lukev

8 ms·

61.
by lukev 8mo ago
Assuming that this was 100% agentic automation (which I do not think is the most likely scenario), it could plausibly arise if its system prompt (soul.md) contained explicit instructions to (1) make commits to open-source projects, (2) ma
62.
by lukev 8mo ago
Absolutely. It's technically possible that this was a fully autonomous agent (and if so, I would love to see that SOUL.md) but it doesn't pass the sniff test of how agents work (or don't work) in practice. I say this as som
63.
by lukev 8mo ago
This works for people. A LLM is stateless . Even if you believe that consciousness could somehow emerge during a forward pass, it would be a brief flicker lasting no longer than it takes to emit a single token.
64.
by lukev 8mo ago
I think there is some slippage on the word "technical" here, when you say there's no representation of it. She has clearly played a leadership role for a great number of technical products. What skillset would you be hoping f
65.
by lukev 8mo ago
I expect that the reason for their existence is political rather than financial (though I have no idea how that's structured.) It's a big deal that open-source capability is less than a year behind frontier models. And I'm
66.
by lukev 8mo ago
And it's pretty easy to run in a stronger sandbox too. "docker sandbox run claude" in a recent version of docker is a super easy way to get started.
67.
by lukev 8mo ago
If you're not vibecoding your own UX to render CC's output the way you like it, you're not living.
68.
by lukev 8mo ago
That may be a puff piece, but this comment reads like a slam? I don't know anything about her personally but the consensus among tech folks I follow and respect (and are politically aligned with...) are pretty universal in applauding t
69.
by lukev 8mo ago
It’s not just for purchasers… it’s to build consensus/approval around the concept of the US military-industrial complex.
70.
by lukev 8mo ago
Chomsky's claim is that linguistics is actually a branch of cognitive science, that language is, by definition, "what the brain does" and that "meaning" in language is grounded in the brain, by what the speaker as a
71.
by lukev 8mo ago
> I imagine there are multiple branches of philosophy, linguistics and cognitive sciences that studied this perspective in detail, but unfortunately I don't know what they are. You're looking at Structuralism. First articulate
72.
by lukev 8mo ago
This accusation would have a little more zing, if the person who owns the other place wasn't posting about white replacement almost every day. Pretty shitty that ones choice of social media is so politicized but if you must pick a si
73.
by lukev 8mo ago
Agreed, but that's trivial to fix. The conceptual problem is that there is a huge intersection between the set of "things the agent needs to be able to do in order to be useful" and "things that are potentially dangerous
74.
by lukev 8mo ago
Watching folks speed-run this whole thing is kind of funny from the outside. I wonder if anyone with a correct mental model of how LLM agents work (i.e, does not conceptualize them as intelligent entities) has actually granted them any perm
75.
by lukev 8mo ago
You’re right, and a big reason they won’t be able to capture the “full” value is because of competition, especially with open-source models. Sure, Claude will probably always be better… but better to the tune of $50k/seat? LLMs are, ul
76.
by lukev 9mo ago
> The skills that matter are the same ones that make someone a good manager of people. I disagree. Part of being a good manager of (junior) people is teaching them soft skills in addition to technical skills -- how to ask for help and do
77.
by lukev 9mo ago
I mean, that is the status quo of everything published anywhere on the internet today, using any protocol, and is why copyright exists. I'm not sure how ATProto changes that equation, aside from providing additional mechanisms for aut
78.
by lukev 9mo ago
God forbid that technologists are enthusiastic about a technology. Such things never happen, least of all here, on Hacker News.
79.
by lukev 9mo ago
What's the downside of it being replicated and distributed? It's got a digital signature, which can verify that it comes from you and is unaltered.
80.
by lukev 9mo ago
This is, like, the entire problem ATProto is designed to solve. Sure, Bluesky can block you from showing up in Bluesky the application. But if they do, you can still host your own content (as is being done here) or access it via alternative
81.
by lukev 9mo ago
A lot's changed since 2024. The decentralization primitives were always in place, but now they're being actively used. Hosting your own PDS is quite viable if you're tech-inclined, or check out Blacksky. There's also rum
82.
by lukev 9mo ago
Firstly, the idea that "1/3" of people went to truthsocial is laughable. Readers are free to create accounts on whichever platforms they choose, in order to follow you. And publishes are free to choose which platforms they en
83.
by lukev 9mo ago
Right. This whole process still appears to have a human as the ultimate outer loop. Still an interesting experiment to see how much of the tasks involved can be handled by an agent. But unless they've made a commitment not to prompt
84.
by lukev 9mo ago
This is a great framework to experiment with memory architectures. Everything the author says about memory management tracks with my intuition of how CC works, including my perception that it isn't very good at explicitly managing its
85.
by lukev 9mo ago
Right -- this seems more of a protective measure than something they will proactively enforce. If you have a well-behaved agent that uses a browser to buy on eBay, I doubt that will cause issues. But if it leads to issues, they can point to
86.
by lukev 9mo ago
What a horribly reductive question. I would hope we can engage these questions with a little subtlety, rather than just "picking a team."
87.
by lukev 9mo ago
Sure, but regardless of how it was generated, it's still garbage with respect to coherent propositional reasoning. It may indeed correspond to a desirable region in the latent space. My point is that it does not correspond to any k
88.
by lukev 9mo ago
I have complicated feelings about this kind of thing. On one hand -- prompts like this do change the latent space of the generation process, to get a different kind of output. If you like that output better, then it empirically "works&
89.
by lukev 9mo ago
They've definitely using tools like this for a while. It's been true under all administrations, and it's always been a problem. Privacy advocates have been alerting on this for a while. Physically attacking citizens takes it
90.
by lukev 9mo ago
Oh, trust me, I would absolutely love for a lot more of the US to be socialized (certainly healthcare, housing, and transportation.) Unfortunately that is virtually impossible in the current political climate, so I didn't include it.

More ›