apriiori

links for ♍️ 2025

you will be subjected to random april interests and you will like it

Respected rationalist bloggers sometimes do these, so I figured it was a good idea to start accumulating links so I have something easy to click publish on if I fail to write a full post some day. Probably the disaster LGBTESCREAL blog minisphere can use someone doing these.

α. You probably know about If Anyone Builds It, Everyone Dies, Eliezer Yudkowsky and Nate Soares’s new New York Times bestseller on existential risk from super AI, but do you know about the page MIRI helpfully set up to make it as convenient as possible to do some concrete helpful actions? I hear it only takes like a handful of calls to a representative before it makes a small difference.

I've always sort of floundered at figuring out how to do anything actually helpful about the whole thing where we very well might be careening towards the largest disaster the world has ever seen—at least if you consider Geoffrey Hinton or Yoshua Bengio, for example, reliable sources on AI. So I figure we owe it to the effort, now that MIRI has made a clear, concrete request and set up a web page to make it as easy as they possibly can for us11 If any of y’all end up reading this, thank you!!! This feels like social technology straight out of a rationalist medianworld to combat the trivial inconveniences that often thwart dysfunctional girls like me. I’m impressed., to go ahead and make that call or write that letter22 I WILL GET TO IT. I promise.. You can also sign up for the potential march, or if you’re worried about travel expenses (as I am) just tell them to let you know if one gets scheduled so you can make the decision then.

β. There is now an official Tetraspace reference sheet, for everyone out there who’s been excited to draw Tetraspace. You can also give the reference sheet to AIs.

γ. Autumn has written up the ways she’s updated partially away from MIRIesque AI existential risk models. I remain fairly uncertain here (though I’m not sure if that’s just the sin of underconfidence), but a lot of her points seem pretty compelling to me. But I don’t want to rely on getting lucky, and the somewhat speculative ideas I've heard about how the luckiness might be robust don’t feel solid enough that I want to place my faith in them, and even if we don’t get a coldly strategizing AI we’re not particularly out of the woods.

δ. Have you ever wanted to play Super Mario Bros. as W. D. “Mystery Man” Gaster? Well now you can. In case that was something you’ve been dying to get the chance to do.

ε. A new Super Mario 63 speedrun category just dropped. The level designer preserve has finally been routed into No Major Skips. I used to speedrun this game a lot when I was younger, and still play it occasionally, so seeing developments like this is very exciting.

ζ. While you may have found the blogs of those friends of mine who started Substacks recently, Thiago has had a blog for longer than the recent burst. Check it out if you want!

η. A few years back, I made a prototype for a chess variant inspired by the popular Chess 960, which my friends Matthew Bolan and Geosquare from back in my Minecraft speedrunning days helped make into the modern Chess 9b60 website. It isn’t really the most beginner friendly thing, if you’re not already used to playing weird chess variants, but I think it’s neat and am not sure how many people are aware of it yet.

θ. I forgot to write a Petrov day post (sorry Saintslav, and thank you from the bottom of my heart for not destroying the world), but I learned from Octavia’s Petrov day post that, apparently, Petrov day is actually recognized by the United Nations as the International Day for the Total Elimination of Nuclear Weapons. Neat! I think I basically support this cause modulo the potential need to nuke an asteroid or something.

ι. Apparently, Oklahoma Governor Kevin Stitt has not read Patrick McKenzie classic Falsehoods Programmers Believe About Names, which I consider required reading33 Or at least, which exhibits some general principles one ought be familiar with, probably there are other ways to get the same nutrients. for anyone who aspires to understand the world. Those ontologies you’ve never questioned in your whole life are flimsy.

κ. Noah Topper is hosting a reading group for Good and Real, which “Eliezer Yudkowsky has said (…) is the best explanation for his own theory of consciousness.” It begins on the sixth. I and my old podcast cohost Eneasz Brodski will be there.

λ. The birth rate decline statistics might be misleading, because parents choosing to have children later can produce an apparent drop even if we ultimately expect them to have the same number of kids. This does not mean we are out of the woods, but it seems important to forming a detailed model of the situation.

μ. Pope Leo defends Chicago Cardinal Cupich’s plans to award pro-choice Senator Richard Durbin of Illinois for work on immigration policy. I've frequently been impressed by the Church lately—at the homily last week the pastor was criticizing Elon Musk for saying empathy of all things was the problem with the world, and there was a USAID mention!—and I pray that American bishops can someday get their heads on straight. And also for JD Vance’s soul.

ν. Vertic got 22:21 in No Major Skips! The sub 22 seems more possible every day, though it’s still a ways off.

  1. If any of y’all end up reading this, thank you!!! This feels like social technology straight out of a rationalist medianworld to combat the trivial inconveniences that often thwart dysfunctional girls like me. I’m impressed.

  2. I WILL GET TO IT. I promise.

  3. Or at least, which exhibits some general principles one ought be familiar with, probably there are other ways to get the same nutrients.