Jay Bailey

Karma: 1,364

Jay Bailey 10 May 2026 0:13 UTC
2 points
0
in reply to: leogao’s comment on: leogao’s Shortform
This is where my knowledge ends I’m afraid—the EA Forum / GiveWell may know more, or perhaps someone here can chime in!

Jay Bailey 9 May 2026 22:31 UTC
6 points
1
in reply to: leogao’s comment on: leogao’s Shortform
I think part of it is what you said. Even in Western countries where people have way more education and slack, people often underinvest in preventative health like flu shots, so it makes sense that people in third-world countries would fall prey to this too, and many more probably don’t even know about the benefits of nets or how to get them.

Another pair is economies of scale and spillover effects. IIRC it’s cheaper per net to protect a whole village at once, and because malaria nets are treated with insecticide, nets kill mosquitos and so they protect other people as well. And people tend to underinvest in things with positive externalities like that—again we go back to the flu shot example.

There’s probably other reasons, but these are the ones I know of.

Jay Bailey 24 Apr 2026 10:33 UTC
10 points
12
in reply to: Richard_Kennaway’s comment on: leogao’s Shortform
I don’t know how many people explicitly believe it but there is a general worldview that inherently assumes it. There are common memes that use this to show the unfairness of pay disparity, such as this one. Inherently it assumes the only fair way for one person to be paid 351x more than another is if they work 351x harder—LTV. https://www.reddit.com/r/antiwork/comments/yrdbyg/ceos_are_not_worth_351_times_the_average_worker/

Jay Bailey 23 Apr 2026 2:43 UTC
3 points
0
in reply to: Bill Seitz’s comment on: There are only four skills: design, technical, management and physical
I can try this:

A human being should be able to: change a diaper, plan an invasion, butcher a hog, conn a ship, design a building, write a sonnet, balance accounts , build a wall, set a bone, comfort the dying, take orders, give orders, cooperate, act alone, solve equations, analyze a new problem, pitch manure, program a computer, cook a tasty meal, fight efficiently, and lastly die gallantly.

Change a diaper—Physical

Plan an invasion—Design

Butcher a hog—Physical

Conn a ship—Physical

Design a building—Design

Write a sonnet—Design

Balance accounts—Technical

Build a wall—Physical. Design if you also need to plan it.

Set a bone—Physical

Comfort the dying—Does not fit (DNF) or maaaaybe Management

Take orders—DNF

Give orders—Management

Cooperate—Management

Act alone—DNF

Solve equations—Technical

Analyze a new problem—Design

Program a computer—Technical

Cook a tasty meal—Physical

Fight efficiently—Physical

Die gallantly—DNF

The ones that don’t fit are generally quite broad cross-domain soft skills (Take orders, act alone) or social/philosophical ones (Comfort the dying, die gallantly) but for the most part, these items can fit pretty well.

Jay Bailey 21 Apr 2026 1:45 UTC
2 points
0
in reply to: bohaska’s comment on: There are only four skills: design, technical, management and physical
Drawing does seem like a physical skill to me. Habryka’s mentioned elsewhere in this comment section that playing an instrument is a physical skill and I think drawing is quite similar to this.

What makes this more complicated is that it seems feasible that a complex piece requires the design skill to configure it, then the physical skill to make it a reality.

Jay Bailey 20 Apr 2026 2:45 UTC
2 points
0
in reply to: davekasten’s comment on: There are only four skills: design, technical, management and physical
Not Habryka, but it seems like according to this article policy creation is pretty solidly a design skill, whereas the art of getting people to get on board with your proposed policy is a management skill. Sometimes policy creation also has technical prerequisites if you’re writing stuff in technical domains that need to engage with specific facts on the ground like FLOP counts and such.

I think producing a policy document is very close to producing a law document, and law is explicitly called out as a design skill.

Jay Bailey 20 Apr 2026 1:36 UTC
11 points
7
in reply to: dan.parshall’s comment on: Reevaluating “AGI Ruin: A List of Lethalities” in 2026
The Orthogonality Thesis says you could have a mind of arbitrary intelligence pointed at any goal, not that any specific training process will wind up pointed at some random target. What you observe is entirely consistent with orthogonality being true.

The main argument this is meant to argue against is “Well, if you had a paperclip maximiser that got really smart, it would realise that maximising paperclips was stupid and decide to do something else instead.” If this seems obviously incorrect to you, then the Orthogonality Thesis has done its job.

Eliezer does make arguments that AI’s will not be pointed at the thing you try to point them at, in a similar way to how evolution didn’t evolve to make humans robustly care about inclusive genetic fitness, but that isn’t quite the same as the Orthogonality Thesis. Orthogonality is a necessary but not sufficient condition for Eliezer’s arguments on this applying to the kind of AI systems we actually train.

Jay Bailey 19 Apr 2026 6:26 UTC
14 points
0
in reply to: habryka’s comment on: There are only four skills: design, technical, management and physical
For the benefit of later readers, this was Scott Young’s MIT challenge, doing a full CS degree in 12 months. I saw this as rather incredible the first time I saw it, but maybe the more incredible thing was having both the circumstances and the agency to attempt it in the first place. https://www.scotthyoung.com/blog/myprojects/mit-challenge-2/

Jay Bailey 19 Apr 2026 6:20 UTC
11 points
1
in reply to: habryka’s comment on: There are only four skills: design, technical, management and physical
Okay, I think I understand this now. For the benefit of other people with Jay-shaped problems in this area, I’ll write down what I’ve learned:

If I were to design a CRUD app, I would immediately begin decomposing the problem into the basic language of the specific domain of web applications. We need a front-end, a back-end, and a database. We need to connect them up in some way. We need to host the application somewhere. Okay, I know what parts are needed.

Now, what is the specific problem I want to solve with this CRUD app? What differentiates it from any generic CRUD app? What information do I need to show to the user and how in order to ensure they can easily do whatever they will want to do with this app? Now I am in “core design skills” land. However, I had to go through that first paragraph to do so, I couldn’t just start here.

So, it feels like I know nothing about interior design, not because I am bad at every step of interior design but because I am bad at the very first step—I cannot yet break down room design into the various components like lighting, furniture, color schemes, and so on that are required. (Honestly, even being able to write that previous sentence with specific examples indicates that I am less clueless than I originally thought of myself as) Since that’s the first step to solving the problem, all the important skills I have that would actually help with this problem are hidden behind the relatively simple ones I lack.

If I’ve got that right, that is a very powerful and new way of looking at things for me! I wouldn’t be surprised if this turns out to be one of the most valuable posts I see this year. It reminds me of when I first picked up an economics textbook—a bunch of things I didn’t understand but were scary black boxes transformed into things where I knew how I could grow to understand them if I wanted.

Jay Bailey 19 Apr 2026 5:29 UTC
30 points
3
on: There are only four skills: design, technical, management and physical
When I look at this article and ask “What is the strongest counter-example that comes into my head”, and I think something like “Being a mathematician or physicist”. I think I have the g-factor to get a math degree if I really wanted, but it’s not at all clear how I could reach the level of a working mathematician in six months, despite already being a decent programmer. There are a few replies my inner Habryka might say, and I’m curious if any of them seem accurate, and if some are very much not what you intended to say:
1. This advice isn’t meant for everyone. In the same way that I’m expecting people to be 140+ IQ to use this advice, I also expect them to be at least +2.5 SD of conscientiousness. If you can put in 60+ hours a week of high quality mental work into this task, you actually could achieve this—most students of mathematics do a fraction of this time. You underestimate how much progress can be made with real focus and drive paired with strong conscientiousness and smarts—it’s a multiplicative effect that adds up very quickly. Such a person really could get to that level in six months, but they’re rare enough that standard career paths can’t accommodate them, which is why society needs specialists in the first place.
2. When you imagine being a working mathematician, you imagine having many years of schooling that teaches you about a very broad variety of maths. This is not what I am imagining—I am imagining having the level of ability to perform a specific type of job a mathematician might do. For example, take this SLT theorist job by Timaeus. https://timaeus.co/blog/updates/2026-04-09-hiring This doesn’t require you to know everything an undergraduate maths degree holds. It requires you to know an important subset and to grasp some universal skills like writing rigorous proofs and doing mathematical research at all. Rolling your own curriculum could get you there in 6 months. Think about your own CS career—does someone need to have learned everything you learned in order to do what you do? Clearly not—if they’d aimed at your specific target from the start, they’d get there in six months.
3. The kind of person who would be an actual research-grade mathematician really is a level of expertise beyond this post. It’s a level beyond, say, a solid mid-level programmer in their field, and to get to that level it really does take years even if you’re good. It would take months to get to the level of doing an easier class of problems that people would still pay you for, and that’s the level I would expect a Lightcone-level generalist to be able to get to in a different area in the same domain within six months.
Or it could be something else entirely.

Jay Bailey 19 Apr 2026 5:08 UTC
9 points
3
on: There are only four skills: design, technical, management and physical
I really like this article, and it feels very empowering and exciting. It’s spawned a ton of thoughts, hence leaving two comments. It’s also quite relevant to me—I’ve recently been asking myself, based on a job offer I’m expecting, how hard would it be for me to become an expert in compute governance? And I actually did think before reading this article that six months of dedicated full-time work could get me there, but I was thinking in a narrower domain than “all technical skills”.

I’m very intrigued about the idea of there being a finite number of domains that I could learn and then be able to train myself to do any one item from like 90% of all things—one of my not-so-secret dreams in life is to be the kind of person that refutes the Character Sheet Hypothesis, the idea that if you’re good at one thing (e.g, smart) you must be worse at some other thing. (e.g, strong, or charismatic)

That said, there are some skills you mentioned as both being in the design domain that feel very different. I’ve always thought of myself as inherently not good at several of those design-shaped skills—but I can write and design software, so according to this model, not only is this false, I already have a lot of the necessary foundations.

So I feel like I can write decently well and architect a software program, but I imagine myself totally bouncing off interior design / UI designs—it’s not clear to me how to get started or how the skills transfer. I don’t consider my writing to use the same skills that I imagine interior design / UI needs—this combination of visual creativity + aesthetic maturity to generate designs and the aesthetic taste to judge them. I envision my writing and system design as much more like a technical problem—what are the requirements I need from this piece of work and how can I create a system that achieves them.

It might be the case that I could treat internal design this same way, but I do not know how. If I were one of your employees, and you knew I could write well and perform systems architecture, so I had this design capacity, but you now needed me to design rooms at Lightcone, how would you tell me to utilise my existing design skills to more swiftly learn this new one?

Jay Bailey 18 Apr 2026 13:41 UTC
2 points
0
in reply to: abstractapplic’s comment on: abstractapplic’s Shortform
I’m not sure if it’s because the follow-up question is doing a lot of heavy lifting or not. But I suspect it may not be the major contributor for the 50%+ probability items, since it did say <15% for long conversations that really were very un-eval-like, like my iterative process of “I want to build this tracking app, and I want it to suit my UX taste, and so I’ll just keep telling Claude to tweak anything I don’t like”.

But, you know, very small sample size, thrown together in an hour or two, LLM’s are weird, insert additional caveats here.

Jay Bailey 18 Apr 2026 13:29 UTC
9 points
0
in reply to: abstractapplic’s comment on: abstractapplic’s Shortform
This was a bit too long for a comment (1600 words, but lots of AI generated replies for obvious reasons) so it’s in a Google Doc: https://docs.google.com/document/d/1pXiVpBz727Ok4Z_TD8XEFViZjKd9g28qWBdwLCET40o/edit?usp=sharing

AI generated text is in italics throughout.

Jay Bailey 13 Apr 2026 13:08 UTC
3 points
0
in reply to: Alexander Gietelink Oldenziel’s comment on: Jay Bailey’s Shortform
Thanks for reminding me to crosspost this to the EA Forum as well, I had totally forgotten to do that.

Jay Bailey 13 Apr 2026 9:57 UTC
15 points
0
in reply to: StanislavKrym’s comment on: Jay Bailey’s Shortform
The specific company I was considering in this example was GDM, but I see Anthropic as broadly similar and OAI as somewhat worse. That’s why I didn’t name it—there was nothing in my logic that was specific to GDM.

If I put on my pessimist hat, here are the ways where this could increase extinction risk counterfactually:
- The work I do winds up being used for capabilities, and advances capabilities more than it advances safety.
- The work I do is largely safetywashing or focused on near-term risks that impact company reputation and/or profit, which has a much smaller or negligible effect on extinction risk. This isn’t strictly worse than doing nothing, but my choices aren’t “Do nothing” or “Join a frontier lab”.
- I end up getting my epistemics changed, falsely decide that alignment isn’t that hard actually, and then probably end up doing one of the other two points. (Especially re: Point 1, the harder you think alignment is, the less value you should put on frontier company work imo. I can go into more detail on why I think this, if it is non-obvious.)
By contrast, the counterfactual thing I could do is to potentially do other useful work that might reduce extinction risk more. My alternate path, which is the one I currently estimate I’m very likely to take (just waiting on security clearance) is to join the new Australian AISI and try to shape their research direction in a way that helps with extinction risk. The expected value of this is well above zero in my book, so it’s not just enough to avoid increasing extinction risk. I also have to decrease extinction risk more at a frontier lab than anywhere else I could otherwise be working.

And since that’s really hard to calculate, it’s easy to wind up doing the things that follow money and status gradients as long as they have a plausible path to impact. Committing to giving away any extra money is a good way of taking away one of those incentives. The money could do a lot of good if donated, but that’s an abstract good, it doesn’t emotionally move me in the way that a large salary clearly does. I don’t endorse this and I am not proud of this, but it would be far worse if I denied it to myself.

Jay Bailey 13 Apr 2026 7:23 UTC
174 points
56
on: Jay Bailey’s Shortform
Recently, I’ve been mulling over the question of whether it was a good idea or not to join a frontier AI company’s safety team for the purposes of reducing extinction risk. One of my big cons was something like:

Jay, you think the incentives are less likely to affect you compared to most people. But most AI safety people who join frontier labs probably think this. You will be affected as well.

So I decided on a partial mitigation strategy, entirely as a precautionary principle and not at all because I thought I needed to. I committed to myself and to several people I’m close to that if I were to join a frontier lab safety team, I would donate 100% of the surplus that I would gain as a result of taking that job instead of a less lucrative job somewhere else.

At this time I was applying for a few jobs, one of which was at a frontier company. Approximately immediately, my System 1 became way less interested in that job. And I didn’t even have an offer in hand for a specific amount of money. I don’t have good reasons to care a lot about getting more money for myself. I have enough already, and I voluntarily live well below my means. This did not stop the effect from existing, and I didn’t notice the effect before. I still don’t notice the effect on my thinking in a vacuum. I only notice it by doing a mental side-by-side comparison.

I now think anyone who is considering joining a frontier company in order to reduce extinction risk should make this same commitment as a basic defensive measure against perverted incentives. I am sure there exist people who are entirely indifferent to money in this way—this is at least partially a skill issue on my part. But it does seem that “Thinking you are indifferent to the money” is not a reliable signal that your thinking is unaltered by it.

This is also an opportunity to say that, if I ever do join a frontier safety team, I officially give you permission to ask me if I’m meeting this commitment of mine in conversation, even if other people are around.
What links here?
- Oliver Daniels's comment on Oliver Daniels-Koch’s Shortform by Oliver Daniels (30 Apr 2026 14:51 UTC; 12 points)

Jay Bailey 9 Apr 2026 6:20 UTC
3 points
0
in reply to: niknoble’s comment on: 11 pieces of advice for children
You’re mostly right about the “culture” interpretation. I was deliberately vague here since I’m not completely sure exactly what the broadest culture that uses terms like “agency” and “directionally correct” is. Is it rationalists? TPOT? Tech workers in the Bay Area? But in any case, I did mean some sort of niche culture that the median adult is not in.

I noticed two terms that a 12-year-old or median adult may be unfamiliar with (“directionally correct” and “agency”) which usually triggers my alarm bells—imo you can use one unfamiliar term in a sentence (as you just very thoroughly showed) but two is harder. (I also noticed “status quo”, but I think the median adult may already know what that is, so a smart 12-year-old could know that too) That said, it is true that a motivated 12-year-old or median adult could figure out how to parse that sentence, even if it wasn’t on the first reading. And “directionally correct” is probably easier to determine from context than “agency” is, so your example uses the harder of the two.

I also think I may have incorrectly specified what I meant, too. When I said: “I don’t imagine being able to understand that sentence when I was a kid”, I meant more “If I read that sentence, I wouldn’t know what it meant” as opposed to “I would be unable to understand that sentence even if I spent five minutes trying”. I recognise my phrasing seems to imply the latter more than the former.

Jay Bailey 8 Apr 2026 3:35 UTC
3 points
0
in reply to: niknoble’s comment on: 11 pieces of advice for children
While I agree with your second sentence, I don’t think that sentence is parsable by a median adult.

Jay Bailey 6 Apr 2026 6:28 UTC
14 points
9
on: 11 pieces of advice for children
I like these pieces of advice, but I feel like the article itself isn’t written for children, and is written for adults who already have some exposure to our culture. Here’s a sentence I really noticed:

“Age is just a number” is not true, but is directionally correct compared to the societal status quo that rids children of agency.

I was a pretty smart kid, as I imagine was most of LessWrong, but I don’t imagine being able to understand that sentence when I was a kid. I might say something like:

Don’t ever tell yourself “I can’t do something because I’m only twelve.” Instead, think of the obstacles you face being a kid in terms of things you can change, like “I don’t know enough maths to be a physicist” or “Adults won’t take me seriously enough to listen to my ideas”. These are obstacles you can work on or around, not just wait to grow out of.

OR

Don’t ever tell yourself “I can’t do something because I’m only twelve.”. Instead, ask what you think you could do at sixteen that you can’t do now, and figure out what you need to get there. Once you know what you need, there’s a good chance you can get there in less than four years, if you try.

I think this still gets the idea across, while using concepts a smart twelve-year-old probably knows.

Jay Bailey 2 Apr 2026 6:16 UTC
2 points
0
on: Have an Unreasonably Specific Story About The Future
But it does lead to another question—is this the best thing you can do? Could you take actions like consulting your local representative or widening your reach online to do better with your advocacy? That’s an entirely different question, and I’ll write about one model I use for it in my next post.
Here are the cliff notes of what I expect that post to look like, so people do still have a pointer to this idea when reading this post now.

My model is a “bullseye” theory of impact, like an archery ring, where you want to be close to the center. Each layer out is perhaps 5-10x less impactful than the next layer. For instance, say you wanted the US and China to sign a treaty on banning ASI development. The bullseye layer here would be literally being Donald Trump or Xi Jinping. The next layer out is being one of their closest advisors, or someone nominally in charge of this area. That’s still probably close to an order of magnitude less impact. Being a policy writer who is read by those advisors is another layer out from that, and so on.

To maximise your impact, it thus makes sense to try and move closer to the bullseye (through taking on closer positions or focusing your influence/advocacy inwards) or to take lots of shots on goal. BlueDot Impact are doing this strategy—they have a theory of impact around stuff like “Teach civil servants about AI risk”. This is pretty low impact individually, but they plan to put thousands of people every year through these courses, so they can make up for this with volume.