Found it: https://preservinghope.substack.com/p/a-victory-for-the-natural-order
noggin-scratcher
Karma: 895
How should we speak about “stressful events”? Maybe instead of, “buying a plane ticket is stressful”, something like, “buying a plane ticket made me stressed.” But the word “made” implies inevitability and still cedes too much power to the event
“I am feeling stressed about buying a plane ticket” would acknowledge that the stress is coming from within you as an individual, and doesn’t foreclose the possibility of instead not feeling stressed.
Pretty sure I’ve seen this particular case discussed here previously, and the conclusion was that actually they had published something related already, and fed it to the “co-scientist” AI. So it was synthesising/interpolating from information it had been given, rather than generating fully novel ideas.
Per NewScientist https://www.newscientist.com/article/2469072-can-googles-new-research-assistant-ai-give-scientists-superpowers/
However, the team did publish a paper in 2023 – which was fed to the system – about how this family of mobile genetic elements “steals bacteriophage tails to spread in nature”. At the time, the researchers thought the elements were limited to acquiring tails from phages infecting the same cell. Only later did they discover the elements can pick up tails floating around outside cells, too.
So one explanation for how the AI co-scientist came up with the right answer is that it missed the apparent limitation that stopped the humans getting it.
What is clear is that it was fed everything it needed to find the answer, rather than coming up with an entirely new idea. “Everything was already published, but in different bits,” says Penadés. “The system was able to put everything together.”
That was concerning the main hypothesis that agreed with their work. Unknown whether the same is also true for its additional hypotheses. But I’m sceptical by default of the claim that it couldn’t possibly have come from the training data, or that they definitely didn’t inadvertently hint at things with data they provided.
Technically it’s still never falsifiable. It can be verifiable, if true, upon finding yourself in an afterlife after death. But if it’s false then you don’t observe it being false when you cease existing.
https://en.wikipedia.org/wiki/Eschatological_verification
If we define a category of beliefs that are currently neither verifiable or falsifiable, but might eventually become verifiable if they happen to be true, but won’t be falsifiable even if they’re false—that category potentially includes an awful lot of invisible pink dragons and orbiting teapots (who knows, perhaps one day we’ll invent better teapot detectors and find it). So I don’t see it as a strong argument for putting credence in such ideas.
Looks like #6 in the TL;DRs section is accidentally duplicated (with the repeat numbered as #7)
Solid point. I realise I was unclear that for face shape I had in mind external influences in utero (while the bones of the face are growing into place in the fetus). Which would at least be a somewhat shared environment between twins. But nonetheless, changing my mind in real-time, because I would have expected more difference from one side of a womb to the other than we actually see between twins.
Even if I’m mistaken about faces though, I don’t think I’m wrong about brains, or humans in general.
In other words, all the information that controls the shape of your face, your bones, your organs and every single enzyme inside them – all of that takes less storage space than Microsoft Word™.
The shape of your face, and much else besides, will be affected by random chance and environmental influences during the process of development and growth.
The eventual details of the brain, likewise, will be in large part a response to the environment—developing and learning from experience.
So the final complexity of a human being is not actually bounded by the data contained in the genome, in the way described.
I wasn’t the one eating it, but having prepared a couple of Huel’s “hot meal pot/pouch” options for my partner (I forget which ones exactly, but something in the way of mac & cheese or pasta bolognese), I can report that I found the smell coming off it to be profoundly unappetising.
Not sure how they went down with her, but there’s a small stash of these pots in the cupboard that she hasn’t touched beyond the first few—so I suspect not very well.
Slight glitches:
The “chapter shortcuts” section of https://www.lesswrong.com/s/9SJM9cdgapDybPksi lists “editPost” links to the chapter drafts (inaccessible to others)
The numbering in the post titles skip over #4
Oh I was very on board with the sarcasm. Although as a graduate of one of them, I obviously can’t believe you’re rating the other one so highly.
This is a general principal
Principle* — unless they’re the head-teacher of a school, the type to be involved in a principal/agent problem, or otherwise the “first”
graduates of the great English universities (both of them)
Shots fired
That definitely looks like the one. Appears I’d forgotten some of the context/details though.
I could swear there was a similar Scott Alexander post, about flirting deliberately skirting the edge of plausible deniability to avoid prematurely creating common knowledge. With an analogy to spies trying to identify a fellow operative without overtly tipping their hand in case they were mistaken and speaking to a non-spy.
Can’t find it now: might have since been deleted, or might have only ever existed on LiveJournal or Tumblr or something.
https://slatestarcodex.com/2017/06/26/conversation-deliberately-skirts-the-border-of-incomprehensibility/ is similar but not explicitly about flirting.
While I can appreciate it on the level of nerd aesthetics, I would be dubious of the choice of Quenya. Unless you’re already a polyglot (as a demonstration of your aptitude for language-learning), it seems unlikely—without a community of speakers to immerse yourself in—that you’ll reach the kind of fluid fluency that would make it natural to think in a conlang.
And if you do in fact have the capacity to acquire a language to that degree of fluency so easily, but don’t already have several of the major world languages, it seems to me that the benefits of being able to communicate with an additional fraction of the world’s population would outweigh those of knowing a language selected for mostly no-one else knowing it.
The strategy above makes all three statements seem equally unlikely to be true. Mathematically equivalent but with different emphasis would be to make all three statements seem equally unlikely to be false.
i.e. Pick things that seem so mundane and ordinary that surely they must be universally true—then watch the reaction as it is realised that one of them must actually be a lie.
I suspect there has to be a degree of mental disconnect, where they can see that things don’t all happen (or not happen) equally as often as each other, but answering the math question of “What’s the probability?” feels like a more abstract and different thing.
Maybe mixed up with some reflexive learned helplessness of not really trying to do math because of past experience that’s left them thinking they just can’t get it.
Possibly over generalising from early textbook probability examples involving coins and dice, where counting up and dividing by the number of possible outcomes is a workable approach.
I know someone who taught math to low-ability kids, and reported finding it difficult to persuade them otherwise. I assume some number of them carried on into adulthood still doing it.
In the infinite limit (or just large-ish x), the probability of at least one success, from nx attempts with 1/x odds on each attempt, will be 1 - ( 1 / e^n )
For x attempts, 1 − 1/e = 0.63212
For 2x attempts 1 − 1/e^2 = 0.86466
For 3x attempts 1 − 1/e^3 = 0.95021
And so on
Ironically, the even more basic error of probabilistic thinking that people so—painfully—commonly make (“It either happens or doesn’t, so it’s 50/50”) would get closer to the right answer.
Cool service/feature, but would it be worth defusing the “jumpscare” with an interstitial that explains the function of the button? At least the first time any given user clicks it.