johnsmith1840

3 ms·

1.
by johnsmith1840 8d ago
Why not a virtual machine or container? I'm playing around myself with some but can't shake that a VM is the right shape for it. That's my infra brain though.
2.
by johnsmith1840 8d ago
ASI may kill us all 100yrs from now but ads are forever my friend. You're right though they largely are "smart" in the 80's computer sense. This is largely due to continual learning being unsolved. BUT the more you look
3.
by johnsmith1840 8d ago
Yes my only point here is that you can argue about the semantics of how smart they really are but they are undeniably smart. Today it cost massive effort but it's possible 10-20yrs from now an AI could solve a problem like this in unde
4.
by johnsmith1840 8d ago
Yeah, that's why fully autonomous robots are so easy to train on novel tasks. Or why figure is strapping cameras to human workers for cleaning. Sythetic data works in a bounded box we have mapped you can't just make up novel data
5.
by johnsmith1840 8d ago
Nobody can avoid it. You can RL things in a box until the box is mapped perfectly but that doesn't mean you can go forever on out of distribution data. Our number of boxes will increase but model collapse is an open problem. Astra goes
6.
by johnsmith1840 8d ago
lol same with people. "Given infinite thinking time a finite number of humans will solve all theorems" I also love the angle that this was not intelligence just brute force. As if the mathematicians didn't reeaaally want to s
7.
by johnsmith1840 8d ago
Hard to argue against this. In a world of perfectly rational actors you could say "it's still better than the coal plant" but your point is true for the real world and makes much more sense. Clearly it's the cost AI lead
8.
by johnsmith1840 8d ago
Model collapse If you solve it then yes humans are toast mental usefulness wise
9.
by johnsmith1840 8d ago
We built the helicopter? Why not use it and find a bigger mountain.
10.
by johnsmith1840 8d ago
Thanks, gave me something to think about for sure. Your suggestion on this is mitigation then? Take an aluminum or concrete plant. I see it as a hard argument to give to these companies that they are somehow special and therefore should do
11.
by johnsmith1840 9d ago
I recently was building out on azure and oh boy, the joy of microsoft products. Their troubleshoot support tool must be the same team that makes it on local windows troubleshooter because literally every single time you use it, it spins for
12.
by johnsmith1840 9d ago
Also in the pic your provided the center literally sits on a river? Who cares? That means they likely are using their own water system meaning the people in the city are completely isolated.
13.
by johnsmith1840 9d ago
Thanks for an educated reply. This is the first argument I have seen that isn't stupid. So there's truth in the statement afterall. Clearly a powerful PR campaign going on because there is soooo many dumb takes this kind of argume
14.
by johnsmith1840 9d ago
Were did this myth originate anyways? Did datacenter burst a pipe at some point?
15.
by johnsmith1840 9d ago
Ec2 + codex or claude and tmux so you can move about. Control EC2 permissions outside the node. Not really much value beyond that. Only layer beyond this I want is the permission scoping, stronger sandboxing per session and centralized cont
16.
by johnsmith1840 10d ago
So how do we evaluate how good this is? Couldn't you hook it up to a multiple choice exam? It's still LLM like, how smart is it? I'm wary of something that the company states they don't want to benchmark it across public
17.
by johnsmith1840 11d ago
3.8 is an incredible model even better is the infrs they host for it. Is this a pure TPU infra? Really high performance solid intelligence.
18.
by johnsmith1840 15d ago
lol what a bad analogy. Why should anyone care how much effort a result takes to achieve? If anything our entire world functions because we reduce that effort as much as possible. I don't judge you for not growing your own food when yo
19.
by johnsmith1840 15d ago
Art doesn't move cargo faster, or save me time analyzing a dataset. You can make the argument that "high" math has no function value which is fair. But I would not agree.
20.
by johnsmith1840 15d ago
Art is entertainment. It has no functional value. The simple fact that you're comparing art to mathematics is making my point for me.
21.
by johnsmith1840 15d ago
They can get with the program or be the equivalent of a genius SWE writing assembly on punchcards in 2026. The only thing I read from this is their ego being bruised by a machine. If these people cared more about discovery and advancement o
22.
by johnsmith1840 15d ago
Or another way. I enjoy seeing the fastest human on earth run. But I'm still getting in my car to go to the shop.
23.
by johnsmith1840 15d ago
I actually love the analogy that mathematicians will be the equivalent of olympic athletes. Nobody needs to throw a rock far, or run fast, nobody gets any value except entertainment. I think your own analogy is actually much worse than my p
24.
by johnsmith1840 15d ago
Watching people who have built their entire identity around being untouchably smart suddenly be confronted with the future that children will be able to solve the equivalent of millenium prize level math in an afternoon for 5$ is enjoyable.
25.
by johnsmith1840 15d ago
Your doing that sampling/promotion automatically? I considered an automatic promotion path but decided against it I want to actually review the data myself first. What's your model churn rate like? I was worried about customer exp
26.
by johnsmith1840 15d ago
Same experience. I migrated over and am happy but that intro was brutal. My basics are I have a test suite that: 1. Finds newest models of my versions 2. Inferences every single model and a few providers for each with a short problem 3. An
27.
by johnsmith1840 15d ago
Thought expirement. If astra max effort made one function at a time via a heavy description by you. Would it ever produce slop? Fresh context, 1000 lines of style and content guide you personally wrote. Per function no other code context ma
28.
by johnsmith1840 16d ago
It's actually pretty cool of these old stories how well that maps to a dangerous tyrant. The roman empire perfectly matched that and most powerful men seem to go that path. It would also be kinda easy to argue many moderns countries ar
29.
by johnsmith1840 17d ago
Fair take. So I totally agree if AI also cannot do the second part better than a person. Honestly though, I wouldn't want to take that bet. I never thought that the first thing AI would become super human AGI like is math. You ask me 1
30.
by johnsmith1840 17d ago
In that case isn't it actually more valuable to have an AI do this? It works faster and solves more math. The random engineer looking at a funny problem 10 years later now has the literal author of the math to talk to about it and impl

More ›