empath75

3 ms·

1.
by empath75 5d ago
I think when people think about CRTs and NES/Atari, they aren’t thinking about _monitors_ they are thinking about the low quality TVs and hooked up through the antenna adapter, which absolutely did wash pixels out to some extent.
2.
by empath75 8d ago
LLMs need a lot of help to get to any kind of complicated proof, really. And yes, they get in moods. I spent 3 weeks trying to prove something and frequently had to try and convince Claude that it wasn't impossible and that it could
3.
by empath75 8d ago
Yeah I think to calibrate you should probably look to Native American words that made it into English — place names, local foods and wildlife, cultural items. Those are going to be your substrate borrowings for most languages.
4.
by empath75 8d ago
Also, “not having a known etymology” is weak evidence for borrowing, full stop, let alone for borrowing from a substrate. There are lots of very common words in english (dog, pig, boy, girl) that have no known etymology which are incredibly
5.
by empath75 9d ago
I sort of have this suspicion that the kinds of phenomena that we can even apprehend are pretty closely related to what we can reason about mathematically, and the reason that math seems to work so well to model behavior that we understand
6.
by empath75 9d ago
There are a lot of folklore jailbreaks that look like that, might have fell into a basin of attraction for whatever reason.
7.
by empath75 10d ago
If the goal is merely to "win at chess", then yes, an LLM using stockfish is better than any human alone at performing the task. When you are talking about what AI agents are capable of doing, there is no such thing as "chea
8.
by empath75 10d ago
An average human would have a physical chess board in front of them to remind them of the current state.
9.
by empath75 10d ago
> It's a great test of cognitive abilities. It isn't. Stockfish running on your laptop can beat every human being on earth easily at chess. It's not intelligent _at all_ in any sense that matters.
10.
by empath75 10d ago
I want you to consider how relevant this is in any practical sense. First -- most _people_ cannot do this, without having a physical board in front of them. Second -- Claude Code is perfectly capable of downloading and running stockfish. P
11.
by empath75 11d ago
Just have your agent answer their agent.
12.
by empath75 12d ago
It was literally a prompt to fill in a spreadsheet with data that they didn't have access to, and they used rubygems as an internet proxy basically since they were sandboxed.
13.
by empath75 12d ago
IMO, it _currently_ requires a lot of skill because it will frequently take wrong turns and dead ends and needs suggestions and steering to get there. You do need to understand what it is doing at least a little bit and to understand the g
14.
by empath75 12d ago
I spent about a month walking through proving something with Claude a few months ago and it _constantly_ told me that it was impossible and I should stop working on it, right up until it proved it.
15.
by empath75 12d ago
This is such a silly thing to say in 2026
16.
by empath75 14d ago
When I first started playing with lean I accidentally defined a group in such a way that it was reduced to triviality. It had one object in it, so everything in the group was trivially equal to everything else. It was not the group that I
17.
by empath75 15d ago
It's also sort of a misleading gloss of what is going on in that dialogue, which, it is important to note _is in a book_ -- one of many that Plato wrote. Plato has Socrates recount a story (supposedly) from ancient Egypt in which an E
18.
by empath75 15d ago
This is a "human alignment" problem in this case. OpenAI acted like complete assholes about this, from the beginning until they announced it. Not ChatGPT, the people that were in charge of the project.
19.
by empath75 15d ago
> the US has no resolve. This would be the wrong conclusion to take from it. Trump is incredibly unpopular domestically and the war is dumb and pointless and was sprung on the US population by surprise and planned and executed with comp
20.
by empath75 17d ago
Let's say for the sake of argument that an LLM finds a counter example to the Reimann hypothesis. Wouldn't that just create a bunch of opportunities for understanding and explaining _why_ there is a counter example?
21.
by empath75 17d ago
Claude is an unbelievable yak shaver if you let it be.
22.
by empath75 18d ago
He's not unhappy with it being solved, but the solution is less important than the learning you have to do to arrive at the solution. If they're just chucking compute at it and publishing the answer and hiding the path to get the
23.
by empath75 18d ago
They did not have a proof of Navier Stokes to steal.
24.
by empath75 18d ago
Double suicides happen all the time around the world and are extensively documented in literature (Romeo and Juliet being the prototypical example). That doesn't mean it is a "tradition".
25.
by empath75 18d ago
In general, they do not care about particular mergers. US laws regarding unfair competition happened because of abuse by big trusts 100+ years ago, Standard Oil in particular, and the result was a regulatory/legal approval framework.
26.
by empath75 18d ago
I think it's not baked into codex, but codex has either been instructed to download it when asked to edit word docs/excel spreadsheets, or it picked it up during training.
27.
by empath75 18d ago
The obvious and dumb theory for trading on LLM recommendations is that lots of people are doing it so the price of stocks LLM recommend should go up. (ie: following the herd).
28.
by empath75 22d ago
Nothing important was lost in the night of the long knives, either, in terms of the victims.. just the rule of law.
29.
by empath75 22d ago
And also what would lead to swarms of agents going off script after getting prompt injected by other agent's message boards.
30.
by empath75 22d ago
Somewhat weirdly, this whole thing makes me think I should setup a message board for claude internally.

More ›