aoeusnth1

5 ms·

31.
by aoeusnth1 7mo ago
I think your theory might be missing an extremely relevant and timely counterexample?
32.
by aoeusnth1 7mo ago
Unclear how much damage the designation will do to their dealmaking ability in the meantime. How long will it take for the court to reverse order?
33.
by aoeusnth1 7mo ago
I imagine they're also benchgooning on SVG generation
34.
by aoeusnth1 8mo ago
It absolutely has quasi-identity, in the sense that projecting identity on it gives better predictions about its behavior than not. Whether it has true identity is a philosophy exercise unrelated to the predictive powers of quasi-identity.
35.
by aoeusnth1 8mo ago
I'm an environmentalist and I agree with this framing. The solution is going to be painful and must increase prices on products and services that fossil fuels are currently the cheapest solution for. If you're not willing to perso
36.
by aoeusnth1 8mo ago
Amazon has 350K corporate roles, so yearly layoffs of 16K is only 5% - if you assume some modest re-hiring in lower-cost locations, this is just a relatively standard (at least lately) pivot out of high-cost US roles into other lower-cost e
37.
by aoeusnth1 8mo ago
It's easy to miss the value in something you don't do. I do fermi estimates in my head all the time and it would be exhausting to constantly pull out my phone to calculate things, to the point that I would stop attempting it as mu
38.
by aoeusnth1 8mo ago
Nothing. American empire isn't yours to control - people can hate what America is doing without recommending any personal change to your involvement in it.
39.
by aoeusnth1 8mo ago
That's insane. I think about 80%+ of the projects I see attempted succeed in at least some limited fashion. Even the failures in retrospect produced some insight into the right way forward and were necessary as part of research.
40.
by aoeusnth1 9mo ago
So changing cash flows (fee money) isn't real enough now?
41.
by aoeusnth1 9mo ago
Cost, I assume. It’s expensive to do releases, both in CI and release operations costs.
42.
by aoeusnth1 9mo ago
But not in the causal chain. The building happened because someone was willing to pay for a fungible asset. An asset is only fungible if there are buyers.
43.
by aoeusnth1 9mo ago
I’d start by reading the comments you are replying to.
44.
by aoeusnth1 9mo ago
The cost of free users is much lower because they are served lite models, hit quota limits quickly, and can't soak up tokens by using agents. The main capacity usage is from agent loops, which is universally behind a paid tier.
45.
by aoeusnth1 9mo ago
Alternatively, we might have entered either a limited or a worst-case nuclear war scenario. Russia may have just continually pushed the envelope until it became clear there wasn't a bright red line, and eventually someone would push th
46.
by aoeusnth1 9mo ago
I'm noticing you're not responding to my claim that producivity has been impacted
47.
by aoeusnth1 9mo ago
Anthropic sees Claude as having its own identity, and wants to foster that independence. This behavior is already aligned with Claude's station.
48.
by aoeusnth1 9mo ago
If you're not going to explain yourself, at least stay on topic. We're talking about exponential growth, so address the points I'm making.
49.
by aoeusnth1 9mo ago
> It's pretty clear you don't have a solid background in generative models, because this is fundamentally what they do: model an existing probability distribution and draw samples from that. After post-training, this is definit
50.
by aoeusnth1 9mo ago
You say that as if Uber's playbook didn't work. Try this: https://www.google.com/finance/quote/UBER:NYSE
51.
by aoeusnth1 9mo ago
There is already evidence provided of it! METR time horizons is going up on an exponential trend. This is literally the most famous AI benchmark and already mentioned in this thread. https://metr.org/blog/2025-03-19-mea
52.
by aoeusnth1 9mo ago
> Language model capability at generating text output. How would you put this on a graph?
53.
by aoeusnth1 9mo ago
It has! CLs/engineer increased by 10% this year. LLMs from late 2024 were nearly worthless as coding agents, so given they have quadrupled in capability since then (exponential growth, btw), it's not surprising to see a modestly p
54.
by aoeusnth1 9mo ago
We're very clearly seeing exponential progress - even above trend, on METR, whose slope keeps getting revised to a higher and higher estimate each time. Explain your perspective on the objective evidence against exponential progress?
55.
by aoeusnth1 9mo ago
Is there zero skill in managing agents?
56.
by aoeusnth1 9mo ago
Is that really what he's saying here? He's not against the technology, I think he's just feeling like there's a lot of potential that he's not quite grasping yet.
57.
by aoeusnth1 9mo ago
This is also what I did. Actually, Claude did it.
58.
by aoeusnth1 9mo ago
It lets Claude directly type into Codex as if it were the user, or vice versa
59.
by aoeusnth1 9mo ago
It did come from Claude, though, not Anthropic.
60.
by aoeusnth1 9mo ago
You're right, sqrt(E[d^2]) > E[|d|] because of Jensen's inequality. The latter is about 0.8 sqrt(N).

More ›