The usual default math assumes debtors want to pay. But founder-controlled firms with a "you lent us too much" attitude introduce a strategic-default option that standard systems of ratings do not fully capture.
The winner would then be the AIs, not us. That is not to say that the China issue is not real, hence the export controls and attempted international agreement.
If this is only due to gains in capabilities, that is extremely bad news, and it means CoT monitoring is unlikely to survive for another year unless we find a way to actively improve it, and it might not last six months.
But it's not meaningfully possible to understand the world at this scale - macroregions are real but analysis of policy, including transportation but also the usual economic and social topics, has to be down at the level of sovereign states first with no illusion that a province in a large country is comparable to a smaller country.
The level of academics that those students know is very low because the parents have basically abandoned that and said, "I care all about the life skills." For parents who really care about academics, those kids crush it.
and I like the idea that since it was a process it looked kind of active and um and then it could be paired with a document that resembled it so our system was kind of a verb and a noun
it's it's not enough to be a good person. I believe that people who are kind and care about people can and and usually do do better than sociopaths in the same roles, but only if they're good at business.
It's like they have all the wisdom of their most senior engineers looking at every single diff and that is fantastic. Which means that you don't have to worry about remembering and looking and nitpicking and all the things that we're not good at anyway
I would say that I expect like full automation of AR&D perhaps somewhere around like 2031 2030 and then getting to like the like beats all humans on the job milestone. Maybe I expect median around 2033
I mean I think like if you look at uh my colleagues work on say alpha fold that was a very specific model for uh protein folding and it was highly successful um and was able to really handle that domain quite well so that all of a sudden you now have this amazing tool and model that can give you answers to questions about proteins and their structure um really effectively um but it's not a general model it's a very specific one and there are other I domains where that kind of approach can work really well. Uh maybe in material science or chip design or things like that that uh will enable you to leverage the capabilities of a very accurate but but niche model uh to do things that are hard today.
Yeah, I mean it's a little different, but I think uh you're going to see more and more uh uh high performance and um low energy uh inference hardware systems because I think everyone is now realizing that inference is the key to making you know these agent-based systems be available to more and more people and that latency is really important and that specialization of the hardware is a really key way you can make uh things that are more energy efficient and lower latency than more general purpose uh computational devices like say GPUs or TPUs
I think startups tend to win when um the sort of like technology landscape is moving very quickly. uh when costs are coming down when cycle times are short and all of those things are happening right now.
it it won't work to try to repatriate uh the the manufacturing that left the US. Um that's a uh you you don't you don't beat China at its own game. What you do is invent the future, invent the next wave of manufacturing uh and build that here.
If you had asked me this question in the lead up to my 2019 book, Ultralearning, I would have said no. I was a strong advocate for doing hard problem solving and the associated mental effort it requires as a way to learn difficult skills.
They just published a secondary analysis of the active trial that showed that um those who did processing speed training with uh booster sessions at one and three years had a significantly decreased risk of dementia 20 years later.
Now I just run everything off the XPS 14, connected to an Apple XDR 6K 32" (nobody has yet managed to beat this, and I've owned it for years). It's a great, simple setup.
But they just by virtue of having less people to coordinate they can often move faster and make um better decisions a little bit less design by committee.
Much of active investing is built around publicly available information and a belief in the power of mean reversion, not original research and seeking information.
Also, not only is it far cheaper to send things by sea, it is far more secure. If you're going to run the Belt Road thing, uh or it's it's the belt part of it, the land part, um you've got to control the whole thing end to end. You're going through some of the most unstable places there are.
Where I get fired up, and this ties back to the AI discussion, is when that's turned into this meme, that programmers no longer have to be competent. I mean the AI is gonna figure it out. The generators is gonna figure it out. I don't need to know SQL, active record is gonna abstract it away from me. No, no, no dude, hold up. The path here is competence.
Their words now
really good programmers are currently more valuable than ever because they're the ones who are able to get the most out of the AI acceleration.
Whenever humans experience situations where we are closer to our animal selves than usual (needing to use the bathroom, having periods, giving birth, raising a baby, being sick), we usually behave as animals do: instinctively, urgently, and impolitely.
a worry I have is that the growth rate could be like 50% in Silicon Valley and, you know, parts of the world that are kind of socially connected to Silicon Valley and, you know, not that much faster than its current pace elsewhere. And I think that'd be a pretty messed up world.
But you cannot outwork the person who's working on a better thing. And so um it is possible if you really grind maybe you can work 10% harder or 20% harder than you are now. But if you work on the right thing you get 100x of the results or thousandx the results. And so what you need I I almost feel like we should change the definition of what it means to work hard. It should not only mean putting in the hours. It should also mean outthinking the other person.
I do, but it's business as usual because we're we're in an intelligence explosion already and have been for decades. And when you look at GDP, it's basically the GDP curve that is an exponential weighted sum over so many aspects of the industry. Everything is gradually being automated has been for hundreds of years.
I saw some of the highest performers just being people that had very high agency, had that clock speed, had that energy. Um, yes, they they cared about the mission, but they didn't necessarily need to have deep experience on that matter. And in fact, sometimes that experience could be a crutch in certain ways. Especially in this world where the grounds are shifting so fast with AI, a lot of your like learn habits actually need to be intentionally discarded. You know, you need to have a beginner's mind on this type of stuff.
I think if you have especially a B2B feature where you may have some lock in reverse trials can be super powerful. You just want to get people in there. You don't need to ask for their credit card because they're using your CRM or they're investing quite a lot of time in like building out, you know, material and content. And so by the time that window drops, you actually like feel, oh man, I probably should keep this and and start paying. I think for a lot of consumer products, it's a little bit harder for that to work. And so I've typically seen more just normal free trials be be the norm.
So I don't think uh learning is really about training. I think learning is about about learning. It's about an active process. The child tries things and sees what happens.
Like people are people and robots are robots. Like the the better analogy for the robot, it's it's like your car or a bulldozer. Uh like uh it has much lower maintenance requirements. You can put them into all sorts of weird places and they don't have to look like people at all.
They're obviously both. A, because we've got things like the bite marks I just described, which is pretty much definitive scavenging, and then we've got the healed bite marks with T-rex teeth buried in bones, which is pretty much definitive active predation. So we've got evidence of it doing both.
You're dominant in contests, particularly with conspecifics. If you're trying to win a territory or win mating rights, bigger things usually beat up smaller things. So there's going to be selection favoring them. But then big things don't usually do well in extinction events, so that tends to reset the clock by killing off the big stuff, and then smaller stuff does better again.
But they they both had something something unique that we didn't know at the time. I'm going to call them concentrated deposits. Not uninsured cuz people are misstating that, concentrated.
it's a luxury good like you're not paying the penalty for that the people who embrace it and think that they can just ask AI for everything or they can just look everything up but they don't have to understand the copy pasta that they slam into their project they're the ones who going to suffer
Their words now
Where I get fired up, and this ties back to the AI discussion, is when that's turned into this meme, that programmers no longer have to be competent. I mean the AI is gonna figure it out. The generators is gonna figure it out. I don't need to know SQL, active record is gonna abstract it away from me. No, no, no dude, hold up. The path here is competence.
As I set out last time, the usual story — that this was caused by deforestation, making firewood so scarce that people resorted to burning an inferior but cheaper fuel — simply does not stack up.
Overall, though, this experience reinforced my belief that the tooling and interface design around LLMs is lagging way behind the actual capabilities, and is an area of active experimentation and development, even aside from any future model improvements.
Precisely how much the ancient continuity of East Asian states matters for development is still a little murky, and the delineation with colonial legacies is still an area of active debate—but the balance of evidence suggests that the legacies of these pre-colonial states stretched through the colonial period, extending even into the post-independence era.
If you buy a handful of index funds, sit on them for decades and never do another thing, you're likely to outperform nearly everyone who tries to beat the market by trading-including most professionals.
In the absence of immigrants, between 2020 and 2050 Spain’s labor force (its active population) will drop from 22.6 to 15.7 million people. You’re not going to fill all the jobs needed in a modern and sophisticated economy like that of Spain’s with 15.7 million people.
A lot of people think that to make deterrence work, you have to be able to beat the Chinese and therefore, you need a much bigger military. And I don't think over time that's possible. I think it's probably not even possible now to beat the Chinese in a war over Taiwan or in a war in the South China Sea.
It was surprisingly unliterary and unpostmodern for Mitchell, but in exchange, he nailed the historical details and gave us an adventure which subverted many of the usual tropes
How to improve your HUGE major life decisions, like whether/who to marry, whether to have kids, where to live, career paths, and such — where you can’t use the usual checklist/data approach. This economist addresses self-identity, “deepest self”, and “something I was meant to do”. You can’t use a pro/con checklist with an item that says “lose respect for myself”.
the electric technology used for cars only has to beat energy hogs that are below 10% in efficiency, our everyday car. But to qualify for aircraft use, the technology has to equal or beat what is used today, which is at the 30% to 40% efficiency level using modern turboprop or turbofan engines.
A fine intro to investing if you think you can reliably beat the market or that active trading is a good use of energy. If you're already bought in on index funds though, you won't find much here.
Adding a dependency is a serious decision which requires consensus within the team, an audit of the new dependency, an understanding of its health and long-term prospects, and an ongoing commitment to re-audit them and be prepared to change course as necessary.
Surprisingly, fuel consumption per passenger mile would be lower in a Celera 500 than in a 737. The larger the fuselage, the harder it is to achieve the same drag reduction.
here's some notable software that I use daily. I also maintain an active list here. Fish Shell. Micro text editor. Toot Mastodon client. cmus audio player.
here's some notable software that I use daily. I also maintain an active list here. Fish Shell. Micro text editor. Toot Mastodon client. cmus audio player.
Managers should support 6-8 engineers. This gives them enough time for active coaching, coordinating and furthering their team’s mission by writing strategies, leading change, and so on.
A korrent is a belief a person has stated in their own words: one
sentence stating the claim, backed by a quote and a source, kept at
korrents.com.
Under a name here, the quoted block is what they actually said.
The korrent beneath it is the claim those words support, in
korrents' wording — tap it to see the record, its source, and who
else holds it.
Nobody here wrote their own korrents. They are compiled from public
statements, and a person can change their mind, which is recorded too.
About the English under a post
Some people here publish in a language other than English. Where they
do, this site shows a machine translation beneath the post, in
this typeface — the site's own, not theirs.
The post itself is never changed, moved or hidden: what is set in the
serif above is exactly what the person published, and it is what to
quote them on. A translation can be wrong in ways that matter,
especially about tone.
Only the post's own words are translated. A quoted post, a linked
article and a belief on korrents.com
are left in their original language.