AI is a notorious confabulator

Chuck Wendig is a well-known author, and unsurprisingly, people are curious about him. He’s the subject of various harmless inquiries, and he has discovered, entertainingly, that AI makes up a lot of stuff about him. For instance, you can ask Google Gemini the name of his cat.

Unfortunately, Wendig is catless.

Well! That answers that. Apparently, unbeknownst to me, I actually do have a cat, as the *checks notes* Wengie Wiki will tell you. This isn’t unusual. Cats are very often little hide-and-seeky guys, right? Dear sweet Boomba is probably just tucked away in some dimensional pocket inside our house.

That leads him down a rabbit hole to discover that he has had and has multiple cats, swarms of cats, that have died and been replaced by other named cats, and he also has more dogs than he expected.

It’s a trivial example, but it illustrates a general problem with our brave new world of AI.

Generative AI is a sack of wet garbage.

Do not use AI for search.

DO NOT USE AI FOR SEARCH.

AI can’t even do the basic math right. Meanwhile it hallucinates endless nonsense things! So many false things! It would generate new false things if I gave it the same question string twice. This is only the tip of the iceberg for the weird things I got it to assure me were true.

I’ll pass the word on to my writing class next semester.

Then I was curious about what chatGPT thinks about my cat, so I asked it, even though I’m nowhere near as prominent as Chuck Wendig. Of course it had an answer!

“Mochi”? Wait until the evil cat finds out. It will be shredded.

I couldn’t resist clicking on the button to find out more about PZ Myers’ pets. I got a whole biography!

That’s a grade-school level essay, full of generic nonsense written to be bland and inoffensive, and could be applied to just about anyone. I’d accept it if it were written by someone in 3rd grade, but I’d still ask them where they got the information.

Notice that it doesn’t mention “spider” even once.

I repeat: DO NOT USE AI FOR SEARCH.

Try it. Tell me all about AI’s fantasies about your pets in the comments.

I envy them their coats

It’s unpleasantly cold this morning, and I was reminded of this old photo from the early years of the 20th century. That’s my great-grandfather Peter on the far left, with his sisters Karen and Marie, and his brothers Ole and Iver. I’m just impressed with their coats; I’d take any of the men’s coats right now. They knew how to dress for a Minnesota winter!

Let’s talk on Saturday

I’m a terrible YouTuber — uncharismatic, dull, lacking in visual skills, and incapable of maintaining a consistent schedule — but heck, I’ll try again. Tomorrow (Saturday, 20 December), I’ll go live around 3pm Central time. I’m open to talking about just about anything, but will center the discussion on this paper:

Christopher J. Kay, Anja Spang, Gergely J. Szöllősi, Davide Pisani, Tom A. Williams & Philip C. J. Donoghue (2025) Dated gene duplications elucidate the evolutionary assembly of eukaryotes. Nature, 3 December 2025, DOI: 10.1038/s41586-025-09808-z.

If you don’t want to read an 11 page technical paper, just contemplate this figure:

Or you can just tune in and maybe I’ll explain it.

DDT causes polio???!?

I think I’m trapped at home today — I tried walking to work, and didn’t get beyond my driveway, because we had a thaw and a refreeze and it’s slick as snot out there. Then we’re supposed to get more snow this morning, with temperatures plummeting down to -15°C with 50mph wind gusts, so I’m cowering at home today. The spiders will go hungry for a day (they are opportunistic feeders, they can handle it).

If you’re similarly stuck at home, here’s an hour long video that I thought was very good. It rips into a couple of self-styled “science” based influencers who are anything but.

The most shocking bit was seeing Joe Rogan getting furious at any push-back on his anti-vax views, and basically shutting down the conversation by claiming that the polio epidemic was co-incident with they years of heaviest DDT use. He also made the standard skeptical claim that vaccines were a late response to an already fading plague, which is sort of true. There are multiple approaches to a serious disease: behavioral shifts, like self-quarantine, and improved hygiene can reduce the incidence and severity of infections, but it takes efficacious medical responses to deliver the coup de grace. And Joe Rogan doesn’t understand science at all if he falls for the correlation equals causation canard. DDT does not cause polio.

The video also jumps on Bill Maher. He’s got this canned response to any claims, saying that we don’t know 100% of everything, more like 20% or 10%, so his weird fads might be true. It’s nonsense. Of course there is much left to learn, but we can say with 100% confidence that you shouldn’t eat cyanide, or that the earth is spherical, or that vaccines don’t cause autism, because smart, skeptical people have studied that stuff and have objective data to back up their arguments. We don’t even quantify knowledge as a percentage fraction of everything, so that’s a bogus metric anyway. I’m willing to go along with a claim that we only know 0.00001% of everything, but that the bits we know, we know pretty damn well, so please, Bill Maher, don’t jump off the roof of a New York skyscraper to test your ‘theory’ of gravity.

Another good topic was about what having a PhD means. It’s not a free pass to make everything you say valuable, important, and true. It just says you passed an apprenticeship. You presumably got some training in critical thinking which the Joe Rogans of the world lack, but you have to demonstrate your skills throughout your life. There are also some really bad theses out there — there is some pressure to get students out the door so you can get a new crop started, and some bad PIs who will let garbage pass as long as they get a publication out of it.

(By the way, I think my PhD thesis holds up. Not only did multiple researchers build on it afterwards, but it wasn’t even just mine — it was the product of a collaboration with several absolutely brilliant mentors and colleagues, which is how every thesis ought to be.)

Avi Loeb is nuts

I’ve been watching the trainwreck named Avi Loeb for a while now, and it’s become obvious that he’s shredding his own reputation, that of Harvard astronomy, and of good science in general. He really ought to step down and retire to pursue his weird hobby — maybe he could get a special on Netflix? Anyway, Rebecca Watson summarized his current record for triumphant farts, and that’s a good thing, because I’m too tired of him to do it myself.

Hey, while I was thinking of YouTube, I figured maybe I’d do a livestream on Saturday afternoon. Would anyone be interested?

The spiders will feast tonight!

When I become an evil overlord, that will be my catchphrase: “The spiders will feast tonight!”

I got to the bait shop this morning shortly after they got a bulk delivery, and just before they parceled them out into smaller batches, so I was able to buy a whole tub of 20 dozen (240) waxworms for $15.99. That’ll take care of feeding supplies for the next few weeks, right through Christmas, so mission accomplished.

Look at all that squirmy cold protein in thin casings! I was tempted to pop a few in my mouth, but that would be taking food from my babies, so I didn’t.

A good use for AI

You can use AI to spy out AI!

GPTZero, the startup behind an artificial intelligence (AI) detector that checks for large language model (LLM)-generated content, has found that 50 peer-reviewed submissions to the International Conference on Learning Representations (ICLR) contain at least one obvious hallucinated citation—meaning a citation that was dreamed up by AI. ICLR is the leading academic conference that focuses on the deep-learning branch of AI.

The three authors behind the investigation, all based in Toronto, used their Hallucination Check tool on 300 papers submitted to the conference. According to the report, they found that 50 submissions included at least one “obvious” hallucination. Each submission had been reviewed by three to five peer experts, “most of whom missed the fake citations.” Some of these citations were written by non-existent authors, incorrectly attributed to journals, or had no equivalent match at all.

The report notes that without intervention, the papers were rated highly enough that they “would almost certainly have been published.”

It’s worse than it may sound at first. One sixth of the papers in this sample had citations invented by an AI…but the citations are the foundation of the work described in those papers. The authors of those papers apparently didn’t do the background reading for their research, and just slapped on a list of invented work to make it look like they were serious scholars. They clearly aren’t.

The good news is that GPTZero got a legitimate citation out of it!

How does this nonsense persist?

I keep seeing lies about biology like this on social media. This is all wrong.

A man’s DNA can stay in a woman for years sometimes. The DNA of other men can even affect her next child’s physical features or mental traits especially if the women had a child with another man or even a miscarriage or abortion. This was probably well know by the ancients or modern day religions like Islam which is why they would strongly prefer to marry and have children with virgins. So let it be known fellas, the more your girl slept around before you met her the more men’’s DNA is flowing through her like ghosts of the past, and her 🐱 is the cemetery gates

I have embryology textbooks all over the place, and this is what they have to say.

Sperm can survive in a man’s testicles for a few months. That’s a pleasant environment for a sperm cell, but even there they’re undergoing a slow process of maturation, and are eventually going to be broken down and resorbed. That’s the maximum longevity for these cells.

A woman’s reproductive tract is warm and moist and allows for limited survival, but is adapted to protect her from infections by maintaining a mildly acidic environment. It’s a somewhat hostile place that has to tolerate foreign cells, but not for long — sperm will last for 3-5 days, not years.

You may have heard of a phenomenon called microchimerism, in which fetal cells, shed during pregnancy, persist for years. These are somatic cells, not sperm cells. Sperm cells are highly specialized and have minimized their cytoplasm and cannot survive for that long.

The textbooks usually mention that sperm can survive for about an hour outside of a human body. That’s optimistic (or pessimistic, if you want to avoid pregnancy). Evaporation or absorption of the surrounding fluid is going to kill the little fuckers pretty fast.

It’s clear how these bad ideas get around — the hint is in the text. The “ancients” or “religion” are terrible sources of information, since none of them had anything but the vaguest notion of how reproduction or inheritance work, and didn’t even know about the existence of cells until, at best, three hundred years ago. Another obvious source is a cultural bias favoring virginity (a bogus concept already), and this is an attempt to rationalize that belief with made-up “facts”.

If we’re going to hang any member of the “loser generation,” start with…

He looks like a genuine American Psycho

Joe Lonsdale. He’s emblematic of the problems of the rich: he’s a billionaire, he’s an entitled dumbass, he despises education, and he wants to kill people less privileged than he is. He has a name for students who seek accommodations for disabilities.

Loser generation. At Stanford, it’s a hack for housing, though and at some point, I get it, even if it’s not my personal ethics. Terrible leadership from the university.

College is only for rich, healthy people who want to join a frat. He also disparages university education.

Claiming your child has a disability to give them a leg up became an obvious dominant game theoretic strategy for parents without honour in the 2010s. Great signal to avoid a family / not do business with parents who act this way.

…

No great companies are interested in the BS games played by universities.

What great companies? Lonsdale was a member of the PayPal Mafia, and he made his fortune at Palantir — not a great company by any means, but a damned evil one. He got a BS degree in computer science from Stanford, one of those elite colleges that the rich pollute with their bogus aspirations. Who needs an education when any dumbass can get filthy rich by gouging government contracts and being buddies with other amoral billionaires.

Remember, this guy was a cofounder of the University of Austin — he’s undermining his own scam by scorning universities, which tells you he’s not very bright.

Now he’s also defending turning murder into spectacle.

Joe Lonsdale defended open-air executions in a series of posts on X, saying their reintroduction would be an example of the “masculine leadership” the country is sorely lacking.

“If I’m in charge later,

Wait. Hold it right there. He has hopes of being in charge? Dear god no.

we won’t just have a three strikes law. We will quickly try and hang men after three violent crimes. And yes, we will do it in public to deter others,” he wrote. “Our society needs balance. It’s time to bring back masculine leadership to protect our most vulnerable.”

Biggest sign of an entitled asshole: he throws around the word “masculine” as if it is a flawless great good, and treats “feminine” as a weakness. I want leaders who are human and humane, and don’t think more than half the population are unsuited for learning or leading because of their biology.

Acquit Luigi Mangione. His job isn’t done.

Another mass shooting

This time, at Brown University. It’s deja vu all over again. For some of the students, this was their second school shooting.

As the deadly attack unfolded at Brown University, leaving students hiding under desks and reeling as gunshots rang out, the scene was eerily familiar for at least two students.

Years earlier, Mia Tretta, 21, and Zoe Weissman, 20, had both survived school shootings. “What I’ve been feeling most is just, like, how dare this country allow this to happen to someone like me twice?” Weissman told the New York Times.

Two people were killed and nine others wounded on Saturday after a man dressed in black opened fire during final exams at one of America’s most prestigious colleges. Hundreds of police spent the night scouring the campus and nearby neighbourhoods as the suspect remained at large.

I’m working on some simple statistical problems for my genetics course — you know, basic stuff like the probability that two rare mutations would simultaneously occur in a single individual, etc. I’m not going to set up any problems around this kind of event in America, because this has become too real and too close to home. Also finals are an easy target: students are under stress, they’ve been openly studying (and complaining!) for weeks, and the finals schedules are easy to find online. It’s like we’re advertising that a herd of students will be gathering in a large space at precisely Day X and Time Y, and nobody plans on bringing a gun to such an event, other than cowards with nefarious intent. Maybe we need to start being more secure and confidential with this kind of information.

This story ends with another coincidence. Every story about a mass shooting somehow ends with a fairly typical and familiar conclusion.

Saturday’s attack has again cast a spotlight on longstanding calls for gun control in the US, where gun laws rank among the most permissive in the developed world. So far this year, there have been 389 mass shootings across the US, according to the Gun Violence Archive, which defines incidents where four or more victims have been shot. Last year, more than 500 mass shootings were reported.

Yeah, how weird that every story has to end with a conclusion deploring the USA’s insane gun policy.

I have another familiar part of the conclusion: nothing will be done.