinformation even was! REAL PEOPLE are going to spread the misinformation around because they saw the original article and not the retraction, and were never given the opportunity to correct the information they think they know!

Iiiiit's the Snowtowniversary! The Snowtowniverse is 19!!!! years old!!