tl,dr; Scientists no longer allowed to use federal funding to publish, attend meetings, or talk to the public. They cannot collaborate internationally. Grants can be cancelled for any reason, at any time, political appointees have a final say over what gets funded, and who gets funded.

The APSA Pres Task Force on AI, Politics, & Political Science's report comes in the form of an edited volume identifying questions & establishing a foundation for the empirical study of how politics & governance are affected by AI. Check out these early chapter drafts: shorturl.at/cMZzI #polisky

Wieder mal ein sehr schlecht gemachtes Papier in Theory and Society, oder wie Christian Lindner sagen würde "Wo ist die Nachricht".

This is bad. Bad for researchers. Bad for science. Bad for universities. Bad for professional associations. For example: 1. Every award would be vetted by political appointees, who could overrule peer reviewers and deny the award if it doesn't advance the president's priorities. 1/🧵

This is a great study and (one of?) the first of its kind! A note of caution from a human contributor though: given the slow academic publishing process, the findings should primarily be interpreted as a snapshot of AI capabilities (and the capabilities of researchers using AI) in 2024. (1/4)

The methods section is the most important section of a paper. Without knowing the methods, the results are literally meaningless. So it's not OK to relegate the methods to the end. Because this suggests they are an optional afterthought rather than the essential core of the entire paper.

Wichtigste Änderung neben dem Nichts, was letztlich drinsteht: Die Übernahme der nicht genutzten Zeiten in die Postdoc-Phase entfällt. Wer also unter sechs Jahre Befristung vor Abschluss der Promotion hatte, bekommt sie danach nicht mehr. #IchBinHanna #WissZeitVG

I am very happy that we are expanding Registered Reports to all disciplines in which we publish, which is all areas of natural, clinical and social sciences, engineering & public health. We’ve learned a lot since 2023 when we first invited RRs; we are ready to expand 🧪 www.nature.com/articles/d41...

Fake journals using real professors’ names on AI-generated papers www.nbcnews.com/tech/interne... Behind a paywall, but the title speaks for itself. Q is: what's the purpose? Do the journals try to build a good reputation and lure real researchers into submitting and paying APCs?

If a scientist uses AI to make strong claims in public about the scientific literature, and some references do not exist, and the generated text does not correcfly reflect the findings, should we inform them privately to fix the text or should we share this information publicly?

👋 Meet the Sociological Science Accelerator (SSA), a network for collaborative experimental sociology. Members propose research ideas 💡, the community votes 🗳️to select projects, and winning studies run across multiple labs/online using shared protocols. experimentalsociology.github.io

Is it just me, or are Zenodo DOI links (for GitHub README markdown files) broken? I just get a broken link symbol, even for repos that used to work. If I use the image link separately, it renders fine in the browser. Example repo: github.com/pgmj/easyRas...

New blog post! Let's say an effect you are interested in varies between countries and you want to explain some of that cross-cultural variation. What could possibly go wrong? www.the100.ci/2026/05/20/f...

What if we treated #OpenScience like medicine and demanded evidence before recommending it? That's the question we put to @floriannaudet.bsky.social, co-founder of @grios-openscience.bsky.social, in our latest interview. 🔗 www.grios.org/prevention-c... #Metascience #SystematicReviews #EvidenceBased

🚨NEW-ish VIGNETTE ALERT🚨 The Bayes factor vignette for {bayestestR} has been completely re-written! We've got all the BFs! - Marginal likelihoods - Model & posterior averaging - Order restrictions - Savage-Dickey density ratios AND MORE! Read it up, here: @easystats.github.io #Bayes #rstats

One of the greatest tricks in coding is, at least for me: When you are stuck, walk away from the desk and take a break. Letting things sink in often helps in finding a solution.

"Peer review will always be a compromise between trust and doubt, efficiency and exploration, order and disruption. The danger lies not in that compromise, but in forgetting that it exists." #AcWri #AcademicSky #PhilSci #STS #HistSci 🧪

#MetaScience How useful are open data when we do not have open results? (At least APA does not claim to be a scientific organization) replicationindex.com/2026/05/16/o...

Published today & free to download by chapters or whole text, my @LSEPress book with Tim Monteath, Denisa Kostovicova and Hannah Boroudjou *Doing Open Social Science* press.lse.ac.uk/books/e/10.3... helps qualitative & quantitative researchers make their work more reproducible & robust

Das sollte nicht nur einmal gesagt werden. Das sollte alle Alarmglocken schrillen lassen. Journalismus in AT steht mit dem Rücken zum Abgrund. Kein seriöses journalistisches Medium profitiert von einem geschwächten ORF. Und unsere Demokratie jedenfalls nicht.

ArXiv to Ban Researchers for a Year if They Submit AI Slop www.404media.co/new-arxiv-ru... Special times call for special measures. "If a submission contains incontrovertible evidence that the authors did not check the results of LLM generation, this means we can't trust anything in the paper.”

I may miss the background here, but this reads like funding startups and companies. (and it reads a bit like megalomania, "reshape entire technology sectors"). Existing institutions are too slow and whatever, new organizations are needed with a focus on product development.

Academia relies on good faith. It cannot be done otherwise. This is precisely why plagiarists, people who fake data, people who deliberately ignore counter-evidence, and fascists need to be driven out. Full cordon sanitaire against this crap.

We are excited to announce the launch of the brand new Open Research Bulletin! Aimed at anyone with an interest in open research, the ORB showcases expert comment, case studies, and practical tips for new researchers and experienced practitioners alike. www.journals.sussex.ac.uk/index.php/op...

Early in my PhD, a colleague finishing his (highly rated) dissertation told me he was stuck on the introduction. After years of hard work, he could no longer find an argument for why any of it actually mattered. This stuck with me ever since. A🧵 about this paper: www.nature.com/articles/s41...

I adopted this workflow and created a simplified version using R, quarto, and only Claude (high effort sonnet 4.6). I kept the prompts with minimal changes, but executed them one after another, and checked the artifacts created (codes, reports, etc.).

It really is somewhere in between hilarious and deeply (if not at all surprisingly) embarrassing that book publishers still haven’t figured out that Bluesky is The Premiere Social Media Platform for People Who Actually Buy Lots of Books.

Terrible example of the power of rankings and their misuse

Interesting. I heard people say agents don't need flashy websites, for example for online shopping, but a genuine agentic web. Same could be said for agentic science that doesn't need papers. Not that I am endorsing it right now, and neither is @hermwerf.bsky.social, I guess bsky.app/profile/herm...

Scientists should be very careful with the term "reproducibility crisis". It is a (highly effective) tool to sow doubts by those who want to undermine science. There is no crisis; non-reproducible results are not ideal, they have always been part of the scientific process. Don't feed the beast.

Science Has Too Many Pet Theories, Not Enough Falsification Science rewards positive findings, but how do we decide what not to keep pursuing? “Fail faster” research models may make science more reliable. #ReplicationCrisis #Falsification #OpenScience #ScientificMethod #MetaScience

oh I'm skeptical too, but more bc I think journals are sticky institutions hard to reform. That said I'd be willing to register the hypothesis that reviews/editors who see (substantive) quant work that does not bring credible human content might reject more than they used to. Mixed methods go up.

What can we learn from automating an entire quantitative social science paper, from prompt to finished product? Thread about ongoing work with @natewilmers.bsky.social 1/12 Paper: osf.io/preprints/so...

Reads a bit like an anti-Coase theorem.

75 Jahre DVPW – 75 Jahre #Politikwissenschaft in Deutschland Heute Abend wird gefeiert! In Lüneburg mit und an der @leuphana.bsky.social diskutieren ehemalige und amtierende Vorstände zu Stand und Perspektiven unseres Fachs. #DVPW75 #75JahreDVPW #PoWi

The Draft Volume of the APSA AI Task Force, cochaired by Josh Tucker and me, is now available. ARTIFICIAL INTELLIGENCE, POLITICS, AND POLITICAL SCIENCE will be published by Cambridge Press later this year, but the draft chapters are available today: apsanet.org/wp-content/u...

On the (Mis)Use of "On the Mis(Use) of..." in article titles (2x in same issue): - On the (Mis)Use of Machine Learning With Panel Data onlinelibrary.wiley.com/doi/full/10.... - On the (Mis) Use of the Fixed Effects Estimator onlinelibrary.wiley.com/doi/10.1111/...

Came across EarlyReview earlyreview.ai You "receive detailed, constructive feedback covering key elements of your research design" using AI. Of course, PAPs and RRs should be scrutinized before submission, but this service may not be the way: 1) There is no name or institution listed. #MetaScience

Kleiner Fehler in der Überschrift: Rate der Kaiserschnitte hat sich von 0,15 in 1991 auf 0,33 in 2025 erhöht. 1991 war Geburtenzahl aber höher als 2025 (830019, de.statista.com/statistik/da...). Absolut gesehen ist die Zahl der Kaiserschnitte in 2025 daher um ca. 73% höher, nicht mehr als 100%.

ReplicationResearch.org is now open for submissions! Submit replications and reproductions from many different fields, as well as conceptual contributions. With diamond OA, open and citable peer review reports, and reproducibility checks, we push the boundaries of open and fair publishing.

Ab 2027 startet 𝘗𝘰𝘭𝘭𝘶𝘹 𝘗𝘶𝘣𝘭𝘪𝘴𝘩𝘪𝘯𝘨 als nicht-kommerzielle Plattform für politikwissenschaftliche Zeitschriften im Diamond Open Access. Als ersten Partner konnten wir Zeitschrift "Totalitarismus und Demokratie" vom @hait-tud.bsky.social gewinnen. Wir freuen uns sehr auf die Zusammenarbeit! 🤗 #DOA

Faculty Concerned About ASU’s New AI "Frankensteinian" Course Builder www.insidehighered.com/news/tech-in... Based on the report, this seems to be legally okay. Ethically, it is not okay to prepare and launch this tool without the instructors' consent. I am not sure what ASU's goal is. 1/

Excited for my talk today at Cornell University on reproducibility in the soc & behavior sciences, results from SCORE w/ a focus on sociology. In case you can't make it, I made a pre-recorded video youtu.be/kMwOwZaYYtE @briannosek.bsky.social @weedenkim.bsky.social #openscience #reproducibility

ASR has no requirement for providing data, code, or other research materials to editors, reviewers, or readers of the journal… Why spend an hour getting real data, running the model, sharing the data and code — when ChatGPT already “knows” the answer? familyinequality.wordpress.com/2026/03/07/t...

Selecting an effect size for power analysis is hard. Many researchers fall back on Cohen's thresholds, but they have no empirical basis and vary wildly by field. Our new paper offers a better option: field-specific effect size distributions built from meta-analytic data doi.org/10.3758/s134...

"The replication crisis, if there is one, cannot be established by the methods used to declare it." yesyes 🤓

A simple solution would be to require grant applicants to do a short, but comprehensive, musical number summarizing their research aims and goals.

How much does it cost to buy your way onto a scientific paper? To answer this question (and better understand the global market for academic reputation manipulation), we compiled BuyTheBy, a massive dataset of thousands of advertisements from paper mills in seven different countries. Link below!

🤔🤔 "Now that AI can generate plausible papers at low cost, the distinctive value of human scientists lies increasingly in judgement: identifying important questions and critically evaluating results. This can be assessed by looking at a scientist’s performance as a peer reviewer"

Re: the Qualitative Inquiry paper on rejecting genAI in qual research. Overall, I agree with the critical stance but I think the authors need flesh out more about what they mean by "using genAI" specifically. here's the paper 1/x journals.sagepub.com/doi/10.1177/...