krona

5 ms·

1.
by krona 3d ago
Well it's an open question. As is generally the case for dog owners whose dogs attack (sometimes kill) other people/animals. There would need to be a degree of negligence demonstrated (e.g. the dog was 'out of control' w
2.
by krona 3d ago
> Whether they have a soul or consciousness or feelings doesn't matter here It does when it comes to accountability for what the model does. If the model is nothing more than the sum of its training data and regime, then the company
3.
by krona 8d ago
Extreme temperature levels (>2.0) can push a GPT of its manifold, essentially producing predictions barely distinguishable from random noise (it flattens the probability distribution of the next token). In theory this could predict anyth
4.
by krona 20d ago
This is a self confession if I ever saw one.
5.
by krona 20d ago
Absolutely not. By most accounts they're terrible at fixing anything other than trivial bugs in complex codebases e.g. Linux kernel, but they're much better at finding them.
6.
by krona 20d ago
You're suggesting that LLMs get better at fixing bugs/vulnerabilities, but at the same time stop getting better at finding them? What if this difference is inherent and essential?
7.
by krona 20d ago
What contract? It's mostly corporate debt. Seems pretty plain and simple to me. The contracts I guess you're talking about are compute obligations which frontier labs have. I don't need to know what those obligations are to k
8.
by krona 20d ago
> The next problem is making the models efficient enough to run a profitable business. In that world, what does Oracle do with a bunch of data centres which 1. It needs to pay for and 2. nobody needs. This is what the article is about: b
9.
by krona 20d ago
GPUs depreciate at a far, far faster rate than houses do. And then they need to be replaced. Who's paying for that? OpenAI and Anthropic don't even have positive free cash flow.
10.
by krona 21d ago
billg: "kaczynski was right afterall"
11.
by krona 29d ago
> Their best bet is hardware Why would Nvidia continue to finance OpenAI to meet its procurement commitments if it became a direct competitor? Where is the money coming from?
12.
by krona 29d ago
It doesn't matter what the valuation is, they won't raise the money to meet their procurement commitments without circular financing. Why would a sovereign wealth fund step in to meet those commitments? The financials of OpenAI ma
13.
by krona 1mo ago
A similar metaphor (I think I came up with but can't remember) is the slop sandwich. You can deliver slop to me but only when wrapped in a thick layer of human intervention (which usually also means removing 50% of the slop)
14.
by krona 1mo ago
From my interpretation, in order to get the data you 'own' (it's not theirs to give away since they can't claim the copyright on it), you need to use their services. The agreement the user has with Antropic is for the se
15.
by krona 2mo ago
Larry is big news in Jerusalem, what do you expect?
16.
by krona 2mo ago
That battle was fought in 2012-2015 and except for a few exceptions, IBM lost. The only real advantage IBM had was in mainframes with 1-10TB RAM. For example, customers who like doing database joins on a time-slice of countrywide telecoms d
17.
by krona 2mo ago
abstract inductive reasoning requires dynamic learning in unfamiliar contexts. All the benchmarks which measure this specifically (in particular, task composition) see LLMs fail catastrophically. There is quite a lot of research published o
18.
by krona 2mo ago
LLMs don't have the executive control to dynamically manage and compose mental sub-routines, when it has never seen those sub-routines in its training corpus. That requires fluid intelligence. ARAOC benchmark measures this specifically
19.
by krona 2mo ago
LLMs have almost no fluid intelligence or capacity for abstract inductive reasoning, relying on crystallized intelligence for basically everything.
20.
by krona 2mo ago
Interpolation vs extrapolation? It's very simple distinction and fundamental. Look at ARC-AGI-3 for empirical evidence, if that's important to you. My point however is purely an a priori one.
21.
by krona 2mo ago
> That's a common meme but it's the opposite of true. But you're restating what I just wrote - We're training a status quo machine and the probability of anything outside that distribution rapidly drops to zero.
22.
by krona 2mo ago
> so the proof to the unit-distance problem was on the manifold, given it was outputed by a LLM? By design this must be the case, even when you account for stochastic sampling i.e. 'temperature'. All it's outputs are a hig
23.
by krona 2mo ago
> They construct space between the points and extrapolate outside of the points. They don't. They interpolate between the points on a manifold.
24.
by krona 2mo ago
The output of a GPT is an interpolation (an estimation of new data points inside the range of known data) rather than extrapolation (estimations outside that range). 99% of the time we don't need a true intellectual breakthrough to get
25.
by krona 3mo ago
How do you know they're not also swallowing a portion of the supplier costs?
26.
by krona 3mo ago
Nazism is as eschatological as Marxism (class vs racial struggle) and many other secular/religious ideologies with utopian endpoints. Depending on your point of view there's no difference (because the end justifies any means), but
27.
by krona 3mo ago
> literal, unashamed Nazi He's a Christian Zionist which is the belief in the fulfilment of old testament biblical prophecy. This seems diametrically opposed to Nazism to me, ideologically. I don't know how you'd square th
28.
by krona 3mo ago
The typing was never the bottleneck.
29.
by krona 3mo ago
I'm actually talking about both. WSJ publishes Anthropic artificial profitability. Days later the reason for the profitability appears in SpaceX S-1; it's compute costs were artificially suppressed. Both are going public. It'
30.
by krona 3mo ago
They need to financially engineer a good looking quarter beforehand. Perhaps Larry Ellison can cut them a nice quid pro quo for a few months to make OpenAI look profitable (like the SpaceX/Anthropic deal), although that's probably

More ›