Joanna Tai

Cartoonist Jon Kudelka has passed away, his wife Margaret has confirmed.
www.facebook.com/margaret.kud...

Vale.

Vale, Jon Kudelka.

And as always, Fuck Cancer.

Cover image of the Journal of AI in Education from 1989.

When most people think of AI in education they probably don't think of scientific journals old enough to exist online as black and white photocopies. Here's a cover from the first volume of the Journal of AI in Education 1989/90. It contains a really signficant paper...

Never mind the jobs you had, tell me five classes you took at university (I had to make this up since it was an integrated curriculum! These were themes...)

🎓 Law & ethics in medicine
🏃Health promotion & knowledge management
🔍 Sociology of medicine
📈 Evidence based medicine
❤️‍🩹 Occupational medicine

It's only Thursday (of a short week) but already feels like Friday ... also, how is it the end of January???

I have also done a lot of writing and commenting this month but still seem to have a stack of papers to be worked on and submitted 😲

Guess who hates university collegiality? Apparently collegiality is inconsistent with 'modern people management'. So it fucking should be. The university is not there to serve its own HR department.

open.substack.com/pub/hannahfo...

Are They Hot, or Is It the ‘Australia Effect’?

Since it’s Australia Day you might like this piece about how Australia makes young people hot.

“Are They Hot, or Is It the ‘Australia Effect’?” www.nytimes.com/2026/01/20/s... (gift link)

The thing about the bloke who lost stuff bc ChatGPT shows the extent to which people still haven't understood that TECHNOLOGY IS NOT BUILT FOR YOU ANY MORE. It is purely, entirely self-serving, with zero obligation to, or care for, its users. Everything is on you; you cannot rely on software.

NeurIPS reviewers typically review 6+ papers, each with 100 or so citations. The reviewers can't possibly check that every citation is real. But why don't we have an automated way to do this checking (and desk-rejecting if fake citations found) *before* the papers are sent to reviewers?

NeurIPS papers contained 100+ AI-hallucinated citations, new report claims | Fortune An analysis of NeurIPS 2025 papers by startup GPTZero reveals how AI-generated citations are slipping into elite academic research.

NEW: NeurIPS,one of the world’s top academic AI conferences, accepted research papers with 100+ AI-hallucinated citations, new report claims

fortune.com/2026/01/21/n...

Competitive drinking (beer bong, beer pong, boat race etc) probably isn't a work transferable skill (welll, maybe in the good ol' days) but the ability to skol a drink comes in handy when you have to do the oral glucose tolerance test...

When two years of academic work vanished with a single click After turning off ChatGPT’s ‘data consent’ option, Marcel Bucher lost the work behind grant applications, teaching materials and publication drafts. Here’s what happened next.

lmao get wrecked imo www.nature.com/articles/d41...

Key Points Question Does acetaminophen use during pregnancy increase children's risk of neurodevelopmental disorders? Findings In this population-based study, models without sibling controls identified marginally increased risks of autism and attention-deficit/hyperactivity disorder (ADHD) associated with acetaminophen use during pregnancy. However, analyses of matched full sibling pairs found no evidence of increased risk of autism (hazard ratio, 0.98), ADHD (hazard ratio, 0.98), or intellectual disability (hazard ratio, 1.01) associated with acetaminophen use. Meaning Acetaminophen use during pregnancy was not associated with children's risk of autism, ADHD, or intellectual disability in sibling control analyses. This suggests that associations observed in other models may have been attributable to confounding. Abstract Importance Several studies suggest that acetaminophen (paracetamol) use during pregnancy may increase risk of neurodevelopmental disorders in children. If true, this would have substantial implications for management of pain and fever during pregnancy. Objective To examine the associations of acetaminophen use during pregnancy with children's risk of autism, attention-deficit/hyperactivity disorder (ADHD), and intellectual disability.

Design, Setting, and Participants This nationwide cohort study with sibling control analysis included a population-based sample of 2,480,797 children born in 1995 to 2019 in Sweden, with follow-up through December 31, 2021. Exposure Use of acetaminophen during pregnancy prospectively recorded from antenatal and prescription records. Main Outcomes and Measures Autism, ADHD, and intellectual disability based on International Classification of Diseases, Ninth Revision and International Classification of Diseases, Tenth Revision codes in health registers. Results In total, 185909 children (7.49%) were exposed to acetaminophen during pregnancy. Crude absolute risks at 10 years of age for those not exposed vs those exposed to acetaminophen were 1.33% vs 1.53% for autism, 2.46% vs 2.87% for ADHD, and 0.70% vs 0.82% for intellectual disability. In models without sibling control, ever-use vs no use of acetaminophen during pregnancy was associated with marginally increased risk of autism (hazard ratio [HR], 1.05 [95% CI, 1.02-1.08]; risk difference [RD] at 10 years of age, 0.09% [95% Cl, -0.01% to 0.20%l), ADHD (HR, 1.07 [95% Cl, 1.05-1.10]; RD, 0.21% [95% Cl, 0.08%-0.34%]), and intellectual disability (HR, 1.05 [95% CI, 1.00-1.10]; RD, 0.04% [95% Cl, -0.04% to 0.12%]). To address unobserved confounding, matched full sibling pairs were also analyzed. Sibling control analyses found no evidence that acetaminophen use during pregnancy was associated with autism (HR, 0.98 [95% Cl, 0.93-1.04]; RD, 0.02% [95% Cl, -0.14% to 0.18%l), ADHD (HR, 0.98 [95% Cl, 0.94-1.02]; RD, -0.02% [95% Cl, -0.21% to 0.15%]), or intellectual disability (HR, 1.01 195% Cl, 0.92-1.10]; RD, 0% [95% Cl, -0.10% to 0.13%1).

In short, acetaminophen exposure during pregnancy is not linked to the risk of developing autism, ADHD, or intellectual disability.

The study has been published in Journal of the American Medical Association (JAMA). YES, it is PEER-REVIEWED.
• jamanetwork.com/journals/jam...

🧪🧵⬇️

WELL NOW, would you look at that?

A massive, 26-YEAR-LONG study of MORE THAN 2.4 MILLION people in Sweden found found NO EVIDENCE to support a causal link between acetaminophen (the API in Tylenol) use during pregnancy and increased risk of autism, ADHD, OR intellectual disability in children.

I'm sorry. As a technology writer, I'm supposed to be telling you that this bet will some day pay off, because one day we will have shoveled so many words into the word-guessing program that it wakes up and learns how to actually do the jobs it is failing spectacularly at today. This is a proposition akin to the idea that if we keep breeding horses to run faster and faster, one of them will give birth to a locomotive. Humans possess intelligence, and machines do not. The difference between a human and a word-guessing program isn't how many words the human knows. I'm sorry. I know that when we talk about "digital sovereignty," we're obliged to talk about how we can build more data-centres that we can fill up with money-losing chips from American silicon monopolists in the hopes of destroying as many jobs as possible while blowing through our clean energy goals and enshittifying as much of our potable water as possible.

Córy Doctorow with another verbal bullseye: pluralistic.net/2026/01/13/n...

Black text on white background. Screenshot of ARC’s Network Message regarding delays to grant announcements because of new security arrangements.

⁉️The ARC has delayed outcomes of ALL grants 1–4 months & increased scheduled outcome windows from 2 weeks to 3 months!

This reverses 4 years of progress in providing greater certainty & ability to plan for researchers, their families & unis.

Their excuse? Security checks under new ARC legislation👇

How's that even going to work with the 2x DP cap, given +1 yr on responses? Seems as well thought through as a paper raincoat.

ECR/MCR precarity is already ridiculous, but this is only going to make it worse. Genuinely have to wonder if it'd be easier for the ARC if we all just left research.

What is the future of feedback in the age of Artificial Intelligence? Join our panel, facilitated by Prof Margaret Bearman, to discuss the manifesto for of feedback in the age of #GenAI and kick off our Seminar Series for 2026!

Read more and register: blogs.deakin.edu.au/cradle/what-...

Let’s not let our analysis stop at the goonerism of it all. I’m begging for our view of this to be wider than some guy’s stained pants.

It’s a labour weapon. A missile to bomb us out of public life. One email to a boss and a woman loses her job, her income, and her independence. That’s the point.

Getting grok to publicly strip a woman’s likeness online or alter her image isn’t just horrifying at the level of the individual, it’ll be used to get us fired, too. It’s a technique to keep us in the home popping out babies because we lost our teaching job when some guy sent porn of us to our boss.

Indexicality lesson for students and others who submit attachments. You need to label your file with your OWN NAME, not "Chapter 6" or "Paris conference presentation". The Paris conference will receive 400 files, 395 of which will have identical/similar names. Recipe for chaos. #academicsky #PhDchat

Websites need to realise that no one ever, ever wants the site to be able to send them notifications.

It seems they had a slight smattering of citrus peel, and nothing else.

Wanted to eat a hot cross bun so dug one out of the freezer. Somehow, they appear to be almost fruitless 🙄 and also possibly homemade. I don't recall committing this atrocity, and yet here I am eating it....

"And I never say no"

We need to have a serious talk about the way "AI companion" apps not only prey on the vulnerable, but are priming their users to ignore consent and to conflate love with control.

We need AI regulations across so many sectors, but this area is particularly horrifying.

Comic. [Banner: Mathematical Society: 2025 Meeting.] PERSON 1 with bun: Any other new developments from the year to cover before we wrap? PERSON 2: Oh, the teens picked a new funny number. PERSON 3 with short hair: Aww, I’m glad to hear they’re still doing that. PERSON 4 with ponytail: I’ll add it to the list. [List: 23 (skidoo!; 42; 69; 420; 1,337; 58,008; [circled]: 67]

Funny Numbers

xkcd.com/3184/

Text reads: About synthetic panels Recruiting the right participants for a study can be difficult. You may not get the exact demographics you need, and the shorter the deadline, the less sure you can be that everyone will answer on time. One possible solution can be to use synthetic panels. Synthetic panels are powered by a first party proprietary AI model developed here at Qualtrics. Our synthetic panel is trained on thousands of responses from a variety of demographic backgrounds in order to more accurately predict how certain populations would respond to a survey. Our synthetic panel is based on the United States General Population, and is only available in English. This panel comes with ready-made quotas and target breakouts in order to represent your chosen population and make it easy to launch your survey right away.

Text reads: Question-writing best practices To get the most reliable and actionable results from synthetic audiences, consider these question-writing best practices: Ask forward-looking and attitudinal questions. Synthetic panels perform best with perceptions, preferences, and intent-based questions. For example, “How likely are you to try…?” Synthetic panels are less applicable for studies on past behaviors, detailed recall, brand recall, or awareness questions. For example, “When did you last visit…?”

Text reads: Discussion The current study aimed to conduct a meta-analysis of the TPB when applied to health behaviours which addressed the limitations of previous reviews by including only prospective tests of behaviour, applying RE meta-analytic procedures, correcting correlations for sampling and measurement error, and hierarchically analysing the effect of behaviour type and sample and methodological moderators. Some 237 tests were identified which examined relations amongst model components. Overall the analysis indicated that the TPB could explain 19.3% of the variance in behaviour and 44.3% of the variance in intention across studies. This level of prediction of behaviour is slightly lower than that of previous meta-analytic reviews which have found between 27% (Armitage & Conner, 2001; Hagger et al., 2002) and 36% (Trafimow et al., 2002) of the variance in behaviour to be explained by intention and PBC.

Did you know that from tomorrow, Qualtrics is offering synthetic panels (AI-generated participants)?

Follow me down a rabbit hole I'm calling "doing science is tough and I'm so busy, can't we just make up participants?"

As it turns out, the only way to stop a bad guy with a gun *isn’t* a good guy with a gun—it can be a good guy *without* a gun. Just another NRA lie.