That's why I'm so against ChatGPT. I'm against letting a machine do your homework for you. I think you should cling to Doing Your Homework Yourself like it's the last opportunity you have to liberate yourself from mental enslavement. Because it basically is.
Anyway shout out to Kai Cenat.
I think a lot of us millennials & older really take it for granted that we were the last of only a couple generations where there was an all-encompassing campaign to encourage literacy. Everyone used to agree upon it. There were commercials with your favorite celebrities just being like, "READ!"
I really like Devezer's comments. Unfortunately couldn't keep myself from reading the original substack post:
Going to "These are solvable engineering and institutional design problems"
from "How do we verify AI-generated claims at scale?"
is just... I honestly don't have a word for that.
2. Of course we should hold AI to the highest standards. That doesn't imply double standards because we can treat AI as technology without anthropomorphizing it and we primarily conceptualize science as a human endeavor. The two can be subjected to different standards. It's normal, not at all weird.
Really super legible 🤯
(I assume it can also be used in less optimal ways...)
Exactly this. Burning a generation for some weird company bets.
I'm constantly trying (/failing) to get this point across.
If you're a trained expert in a field, then it may be worthwhile to question the scientific consensus of your peers.
If you're not, the scientific consensus is absolutely the best you can do and it's arbitrary foolishness to disregard it.
These seminars are always worth it, and what's fantastic on top: when you have to suddenly run halfway through a talk the recordings are freely available!
Longevity influencers tell millions to take rapamycin, sit in hyperbaric oxygen chambers, blast themselves with red light therapy, and take 50+ supplements daily.
Meanwhile something that cuts dementia risk by 20% is invisible.
overmatter.substack.com/p/your-anti-...
Saving this for teaching...
New YouTube video posted
“The Joy of Coupling”
Happy Valentine’s Day!
youtu.be/rpW2ZYtXn6o?...
"Model to Meaning" just received a super generous review in JASA.
Check it out! (I'm blushing 😊)
Reminder: You can get the paper copy from CRC, and the full **free** version will remain online forever at marginaleffects.com
www.tandfonline.com/doi/full/10....
Oh wow thx heaps!
*didn't
Did manage to find it, could you tell me the title? Or is it just in that volume?
Neyman is my hero
Screenshot one of three describe a new pair of R and Python packages demonstrating easy object interchange between R and Python.
Screenshot two of three describe a new pair of R and Python packages demonstrating easy object interchange between R and Python.
Screenshot three of three describe a new pair of R and Python packages demonstrating easy object interchange between R and Python.
chronometre: A new package (pair) demo for R and Python
Efficient yet simple object exchange between R and Python
dirk.eddelbuettel.com/blog/2026/02...
#rstats #rcpp #python #datascience
Google with AI buttons
Google without AI buttons
I made a filterlist for uBlock Origin to remove Generative AI features on websites. Includes blocks for
* Google AI Summaries
* YouTube Ask button & chat summaries
* GitHub Copilot
* Facebook AI chat
* X's Grok buttons
* Deviantart DreamUp
* Booru AI images
* And more
github.com/Stevoisiak/S...
It lives! 🎉
dx.doi.org/10.1002/sim....
Quite some time ago I mentioned the arxiv version of this paper comparing imputation, raking, and TMLE in a reasonably large set of simulations, by authors including moderately strong partisans of the various approaches.
Screenshot from Rotten Tomatoes showing 12 actual reviews panning the Melania movie while '99% score' on Popcornmeter smelling of astroturfing campaign
As Rotten Tomatoes might be getting astroturfed, a quick quiz:
- Amazon paid $40 million to make it
- Amazon paying $35 million to promote it
- Amazon own IMDB
What is going to be the opening score at IMDB?
Everytime some asshole starts taking about how "EU regulation stifles innovation", this is the kind of stuff they are hoping to enable.
Then prove it!! Where is the evidence!? Where are the evaluations!?
Fantastic talk, a must see!
Info, schedule, and recordings: cls-data.github.io/CIIG/
Join mailing list: www.jiscmail.ac.uk/cgi-bin/wa-j...
If you are interested in all things ✨causal inference✨, please join our multidisciplinary Causal Inference Interest Group (CIIG). We host monthly seminars featuring speakers with various academic backgrounds and research interests.
Links below.
cc @clscohorts.bsky.social @pwgtennant.bsky.social
Woah terrible, glad to be on Fedora here.... Also good description on how to think of a quick local fix.
I think that the mass enthusiasm we see around gen AI is manufactured, and I think this because it doesn't exist for renewable energy, despite it being more measurably impactful (and also technologically wonderous)
In which I look at simulations to see how well a Normal distribution fits something, but with complications
notstatschat.rbind.io/2026/01/05/s...
Can send you the high-res version! It's 1907, Author is G. Freytag, it shows voting in the Austrian-Hungarian monarchy.
This smells distinctly like collider bias and/or selection bias and/or regression to the mean... You simply can't select teen prodigies, and world class athletes rom databases, and go run regressions without serious consideration of the selection process!
