I should add that I have a book published with Springer. So, yeah, my work is being directly devalued here. Fun fun fun.
AI slop in Springer books:
Our library has access to a book published by Springer, Advanced Nanovaccines for Cancer Immunotherapy: Harnessing Nanotechnology for Anti-Cancer Immunity. Credited to Nanasaheb Thorat, it sells for $160 in hardcover: https://link.springer.com/book/10.1007/978-3-031-86185-7
From page 25: "It is important to note that as an AI language model, I can provide a general perspective, but you should consult with medical professionals for personalized advice..."
None of this book can be considered trustworthy.
https://mastodon.social/@JMarkOckerbloom/114217609254949527
Originally noted here: https://hci.social/@peterpur/114216631051719911
And a new language feature, generating a list by lack-of-comprehension
Sloppenheimer
alt text
Drunk woman yelling into man's ear (meme image). Captioned as though she is speaking:
Their foundational text is a Harry Potter fanfic that supposedly teaches science
but it gets 9th-grade biology wrong by fucking up Punnett squares
A Bluesky post by Jamelle Bouie prompted me to reflect on how I resent that my knowledge of toxic nerd deep lore is now socially relevant.
alt text
Breaking Bad meme. Jesse: They always say "Read the Sequences", right?
Walter White:
Jesse: But the Sequences are all cult shit, like everything Yud says about quantum mechanics
Jesse: It's all "The scientists are insufficiently Rational(TM) to see the truth, don't trust the scientists, trust me instead"
Walter White: Jesse, what the fuck are you talking about
A lesswrong declares,
social scientists are typically just stupider than physical scientists (economists excepted).
As a physicist, I would prefer not receiving praise of this sort.
The post to which that is a comment also says a lot of silly things, but the comment is particularly great.
Are we actually going with vibe coding as the name for this behavior? Surely we could introduce an alternative that is more disparaging and more dramatic, like bong-rip coding or shart coding.
The phrase "trying to gatekeep what was once their moat" makes me feel like a character in A Scanner Darkly who has reached the "aphids, aphids everywhere" stage of Substance D abuse
So a wannabe DOGEr at Brown Univ from the conservative student paper took the univ org chart and ran it through an AI aglo to determine which jobs were "BS" in his estimation and then emailed those employees/admins asking them what tasks they do and to justify their jobs.
I write science for my job and fiction for fun. The mental processes are not that different between a murder mystery and a theoretical physics paper. In both cases, you've got a tangle of pieces floating in abstract space, be they preconditions for a theorem or clues to whodunit, and you have to instantiate them somehow, picking a linear order of text to lock down the loose assemblage. You're trying to cast a shadow of this damn strange thing made of parts that stick together or split apart depending on how you turn the whole.
What happened was that I had a handful of articles that I couldn't find an "official" home for because they were heavy on the kind of pedagogical writing that journals don't like. Then an acqusitions editor at Springer e-mailed me to ask if I'd do a monograph for them about my research area. (I think they have a big list of who won grants for what and just ask everybody.) I suggested turning my existing articles into textbook chapters, and they agreed. The book is revised versions of the items I already had put on the arXiv, plus some new material I wrote because it was lockdown season and I had nothing else to do. Springer was, I think, the most likely publisher for a niche monograph like that. One of the smaller university presses might also have gone for it.