Alex_Altair

• Alex_Altair 6 Mar 2026 1:30 UTC

  13 points

  1

  on: An Align­ment Jour­nal: Com­ing Soon

  I’m sure you’ve had lots of discussion about this; why the label “AI alignment”?

  I think “alignment” refers to the somewhat specific task of aligning an AIs values to human values. But my understanding of your actual scope is more like “theoretical AI safety”. A lot of foundational work is done with the intention that it will eventually help with alignment, but which definitely isn’t about alignment, and a lot of theoretical AI safety work isn’t about alignment per se at all. For example, some of my research problems are trying to understand which types of AI systems are not dangerous, not because their values are aligned with ours, but because they’re not unrestrained consequentialists.

• Alex_Altair 24 Feb 2026 16:55 UTC

  5 points

  4

  in reply to: Ronny Fernandez’s comment on: The truth be­hind the 2026 J.P. Mor­gan Health­care Conference

  I mean, I just think the post isn’t very good. It’s kinda funny, but it’s not THAT funny, and also it’s not the kind of funny that is appropriate for curation. Overall I think curating it wasted a bunch of people’s time and somewhat eroded the commons.

  But, I also don’t expect to convince you personally? Because you are a trickster archetype. Trickster archetypes have a valuable place in society but that place is not in choosing curated posts for LW. So the people I am petitioning to un-curate this are the other LW mods.

How much in­for­ma­tion does an op­ti­mal policy con­tain about its en­vi­ron­ment?

Up­com­ing Dove­tail fel­low talks & discussion

• Alex_Altair 31 Dec 2025 19:26 UTC

  10 points

  2

  on: Open Thread Win­ter 2025/​26

  Hi all! I’m a long-time LWer, but I’m making a comment thread here so that my research fellows can introduce themselves under it!

  For the past year or so I’ve been running the Dovetail research fellowship in agent foundations with @Alfred Harwood. We like to have our fellows make LW posts about what they worked on during the fellowship, and everyone needs a bit of karma to get started. Here’s a place to do that!

When bits of op­ti­miza­tion im­ply bits of mod­el­ing: the Touchette-Lloyd theorem

• Alex_Altair 14 Dec 2025 19:27 UTC

  28 points

  18

  on: The Ax­iom of Choice is Not Controversial

  I think your title “The Axiom of Choice is Not Controversial” is literally false, because “controversial” is a social property referring to how much disageement there is, and not whether something is true. Rejecting the AoC is a minorinty view but it is not that minority. It’s also a respected alternative view, and texts often go out of their way to mention when they are acceping the AoC. It may if anything be the most controversial thing in math.

An­swer­ing a child’s questions

• Alex_Altair 2 Dec 2025 21:05 UTC

  6 points

  2

  in reply to: Daniel Kokotajlo’s comment on: Fu­ture Proofing Solstice

  I basically agree with all of this. I had this thought in part because I have the fairly strong sense that (current, US) society has a belief that there is a healthy state we should be aiming for, and does a lot of question substitution like “does this feel good/​acceptable”. There are much better and worse ways to orient to x-risk, but it may be that all of them feel “bad/​unacceptable” and it’s hard for me to see how they could be accurately labeled “healthy”.

• Alex_Altair 2 Dec 2025 5:05 UTC

  15 points

  1

  on: Fu­ture Proofing Solstice

  I’ve been worried that the Solstice isn’t really set up to handle “actually looking at human extinction in nearmode” in a psychologically healthy way

  A thought I had when I read this sentence was, “what makes you think there is a healthy way?” It kinda feels to me like, just as there’s no way to get in a car accident that’s healthy for your body, there’s no way to actually believe the world is going to end that’s “healthy” for your mind. There could be better or worse ways it could impact you, but you’re gonna get some impact trauma.

• Alex_Altair 29 Nov 2025 6:26 UTC

  2 points

  0

  on: The Miss­ing Genre: Heroic Par­ent­hood—You can have kids and still punch the sun

  Is there any polyamorous fanfiction of The Incredibles?

• Alex_Altair 21 Nov 2025 7:03 UTC

  64 points

  27

  on: Dom­i­nance: The Stan­dard Every­day Solu­tion To Akrasia

  I think something like “accountability” is a much more accurate term for this generalization than “dominance”. Parents, teachers, bosses etc. are roughly as motivated by their accountability to their children, students and employees.

• Alex_Altair 19 Nov 2025 22:23 UTC

  3 points

  0

  on: Re­search Reflections

  Finite Factored Sets

  Nitpick; I thought “finite factored sets” were being rebranded as factored space models, because the finiteness was inessential. I think this name is better and would like to see the switch-over happen before the framework gets more popularity. But maybe I’m wrong about this intention?

A re­view of Red Heart, the new al­ign­ment novel by Max Harms

I store some mem­o­ries spa­tially and I don’t know why

What would my 12-year-old self think of agent foun­da­tions?

• Alex_Altair 16 Nov 2025 6:06 UTC

  46 points

  0

  on: Alex_Al­tair’s Shortform

  A hot math take

  As I learn mathematics I try to deeply question everything, and pay attention to which assumptions are really necessary for the results that we care about. Over time I have accumulated a bunch of “hot takes” or opinions about how conventional math should be done differently. I essentially never have time to fully work out whether these takes end up with consistent alternative theories, but I keep them around.

  In this quick-takes post, I’m just going to really quickly write out my thoughts about one of these hot takes. That’s because I’m doing Inkhaven and am very tired and wish to go to sleep. Please point out all of my mistakes politely.

  The classic methods of defining numbers (naturals, integers, rationals, algebraic, reals, complex) are “wrong” in the sense that it doesn’t match how people actually think about numbers (correctly) in their heads. That is to say, it doesn’t match the epistemically most natural conceptualization of them: the one that carves nature at its joints.

  For example, addition and multiplication are not two equally basic operations that just so happen to be related through the distributivity property, forming a ring. Instead, multiplication is repeated addition. It’s a theorem that repeated addition is commutative. Similarly, exponentiation is repeated multiplication. You can keep defining repeated operations, resulting in the hyperoperator. I think this is natural, but I’ve never taken a math class or read a textbook that talked about the hyperoperators. (If they do, it will be via the much less natural version that is the Ackermann function.)

  This actually goes backwards one more step; addition is repeated “add 1”. Associativity is an axiom, and commutativity of addition is a theorem. You start with 1 as the only number. Zero is not a natural number, and comes from the next step.

  The negative numbers are not the “additive inverse”. You get the negatives (epistemically) by deciding you want to work with solutions to all equations of the form $a+x=b$ for naturals $a$ and $b$. The fact that declaring these objects to exist is consistent should be a theorem, as should the fact that some of the solutions to different equations are equal (e.g. that the solution to $2+x=1$ is the same as the solution to $3+x=2$).

  This idea is also iterated up through the hyperoperations. The rational numbers are (again, epistemically) the set of all solutions to the equations $a\ast x=b$ where $a$ and $b$ are integers. The fact that this set is not consistent when $a$ is $0$ should also be a theorem.

  Since the third-degree hyperoperation $a^b$ is not commutative, you can demand two new types of solutions, those of $x^a=b$ and those of $a^x=b$. This gives you roots and logs. The fact that roots require you to define the imaginary numbers, and therefore lose the total ordering over the numbers, should be a theorem.

  Some of the solutions to these equations are numbers we already had, and some of them are new numbers that we didn’t have This setup leads to the natural question of whether we will keep on producing new types of numbers or not. The complex numbers are special in part because they are closed under all these inverse operations. But do we need new types of numbers if we demand solutions to the fourth-order hyperoperator? What happens if we go all the way up? I have no idea.

• Alex_Altair 14 Nov 2025 4:17 UTC

  9 points

  0

  in reply to: RobertM’s comment on: 8 Ques­tions for the Fu­ture of Inkhaven

  Yeah, I agree. An earlier deadline would cut out procrastination time, or more accurately, it would just make people write more efficiently per unit time. I’d honestly be pretty excited to try having a deadline even as early as 6pm.

Types of sys­tems that could be use­ful for agent foundations

• Alex_Altair 6 Nov 2025 17:23 UTC

  7 points

  0

  in reply to: Ben Pace’s comment on: Peo­ple Seem Funny In The Head About Sub­tle Signals

  Some subtle signals perhaps?