NEW #opensaccess paper "Evidence of unreliable data and poor data provenance in clinical prediction model research and clinical practice" link.springer.com/article/10.1... #researchintegrity #dataprovenance #predictionmodels #predictiveAI

🚨 Protocol -> preregistration -> preprint -> publication 🚨 125 clinical prediction model articles developed on unreliable data on stroke and diabetes. Since preprinting our results 7 articles have been retracted. doi.org/10.1186/s129...

I like Messi but this time I want Egypt to win

I like my memes the way I like my science: open

"tell the truth" I was reflecting a lot today on this. On this era when you can read and hear bullshits everywhere

Society for Epidemiologic Research@societyforepi.bsky.social · last mo.

Thank you for sharing your journey and being such an inspiration to all of us. #SER2026

I want to be at SER meeting too 🥲

Indeed the primary causes of bad science come from within: academics are incentivized for quantity, not quality; for novelty, not reproducibility. Toss in poor #Statistics education (including measurement and experimental design) and you have a disaster.

Alejandro Montenegro@aemonten.bsky.social · 2mo ago

"The crisis of science stems not from paper mills or profit motives but from systemic incentives that reward volume, novelty and prestige over validity and reliability" The problem lies within. doi.org/10.1098/rsos...

It's not a nice feeling to realize that so many people who do research aren't actually interested in finding out the truth, but only want p<0.05

The physical-sciences repository arXiv is banning researchers from posting their manuscripts on the platform for one year if a submission is found to contain references that have been hallucinated by artificial-intelligence tools. go.nature.com/4nDfXvv

Researchers who use hallucinated references to face arXiv ban

The preprint server is the latest to impose stiff penalties on authors who contribute to AI ‘slop’ — but not everyone is convinced it’s the right approach.

go.nature.com

On my next birthday I'm gonna get myself a dinosaur encyclopedia!

"Anyone who looks on the world as if it was a game of chess deserves to lose."

"If you don't care about the methods, I don't care about the results"! Most observational research published in biomedical journals is very weak. That's why the methods section is the single most important section, and why it's so ridiculous that Nature journals relegate the methods to the end!

Katie Mack@astrokatie.com · 3mo ago

Nice reminder that because a study is published in a journal does not mean it’s a good study or that its stated conclusions are reasonable! When people say they are “doing their own research,” even if reading primary literature, missing the context & subtleties can lead to wildly wrong conclusions.

What 6,000 researchers think about the future of science. Noteworthy how this discusses well-known issues, without fully contextualising them or holding systemic harms accountable. Example: discussion of ECRs considering leaving, but not widespread redundancies www.nature.com/articles/d41...

What 6,000 researchers think about the future of science

Funding worries and publishing trends loom large in the minds of scientists surveyed by Nature Index.

nature.com

Where have people's sense of smell gone? Everytime I get on the bus I am hit by really strong perfumes. Even some hours later I could still smell it from my hair! 😵‍💫 In case you're not aware, some people can get headaches from these strong smelling perfumes.

SV: Peer review is not magic. Post-publication critiques are part of the 'Science is self-correcting' process. Love your critics - those are the ones who help you be stronger. Another lesson I learned: A crisis can look a lot like normal science. It's normal that things fail to replicate.

"...the big problem in science is not fraud but rather well-intentioned bad work." !!!

“I have seen the future of science. It is ruled by bitter competition instead of collaboration, pageantry instead of exploration. . . .Those willing to toil over genuine questions will necessarily lose out to those that can furnish cheap answers. . . .” statmodeling.stat.columbia.edu/2026/04/20/94/

It's kinda disappointing when people whose views you generally respect start using LLMs to write their contributions to collegial discussions, especially when they leave in clues like sycophancy from the bot ("this is your killer point"). Apparently they assume that we won't notice that it's slop. 😥

…students, by definition, cannot reliably determine if a re-wording of their writing changes the underlying ideas. Their ability to communicate their own understanding is exactly what we’re meant to assess. If they were already competent to tell, they wouldn’t need to be students. (4/7)

Question for AI haters who are also academics: How do you navigate co-authorship situations, where you don’t want to be involved in manuscripts where an author is using AI? Do you have explicit discussions with potential co-authors? How do these go?

-- New Open Research Game -- I reached the rank of Asst. Prof in #TenureRun! 🎓 I published 98 papers and paid $115.000 in APCs to publishers. My grant money is gone and I am exhausted. 💸 How far can you make it? Play now at forrt.org/TenureRun/

I do not want to use AI to write or think. Can someone train it to handle annoying tasks like formatting journal submissions instead of trying to get us all to use it for things that take critical thought and expert knowledge?

online.hbs.edu/blog/post/th...

Fundamental Attribution Error: What It Is & How to Avoid It

The fundamental attribution error plays a central role in how we understand the actions of others and how we justify our own.

online.hbs.edu

I'm glad to see accountability for academia's bad apples. But I also hope that resignations like this one won't be the end of the story. Because we still need to reckon with how the structures and cultures in "elite" academia have created ripe conditions for rot. 1/🧵 www.nytimes.com/2026/02/25/u...

Lawrence Summers Will Resign From Harvard After Epstein Revelations

nytimes.com

"Be teachable. You're not always right."

I don't think the "Einstein" or other AI replacing the important steps of learning, esp in higher ed, would be posing quite the challenge it is right now if students hadn't been sent the message their entire lives that the point of college is to get good grades and a high paying job, not to learn.

So inspirational. I always believe if you can have freedom to be yourself and do things on your own terms, you will have joy and be able to be at your best 🙂 that's also my experience so far www.theguardian.com/sport/2026/f...

Alysa Liu released the pressure, reclaimed her joy and turned it into Olympic gold | Bryan Armen Graham

After stepping away from figure skating, the US star climbed back on her own terms. Her journey culminated in a medal, but it was about much more than that

theguardian.com

Doing things the "right" way, burning out and then realizing there is another option, coming back and winning it all with no fucks given? Yes please tell me that story forever.

Frank Cifaldi @frankcifaldi.bsky.social · 6mo ago

I urge you to stop what you're doing and watch Alysa Liu's free skate from yesterday. I have never seen anything like this. I understand why sports writers are born now. I have never seen anyone as free from the burdens of this world as this pierced Oakland girl spinning to Donna Summers.

"Too many people – even those who were not involved directly in exploitation – took the deal because it was easy to do and because it felt good." www.theguardian.com/commentisfre...

Why are so many academics in the Epstein files? It’s not just about money | Christopher Marquis

In a university ecosystem that breeds hunger for status, Epstein made scholars feel like celebrities

theguardian.com

Really like this one! xkcd.com/1053/

Ten Thousand

xkcd.com

Dinosaurs And Non-Dinosaurs xkcd.com/3204/