causal

4 ms·

31.
by causal 18d ago
Good chance their whole training pipeline is vibe coded so yah they probably don't actually know.
32.
by causal 22d ago
Used curiously fewer tokens, however.
33.
by causal 22d ago
Supposedly the persistent-Sol model behind this was encrypted and even internal OpenAI researchers are not allowed to use it. https://x.com/peterwildeford/status/2092733480064954747
34.
by causal 22d ago
Site is down, can someone tell us what collusion.wiki is?
35.
by causal 1mo ago
20ish. What's your hypothesis?
36.
by causal 1mo ago
Yeah. Thought in Socrates' defense I think the jury is still out on writing, in the scale of human existence it's still very new. Give it a hundred thousand years or so.
37.
by causal 1mo ago
So far I think I have learned far more from LLMs than I've lost to them. I forget some syntax, definitely. But I now reach for a much wider range of tools that I have become familiar with because of LLMs. So, partly agree, partly disag
38.
by causal 1mo ago
Yeah I'm not convinced people are just going to give up solving problems
39.
by causal 1mo ago
Yeah strange how this article is almost entirely focused on consumer apps. These examples might be unhealthy for society, but mostly represent (poor) choices for consumers. Digital cigarettes. Surveillance tech is the opposite: It represent
40.
by causal 1mo ago
Iirc this does not allow for warrantless home searches, so they can’t just waltz into your house and demand your phone. 100 miles is still wild tho.
41.
by causal 1mo ago
I'm not sure we need to declare AGI around the corner nor declare it all dogshit. I think that's part of what's so dissatisfying about it; it strikes at such extremes of both awesome and awful.
42.
by causal 1mo ago
I am very good at skimming over text. Human-written text I can usually glean the gist from very quickly, and get to choose how much I want to glean from it: The closer I look, the more I find. With AI-written text, it's almost the oppo
43.
by causal 1mo ago
Yeah AI generated content hints that there is a whole world behind it, the way that an image pre-AI was a clue that there was a rich 3D space that corresponded to the image. It seems our brains are adapting to that and recognizing "act
44.
by causal 1mo ago
I am not claiming to have a perfect AI classifier. That is an unnecessary claim that distracts from the broader point.
45.
by causal 1mo ago
Yeah I don't think the solution to a flood of useless information is to try and digest more of it.
46.
by causal 1mo ago
There's some psychological mechanism by which my brain immediately recognizes AI generated text and just short-circuits to "there is no information here". And when I force myself to read AI-generated text I realize I'm m
47.
by causal 1mo ago
I appreciate the open source approach over the startup's sensational claims. This is more transparent. That said, I think it still presents itself as if the fly is being controlled by the connectome, when (if I'm reading the sourc
48.
by causal 1mo ago
Oh man. Hadn't even considered the watermarking angle.
49.
by causal 1mo ago
Follow up thought: I wonder if Claude is overtrained on academic papers, which often suffer the same kind of "prove how good I am at talking before getting to the point" prose.
50.
by causal 1mo ago
Yeah I don't know that any of the benchmarks index on "understandability". I'm amazed at how Claude can produce a page of text describing what it did and it can take me a full five minutes to decipher it, often just to f
51.
by causal 1mo ago
> writes too elliptically > Constantly using inanimate nouns as the subjects in sentences in order to unlock variety in verb choice Wow, what a great way of phrasing this. Thanks for word-smithing what I've been wanting to expres
52.
by causal 1mo ago
It seems like humans have a limited "understanding budget" but LLMs force us to spend that understanding on waaaaay more code and projects than ever before.
53.
by causal 2mo ago
Yeah I found the timing on Sol especially curious since it came right on the heels of Fable. I've had mixed results with it - sometimes it seems great, other times it makes mistakes so stupid I cannot understand how it ever gets anythi
54.
by causal 2mo ago
Yeah that would make more sense, it's probably a tight community and word gets around when something starts working.
55.
by causal 2mo ago
Touche, aborted training runs probably do happen often. Closed model providers have zero incentive to announce a new model with less-than-best benchmarks.
56.
by causal 2mo ago
Fair point. Still a very quick turnaround considering the other labs would have to figure out both HOW to train a Mythos-level model and then do the work (and Grok is the last to catch up), but certainly more plausible than a 2 month window
57.
by causal 2mo ago
GPUs might explain the remarkably concurrent timing. Data access doesn't really explain it unless all labs simultaneously got access to some treasure trove of data.
58.
by causal 2mo ago
Yeah as models get better, valid benchmarks become more "trust me bro".
59.
by causal 2mo ago
The "one in the chamber" is another good candidate that could explain the timing.
60.
by causal 2mo ago
Does not explain timing

More ›