I’ll stop when they do

C'mon, @mdpiopenaccess.bsky.social - after all the effort sleuths put in to hunt down the papermill products, is this the best you can do? A "correction"? If you want to be taken seriously as a publisher you need to weed out the gobbledegook. pubpeer.com/publications...

ArXiv will quarantine you for a year if you submit AI slop, and somehow people are freaking out about having to check that their citations aren’t made up. Now call me old fashioned, but isn‘t reading what you cite and meticulously copyediting your whole manuscript part of the job description?

My benign take is that (1) you’re not responsible for checking all of your coauthor’s citations but also (2) the new arXiv policy is totally reasonable because there should be consequences for fake citations and there’s no practical way to adjudicate multi-author cases.

I go back to this passage of the the Ezra Klein interview with @tressiemcphd.bsky.social about the alignment between message, register, audience, and medium and, I’ve gotta say, folks. Some papers out there could have been tweets.

SHOCKING stories of data fabrication Tim Errington shares a never-before-told story of data fabrication in cancer biology. Scientists go to extreme measures, even engaging in data fabrication, to secure a publication in prestigious journals. #science #truth #medicine #stem #shorts

AI psychosis is just “I really think the stripper likes me” for guys who don’t leave their house

Partiet vars partiledare har 93% frånvaro fuskar med kvittningssystemet för att vinna en omröstning mitt i valåret med flera viktigare kommande omröstningar. Var det här som heter ”att kissa i byxan för att hålla sig varm?”

Martin Melin, vilken shitshow. Du sitter där när människovärdet rasar och nickar. Kan du inte ryta till när du i en timme lyssnar på en rasistisk flodvåg om somalier, “bidragsblattar” och “parasiter”, klassisk BSS-retorik? Vad är det som hindrar dig från att protestera när hat sägs i klartext? 1/3

🤷🏻‍♂️ Systematisk israelisk förstörelse av livsvillkor också i Libanon nu 🤷🏻‍♂️

good news! hungary has replaced its authoritarian racist leader with a liberal racist

AI slop has reached big scientific publishers: irisvanrooijcogsci.com/2025/08/12/a... As if it was not enough the amount of additional crap that is polluting internet searches lately… The post contains many interesting resources and links if you are interested in this topic #academicSky #eduSky

The fundamental asymmetry in scientific publishing today is that people can submit absolutely mind boggling crap, and face zero real costs. Unless these kinds of things are tied to substantial costs/risks, and while we rely on volunteer labor, nothing will get better.

Although AI improves performance during assisted sessions, performance drops sharply once it's removed. And relative to controls, AI-assisted participants also gave up more frequently on test problems.

Why does this happen? Two candidate mechanisms: 1) AI resets your reference point for how long things "should" take. Unaided work suddenly feels harder than it actually is, -- a kind of hedonic adaptation. 2) AI removes the productive struggle through which you learn what you're capable of.

People still hating on the No Kings stuff are telling on themselves. They don't *want* a popular resistance movement. They want to be little rare political flowers who get to complain about the brunch moms but oops actually the brunch moms gave up on brunch and want blood, now

In any sane world CPAC would be treated as an organised attempt to overthrow democracy and replace it with a fascist government.

Congrats to all those journal editors & peer reviewers who legitimised Lynn’s “national IQ” database by letting it into their journals. Truly, it’s a victory for academic freedom to publish papers which claim that children living in a refugee camp constitute a nationally representative sample 👏👏

Wikipedia now has higher standards than all universities

Didn’t think that they meant alt-right rants when the new editors said that they were bringing science back to “Theory and Society.” Between this and the LLM garbage article someone circulated the other day, I’d say I’m probably placing that journal along the likes of MDPI…

Scrolling through this now and it's SO good. Beyond first-year college students, this would be really good for just having conversations with like... your parents. Your colleagues. Really clearly written, nicely designed, breaks for discussion questions. Sharing this widely!

Von Papen, von Hindenburg och von Schleicher annonserar sin strategi för att tygla nazisterna (1932, färgläggd från originalet i svartvit).

If you, as a social scientist, believe that AI is now a better social scientist than you, then (a) you’re probably right, and (b) sure sounds like a skill issue, y’know?

borrowing @anthonymoser.com's template. it actually doesn't matter whether "AI" is "useful".

Funny how tech buzzwords make their way into unexpected corners of our existence. Before AI there was blockchain and before that there were apps and before that everything was .com.

In sociology this would be that one generation where we went “we are educating sociologists, not statisticians, so let’s just teach them to press the right buttons on SPSS and it will be fine.”

For the ABM and networks enthusiasts: Special Issue of Social Networks on "Agent-based Modelling for Social Network Research" - co-ed by Andreas Flache, @squazzoni.bsky.social, Károly Takács and me www.sciencedirect.com/special-issu... Deadlines Extended abstracts: 1 April Full papers: 15 November

“Within two hours of subscribing to NatSocToday for the purposes of this investigation, the Substack algorithm directed the Guardian’s account to 21 other profiles featuring similar content.” substack continues to be a cesspool that should be abandoned just like x

Up next: some Mark Regnerus type writes a textbook based solely on Spencer’s social evolutionism and late Werner Sombart “blut und boden” geisteswissenschaft. I bet that one will clear censorship.

We don't sample from a distribution. Meanings are not random, and they are not individually determined. Nothing means what I say it means. Meanings are culturally determined. They are arbitrary because they are constructed, but not simply by individuals.

Also because co-occurrence and embeddings are trying to measure meanings in all their complexity and (pun unintended) embeddedness within physical and social contexts. Saying that embeddings are meaning is akin to looking at a termometer and saying “that is heat/cold.”

Vi alla kommer ihåg vad sossarna gjorde efter 2015, oavsett hur förvånade och förskräckta de låtsas vara nu när konsekvenserna börjar dyka upp. Det var S som bäddade för Tidö.

Men jag läser inlägg i sociala medier där socialdemokrater nu förvånas och förskräcks över att den strama migrationspolitiken är stram 😱

Decent people resist Gestapo.

There we go. That there’s a headline.

I’m sure that somebody is already making this argument, and I’d love to read about how the current turn to fascism is partly a response to the current crisis of social reproduction. Any pointers?

I'm teaching Statistical Rethinking again starting Jan 2026. This time with live lectures, divided into Beginner and Experienced sections. Will be a lot more work for me, but I hope much better for students. I will record lectures & all will be found at this link: github.com/rmcelreath/s...

outra distinção importante é quem é de esquerda no mundo inteiro, e quem é só da fronteira pra dentro

If he carries on like this it's going to make the FIFA Peace Prize look ridiculous

Doctorow's enshittification closing 2025 with a strong play for Word of the Year futurism.com/artificial-i...

'The fall of a prolific science journal [Science of the Total Environment] exposes the billion-dollar profits of scientific publishing' english.elpais.com/science-tech... '(It) has been expelled from the indexing system for irregularities. Its publisher, Elsevier, has a 38% profit margin in 2024'

There is no ethical use of AI because the continued existence of the AI industry is directly harmful to (alongside everyone else) the interests of everyone whose labour makes books exist—writers of fiction & nonfiction, illustrators, translators, researchers, editors, everyone.

Professional societies keep beclowning themselves buying into a lie about what an LLM "summary" is. They are inherently counterfeit: not an epistemic product of the ideas in the source, but summary-shaped text linguistically based on *other* works (in the training corpus) that use related language.

“Kin Propinquity, Residential Mobility & Segregation”: @benjarvis.bsky.social, @kchihaya.bsky.social & @eduardotapia.bsky.social examine ancestry & segregation; they find ancestry sorting effects are 3X greater than kin propinquity effects. @iasliu.bsky.social read.dukeupress.edu/demography/a...