Garrett Baker

Karma: 6,227

I have signed no contracts or agreements whose existence I cannot mention.

They thought they found in numbers, more than in fire, earth, or water, many resemblances to things which are and become; thus such and such an attribute of numbers is justice, another is soul and mind, another is opportunity, and so on; and again they saw in numbers the attributes and ratios of the musical scales. Since, then, all other things seemed in their whole nature to be assimilated to numbers, while numbers seemed to be the first things in the whole of nature, they supposed the elements of numbers to be the elements of all things, and the whole heaven to be a musical scale and a number.

Metaph. A. 5, 985 b 27–986 a 2.

Garrett Baker 14 Dec 2025 19:46 UTC
11 points
0
in reply to: niplav’s comment on: niplav’s Shortform
None of the evidence or framing here is particularly worse than what I would expect from a biased against LW search & retelling of the history, and that is what you’re getting if you read Bouchard’s writing, so this seems like effectively zero evidence.

It is also notable that they have nothing good to say about LessWrong’s response. I think at minimum you have to credit the trading gains many LessWrongers made at the time, and MicroCovid.

That is not to say they are wrong about how early LessWrong was, just that it provides extremely little evidence. Better to read DirectedEvolution’s review of Seeing the Smoke.

Garrett Baker 11 Dec 2025 0:20 UTC
2 points
−6
in reply to: Screwtape’s comment on: Screwtape’s Shortform
My guess is the aversion is real, and maybe somewhat cynically I see the rejection of phone calls and emails as motivated mostly by rebellion/fashion cycles rather than anything else. Phone calls and emails just aren’t cool anymore. Your parents used them, yuck! And all the boring, very uncool workplaces and companies require them. Very similar to how gen-z dislikes Facebook in favor of Instagram. Facebook is where the boomers are.

Garrett Baker 10 Dec 2025 23:29 UTC
5 points
0
in reply to: Chris Wintergreen’s comment on: Eliezer’s Unteachable Methods of Sanity
The book in which Alex Wellerstein really makes the case was also released yesterday, buy it here!

Garrett Baker 10 Dec 2025 23:27 UTC
17 points
2
in reply to: Eliezer Yudkowsky’s comment on: Eliezer’s Unteachable Methods of Sanity
There is also recent debate about whether Truman was even well informed about the fact that Hiroshima was a city rather than a “purely military target”, eg see the book The Most Awful Responsibility, well reviewed by many including Richard Rhodes, as well as the excellent interview with the author by Dan Carlin.

Garrett Baker 10 Dec 2025 1:38 UTC
2 points
0
in reply to: faul_sname’s comment on: faul_sname’s Shortform
There is a dis-analogy, in the former case you have a single goal, get good at chess. In the latter case there are many goals we want AIs to do, ranging from coding to running scientific experiments to curing diseases and even making art. Obviously if you want a generalist you will want to teach general skills.

Secondly, a big reason labs are focusing on ML research is to get on the recursive-self-improvement super-exponential curve.

Your analogy addresses neither of these points, and I do think that these points are the primary reasons why people are trying to get AIs to do well at ML research. Therefore I think your analogy is bad and you should not make inferences or plans using this logic.

Garrett Baker 9 Dec 2025 5:01 UTC
65 points
41
on: D0TheMath’s Shortform
I feel like there was a time just after ChatGPT became big where there were so many different new & competing frameworks for what exactly was going on, shard theory and simulators in particular but also active inference the MIRI views and all the people rolling their own ontologies with which to understand LLMs, and notably trying to make those ontologies explicit and comparing them against each other.

Maybe I’m just around very different people now or doing very different work than I was (I am on both counts), but those conversations aren’t really happening anymore. I don’t know whether its for good or ill all things considered, but I do get nostalgic for them sometimes.

Garrett Baker 5 Dec 2025 20:09 UTC
4 points
0
in reply to: Tomás B.’s comment on: Tomás B.’s Shortform

Interestingly, my tweets were (for whatever reason) much better than usual during this period.

Presumably because you were drunk and angry

Garrett Baker 3 Dec 2025 21:39 UTC
3 points
0
in reply to: TsviBT’s comment on: Buck’s Shortform

The progress I listed doesn’t seem like it’s going slower due to medical regulation.

I mean the basic research aspect sure (except for stem cells), but applications of each of the progress areas you listed basically involve either clinical applications or selling GMOs. Both of which have very bad regulatory bottlenecks, especially from a world-wide perspective.

There has been, as you mention, enormous progress in bio-tech and our broader understanding of biology in the past 50 years, but comparatively little application of that knowledge. This is not what you would expect if the science is “deep” but applications easy. How exactly does the progress you listed support this conclusion?

Garrett Baker 3 Dec 2025 21:02 UTC
2 points
0
in reply to: TsviBT’s comment on: Buck’s Shortform
My guess is that the big difference in the speed of biotech compared to early-20th-century-advancements is the relative conservatism of the medical field, and the money & time-consuming certifications you need to get before releasing anything to market. This, in my view, is much less a function of the science, and much more a function of the sociology around the science.

Garrett Baker 3 Dec 2025 4:07 UTC
5 points
0
in reply to: the gears to ascension’s comment on: oligo’s Shortform

so yup np-complete. are they halting oracles?

You may be interested in Scott Aaronson et al’s paper on the subject of computability theory of closed timelike curves

We ask, and answer, the question of what’s computable by Turing machines equipped with time travel into the past: that is, closed timelike curves or CTCs (with no bound on their size). We focus on a model for CTCs due to Deutsch, which imposes a probabilistic consistency condition to avoid grandfather paradoxes. Our main result is that computers with CTCs can solve exactly the problems that are Turing-reducible to the halting problem, and that this is true whether we consider classical or quantum computers. Previous work, by Aaronson and Watrous, studied CTC computers with a polynomial size restriction, and showed that they solve exactly the problems in PSPACE, again in both the classical and quantum cases.

Compared to the complexity setting, the main novelty of the computability setting is that not all CTCs have fixed-points, even probabilistically. Despite this, we show that the CTCs that do have fixed-points suffice to solve the halting problem, by considering fixed-point distributions involving infinite geometric series. The tricky part is to show that even quantum computers with CTCs can be simulated using a Halt oracle. For that, we need the Riesz representation theorem from functional analysis, among other tools.

We also study an alternative model of CTCs, due to Lloyd et al., which uses postselection to “simulate” a consistency condition, and which yields BPP^path in the classical case or PP in the quantum case when subject to a polynomial size restriction. With no size limit, we show that postselected CTCs yield only the computable languages if we impose a certain finiteness condition, or all languages nonadaptively reducible to the halting problem if we don’t.

Garrett Baker 2 Dec 2025 23:25 UTC
8 points
3
in reply to: Daniel Kokotajlo’s comment on: Buck’s Shortform

Re: Politicians: Andrew Yang isn’t a major politician I guess, but his main schtick was “AI is coming” basically right?

Not really, from my memory and checking wikipedia, his campaign was mainly focused on advocating for UBI, and used whatever arguments it could to defend that policy position, including but certainly not limited to an argument that automation was coming, but mainly for menial tasks like truck driving.

Garrett Baker 2 Dec 2025 8:06 UTC
2 points
0
in reply to: Erich_Grunewald’s comment on: Charlie Steiner’s Shortform
I noticed that many posters not doing Inkhaven were more vibrant, which I credit to the Inkhaven spirit, which also spawned Halfhaven.

Garrett Baker 1 Dec 2025 18:35 UTC
34 points
15
in reply to: Charlie Steiner’s comment on: Charlie Steiner’s Shortform
I strongly disagree, I think LessWrong has become a much more vibrant and active place since Inkhaven started. Recently the frontpage has felt more… I can’t think of a better word than “corporate”… than I’d like. Maybe what I mean is that the LessWrong posters have started catering more and more toward the lowest-common-LessWrong-denominator.
For example, here are the top posts from September 2025 (I think October had a reasonable amount of Inkhaven spirit, considering all the people doing Halfhaven)
Tomas is always nice and refreshing, but imo the rest of this is just really extremely uninteresting and uninspired (no offense to anyone involved, each post is on its own good I think, but collectively they’re not that interesting), and seem very much catering to the lowest-common-LessWrong-denominator.
Contrast this selection with the following
I have not read as many of these (because there were more overall posts and these are more recent), but collectively the range of topics is so much more interesting & broad, you still get some lowest-common-denominator catering, but collectively these posts are so much more inspired than they were just two months ago.
I really would very much enjoy a world where each person participating in Inkhaven continued posting every day even after Inkhaven.
Edit: I would also say that concretely this has changed how I use the website a whole bunch too. In September I think most of the interesting parts of LessWrong discussion was happening inside shortforms, but now I think most of the interesting discussion has moved to posts again, which I consider very healthy.
What links here?
- Toss a bitcoin to your Lightcone – LW + Lighthaven’s 2026 fundraiser by habryka (13 Dec 2025 19:32 UTC; 222 points)

Garrett Baker 25 Nov 2025 2:40 UTC
2 points
0
on: Why Reality Has A Well-Known Math Bias
I wrote a post in partial response to this one.

Garrett Baker 24 Nov 2025 0:30 UTC
3 points
0
on: I’ll be sad to lose the puzzles
I think from the outside view, people will adapt much easier than they expect. For the longest time, people were not only fine with a God being able but not choosing to solve all their problems (until they died of course, in which case all their problems would be solved), but derived literally all the meaning of their life out of that God existing and having a plan for them personally. So much so that many, even now, are scared to acknowledge that God’s non-existence out of fear of losing that meaning.

From this perspective, in what sense is an AI god different from… you know… the actual god people theorized to exist? For one, the AI god will likely be a lot nicer (if all goes to plan)!

Garrett Baker 24 Nov 2025 0:09 UTC
3 points
−1
in reply to: Taylor G. Lunt’s comment on: Literacy is Decreasing Among the Intellectual Class
I guess my point is that the fact the 2022 version sucks is predicted on my model from the fact that we just don’t use too much complicated etiquette anymore. The fact the sentences are hollow is a fact more about the subject being written, not the skill of the author.

Concretely, using the Wikipedia page for modern handshaking in the US I think gives better prose than the modern 2022 version of the etiquette guide

The handshake is commonly done upon meeting, greeting, parting, offering congratulations, expressing gratitude, or as a public sign of completing a business or diplomatic agreement. In sports, it is also done as a sign of good sportsmanship. Its purpose is to convey trust, respect, balance, and equality.[10] If it is done to form an agreement, the agreement is not official until the hands are parted.

[...]

In the United States, United Kingdom and Canada, a traditional handshake is firm, executed with the right hand, with good posture and eye contact. A handshake where both parties are standing up is deemed as good etiquette.

Garrett Baker 23 Nov 2025 19:00 UTC
2 points
−1
in reply to: Taylor G. Lunt’s comment on: Literacy is Decreasing Among the Intellectual Class
Yeah, I agree that the modern version of Ettiquite is worse, but that’s just because we don’t have very complicated ettiquite! They need to waste space here, because everyone already knows what a handshake is, and you need at most a one-sentence description. If you didn’t turn every sentence into a paragraph the book would quickly turn into a blog post.

The solution is not in fact to add more flowery prose or complicated sentences, its to write about something else.

I also don’t know whether those reading in 1922 would say the same thing. We read the 1922 version and think “Oh wow! So informative!”, but perhaps the girls forced to read the book at a 1922 finishing school were thinking the same thing, or maybe ettiquite in 1922 was just a lot more complicated than it is today (it is).

Perhaps, similarly to the hypothetical bored girl in finishing school, in 100 years the future will look at the modern version and think “So informative!”.

Garrett Baker 23 Nov 2025 5:26 UTC
13 points
−3
on: Literacy is Decreasing Among the Intellectual Class
The more recent versions here just seem like better writing to me honestly. Short simple sentences which communicate a single idea are good.

If that makes them easy to read for the less literate, so much the better; they’re even easier to read for the more literate, no?

Garrett Baker 23 Nov 2025 0:46 UTC
12 points
4
on: Easy vs Hard Emotional Vulnerability
I don’t think I understand the constraints here, taking the anger case as a concrete example. My initial thought is to 1) talk openly about the anger which is different from actually releasing the anger, then 2) find a safe way to release that anger which hopefully doesn’t harm the partner, generic examples would be a rage room, taking a defense class, hunting, maybe voice acting(?), or directing the anger toward productive ends, eg using it to be more assertive or motivated (eg by spite).

But these all seem like things you would’ve thought of, so what am I missing? Is the constraint that the temper must be expressed toward those the person cares about?

Garrett Baker 23 Nov 2025 0:24 UTC
2 points
0
in reply to: Dentosal’s comment on: Complaining about my inability to focus on uninteresting things
I mean it really just sounds like you have ADHD is what I mean.

But what’s the reason to be productive beyond my natural abilities?

You are complaining about not being productive enough, presumably you care about productivity, why care also about whether its “natural”. If you think you have depression instead of ADHD, then the reason is so that you feel better! If Nature decided to make you miserable or distractable, then imo Nature had it coming.

I fear I would just use it all to make even more money, which doesn’t matter to my wellbeing at all.

You can always stop (modulo slowly decreasing your dosage so as not to trigger withdrawal if necessary)? For depression, clearly if you treat that with medication, your wellbeing will just definitely go up, regardless of whether you use your extra motivation to make more money.

The admiration of others? That would be cheating.

Plenty of people take drugs to treat mental illness, that is not “cheating” any more than antibiotics, surgery, or any of the other modern miracles we’re able to use to better the human condition.

Self actualization? I don’t think depression will go away by just doing more stuff, the problem isn’t not doing enough, it’s not enjoying the results

Then try an anti-depressant. Note also that this is an incorrect theory of depression, depression does indeed get better by doing stuff, and it is a common bias that depressed people have to believe nothing they do will work. If you think you have depression rather than ADHD, then take some antidepressant or something, or actually try to treat what you think you have.

I suggest you stop trying to come up with silly excuses not to fix your problems. That is not to say that there aren’t better options than medication, but each of your excuses here either reveal an ignorance about what treatments of mental illnesses actually do, or boil down to “but here’s a big brain reason why if it worked that’d actually be bad, therefore I will continue to wallow in misery for my whole life”.