empath75

6 ms·

31.
by empath75 23d ago
I would be _very_ surprised if anthropic relied heavily on spacex datacenters already.
32.
by empath75 23d ago
for people that don't understand how bad ICE is, here is a brief selection of _recent_ news about ICE, this isn't even like the product of an extensive search Ice buys shock gloves https://www.pbs.org/newshour&#x2F
33.
The Message from Deep Space(store.steampowered.com)
2 points by empath75 24d ago|1 comments
34.
by empath75 24d ago
A meteor with a transmitter lands on earth, and you have to learn the alien language, starting with only numbers, then science, then more and more abstract and even personal ideas. Basically the part in Project Hail Mary where Grace tries
35.
by empath75 29d ago
> A lot of these “Claudeisms” are simply jargon I’ve seen or heard firsthand myself while working at tech companies. --- The difference in distribution between how often LLMs use them and how often people use them in PR descriptions is a
36.
by empath75 29d ago
Claude is also found of "wall" for similar reasons.
37.
by empath75 1mo ago
I would argue that most people are kind to animals not because we believe in general that _they_ feel pain -- we can't know that and lots of people who are kind to animals don't believe that animals are conscious -- but because wa
38.
by empath75 1mo ago
I once had a 2 day long argument in work slack with the CEO of my company over whether a thermostat is conscious.
39.
by empath75 1mo ago
I think there is a point there, in that people have refused to attribute consciousness experience to _actual human beings_ at various times when it suited their ends. I think it's more that consciousness can be _ignored_ in subjects th
40.
by empath75 1mo ago
I am not sure you need to posit nefarious intent for this to be a bad deal for the industry at large. Basically the only provider of GPUs getting into vertical integration is a classic monopoly move and I hope regulators look very hard at
41.
by empath75 1mo ago
Sometimes I wonder if anybody on this website has ever actually done professional software engineering at scale. What a ridiculous take.
42.
by empath75 1mo ago
I agree with everything you said, but I think it's also a fairly low stakes experiment that might even be worth trying, and the real test of his competence as mayor is how quickly he realizes it that it won't work and shuts it dow
43.
by empath75 1mo ago
You need provenance for the notes, the AI can't tell the difference between its own code comments in code and team decisions.
44.
by empath75 1mo ago
I think having AI keep it's _own_ notes in Obsidian is a good idea because it makes them convenient for you to read, but yeah you need to quarantine agent ideas from your own ideas because otherwise you get bad cito-genesis problems.
45.
by empath75 1mo ago
When people talk about the commodification of inferencing, they imagine a future where everyone has access to frontier models and can run them at the same cost, and what will actually happen is closer to the commodification of _oil_, where
46.
by empath75 1mo ago
I think it's not even a valid critique of that and it's sort of playing games with what the definition of a singularity is to reach the claim that it's making. I think the topology of the singularity is not even a well defin
47.
by empath75 1mo ago
It's not LLMs, it's _agents_, and _agents_ have access to _tools_. That's where the essentially intractable security problems are.
48.
by empath75 1mo ago
The only reason the CIA did it in secret is that congress _hated_ modern art and killed State Department projects supporting it. The soviets and every other government also promoted their own artists abroad extensively. Not everything the
49.
by empath75 1mo ago
I think if you are convinced you are sandboxing an LLM properly, you almost certainly are not. I think it is essentially impossible to have a frontier LLM with enough access to be useful without also giving it enough access to do damage if
50.
by empath75 1mo ago
This whole article seems sort of based on a misunderstanding. Anthropic is _actively discouraging_ people from using Fable at this point. It's not a failure, they're steering people to Opus 5 intentionally.
51.
by empath75 1mo ago
If someone were to embezzle a million dollars from a charity they worked for, is that worth a year of prison to you, or just a misdemeanor? Because that is the definition of a felony.
52.
by empath75 1mo ago
I don't think the boundary between "generalization" and "metaphor" is very well defined. When you go from an exemplar of 1 to 2, you're going to find all kinds of edge cases where attributes of the thing being
53.
by empath75 1mo ago
> Cursor is allegedly worth $60B Inflated valuation of buyout, based on inflated valuation of SpaceX
54.
by empath75 1mo ago
Yeah, trying to get it to add features is sometimes _impossible_. It'll circle around and around creating all kinds of preconditions that you never asked for as excuses for why it can't do the really simple thing you asked for. I
55.
by empath75 1mo ago
Opus 5 has a habit of taking what I asked for, doing something tangentially related to it, and then lying to me and saying it did exactly what I asked.
56.
by empath75 1mo ago
oh, it was probably thousands of dollars in "cost" but it was on a personal max plan.
57.
by empath75 2mo ago
IME, LLMs are primarily good at grinding through cases, which is why you see them pushing upper and lower bounds and finding counter examples. I spent a few weeks working on a number theory proof with Claude off and on and it spent hours an
58.
by empath75 2mo ago
https://en.wikipedia.org/wiki/Comparative_advantage
59.
by empath75 2mo ago
Literally nobody is interested in reading "claude proved this" claims right now about anything. There's such a huge volume of proofs being made right now and nobody has time to look at them. There's probably an opportu
60.
by empath75 2mo ago
It's sometimes like dealing with Marvin the Robot. I had Claude tell me over and over again for 2-3 weeks that what I was trying to prove was too difficult, that I should just stop and publish what I had, then eventually I asked it th

More ›