All those things you can easily measure in isolation, and you can also optimize for them quite locally. But these local optimizations do not produce global optimums, and the fewer of us are looking at the output, the less it matters.
Managing the transition to a world with abundant and powerful AI to optimize for safety and benefits to people should be one of the highest priorities in the world.
If every uh software engineer knew to watch out for false serial dependency chains, things where they were creating series of dependent operations that could not be optimized away or other sorts of architectural problems like that that cannot be easily fixed, then the world would look more like just wait and optimize the hotspot, right?
if you treat computing as this onion where you just keep peeling back the layers, there's always something interesting behind the scenes. And the more of those layers that you can peel back and understand, I think in some ways the better you can optimize for that world.
Now maybe this isn't completely out of the question but this would be quite challenging uh to do and that's because the calculus is a little bit different. We're not just running compute to optimize for a use case. We're actually running the robot in the real world and using the hardware and attempting the task in the real world.
Like I think we've seen internally at Meta um cases where if you can develop the right agentic loop and you have the right eval or the right metric for the agents to optimize, you can have a swarm of agents accomplish more than like a team of 100 engineers in, you know, uh very very handily actually, very very easily.
The evidence would show that and it makes perfect sense that AI and automation is creating jobs everywhere. The narrative about AI destroying jobs is exactly backwards. AI eliminate tasks. AI automates tasks away. But it doesn't necessary doesn't necessarily eliminate jobs.
PE funds optimize for a 4-7 flip and typically resort to aggressive cost-cutting and near-term optimization at the cost of the "soul" of the business or what made it successful in the first place.
This gives us a new perspective on why micronations like Sealand don’t work: they start backwards, from the territory and the government, rather than working forwards from a people and their culture.
this is a period in which backwards is forwards, which is to say this is not a period that like us thinks of the future as where potential is and humanity might get better and better over time. The potential of humanity is recapturing Rome.
And our harness wasn't very good for like the first like five months of open code. But it was good enough. It was good enough that most people couldn't really tell a difference. And once we won enough share, then we went back and like tried to make our harness like good and smart and optimize and all those things. But uh it was inverted from what everybody else was doing. Everybody else was being like you have to build the smartest harness and that's how you win.
OpenAI and Anthropic have both realized that code is the most important thing to optimize the models for, cuz that's where the money is. Like coders will spend $200 a month on a plan if it's good enough, it turns out.
this is a very deep truth about habits, which is a habit must be established before it can be improved. You know, it has to become the standard in your life before you can scale it up and optimize and turning it into something more. You need to standardize before you optimize.
And so I would say, yes, we are just giving proxy names to things we don’t understand, but to dismiss that as some kind of, “Oh, they just don’t know…” It is actually quite the opposite.
which is number one, you can measure product market fit. Number two, you can optimize product market fit. Number three, you can systematically even numerically increase product market fit. And number four, you can even have an algorithm write your road map for you.
Growth team can optimize. Growth can maybe lift it by 10, 15% maybe that's enough for you even that like is on the upper end of what growth team would be able to do if there is a slow down trajectory
If you have a growth model that works for you, that's wonderful. Good for you. Optimize it, grow it, scale it, create a team that will be nurturing it and that will be amplifying it. But you're going to need to evolve it. And that evolution needs to come through overlaying other growth models on top of it.
I think that that story about the latter part is completely wrong, almost 180 degrees wrong. I think that Einstein understood quantum mechanics as well as anyone, at least up through the 1930s. I think that his philosophical objections to it are correct. He should actually have been taken much more seriously about that.
I think this is one of the biggest things that physics can help with, and it’s an obvious kind of low-hanging fruit situation where the heat generation, the inefficiency, the waste of existing high-level computers is nowhere near the efficiency of our brains. It’s hilariously worse, and we haven’t tried to optimize that hard on that frontier.
I expect that the delta between 5 and 4 will be the same as between 4 and 3 and I think it is our job to live a few years in the future and remember that the tools we have now are going to kind of suck looking backwards at them and that’s how we make sure the future is better.
What what I find generally is if I don't understand something, I I try to read backwards and and and kind of the further back back I read, the more universal things I find. And the more things that I think are kind of remarkable and unique to our era actually turn out to be quite universal.
We as a profession misunderstand and misuse the concept of backwards compatibility, both upstream and downstream, by focusing on narrow legalistic definitions instead of outcomes.
We provide comprehensive empirical evidence showing that these residual networks are easier to optimize, and can gain accuracy from considerably increased depth.
We provide comprehensive empirical evidence showing that these residual networks are easier to optimize, and can gain accuracy from considerably increased depth.
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.