jacobgold

6 ms·

61.
by jacobgold 3mo ago
In terms of law enforcement, sure. But what we really need is competent white hat hackers doing battle with the black hats. Even if no one goes to jail, the NSA could make residential proxies much harder to operate in the US simply by detec
62.
by jacobgold 3mo ago
It may actually be costing them money through higher peak usage (requiring more capacity). I'm not sure but either way the government should have a role if market incentives aren't sufficient.
63.
by jacobgold 3mo ago
There's no spying required. The NSA and ISPs can find open proxies through infiltration and report them to (e.g. abuse@comcast.com), then Comcast simply has to act on it robustly. ISPs already deal with abuse reports like this, the sys
64.
by jacobgold 3mo ago
US residential proxies are the bane of the internet. They're the major source of social media manipulation and spam. Services can dramatically reduce abuse by blocking entire IP ranges based on country of origin, organization, or type
65.
by jacobgold 3mo ago
There's a new feature being added to AT to support "permissioned data" that would handle a Slack-like use-case: https://github.com/bluesky-social/proposals/tree/main/0016-p...
66.
by jacobgold 3mo ago
Slack exists largely because IRC was insufficient. It didn't natively support channel history, search, etc. For AI agents to flourish, Slack has to either truly open its network with a protocol or eventually be replaced. I'd like
67.
by jacobgold 3mo ago
Anyone is free to draw whatever analogies they want, but they either make sense or they don't. Outside of philosophical discussions or science fiction, comparing influencing humans to prompt injection is silly.
68.
by jacobgold 3mo ago
That's a fair point. I agree that the behavior can look similar even when the mechanism is different. But in practice, all the analogies I've seen are in fact superficial, including this one. The LLM that abruptly switches languag
69.
by jacobgold 3mo ago
In the past, when people built programs to extract text from PDFs, they didn't wrap them in a chat interface which claimed it had the human cognitive ability to "read" human languages. They could have done this. They could ha
70.
by jacobgold 3mo ago
What I'm pushing back on is, apparently, that some people genuinely believe these LLM-based computer programs are human-like intelligences. In reality, they're more like very good search engines that output relevant snippets of te
71.
by jacobgold 3mo ago
> So just like me, including the prompt injections if you count "nerd sniping" as such? Who would count that as prompt injection? It's a superficial analogy. If you were vulnerable to prompt injection, I could order you
72.
by jacobgold 3mo ago
Yes, people speak multiple languages and switch between them. But it's a superficial analogy to the behavior of LLMs which do something different and for different reasons.
73.
by jacobgold 3mo ago
https://en.wikipedia.org/wiki/Ad_hominem
74.
by jacobgold 3mo ago
This isn't even controversial. The proof is available to anyone who uses these systems: They hallucinate tool state, drift from the objective while seeming to comply, switch languages randomly (Cyrillic or Japanese characters in output
75.
by jacobgold 3mo ago
The idea that anyone would think anyone else would think that is fairly revealing.
76.
by jacobgold 3mo ago
Taking issue with using a computer to power the calculator? If so, you should know that all modern calculators are computers under the hood. Everything I've written about calculators applies to computers doing any kind of traditional d
77.
by jacobgold 3mo ago
> Yeah it's a trick question, the human error rate for it was about 30% (higher depending on the country). Error rate doesn't prove anything. The nature of the errors is what matters.
78.
by jacobgold 3mo ago
Sorry to tell a fellow Jacob that you're the one who is out of date. The kids are calling everything "AI" and they mean "AGI", and that's the complaint.
79.
by jacobgold 3mo ago
If an alien landed we could decide whether it seems to have a human-like intelligence or not. It could be incredibly intelligent but very non-human-like.
80.
by jacobgold 3mo ago
> If you're using the existence of flaws in LLMs to deny the claim of intelligence to them... That's not the point at all. It's the fact that they fail in ways completely unlike humans. You also have the burden of proof
81.
by jacobgold 3mo ago
> If someone unskilled at math brings a calculator to an international math competition, they will not succeed at solving many problems. > But if they bring a frontier LLM (and succeed at concealing it from the organizers), they ca
82.
by jacobgold 3mo ago
I get what you're saying but this is kind of a semantic game. These LLM models/agents absolutely do not reason in the sense that humans do, so you're quietly redefining the word. You can say of course decide to call them an &
83.
by jacobgold 3mo ago
Drawing the conclusion that "humans fail" and "models fail", so they must be similar, is very wrong. You could have humans calculate 2+2 all day and get a surprisingly high error rate. That reveals a flaw in how humans o
84.
by jacobgold 3mo ago
> Or you can keep calling them stochastic parrots as they solve decades-old open problems. I didn't use that phrase at all. But computers calculated digits of π to trillions of digits. With a chat interface for a Python math progr
85.
by jacobgold 3mo ago
We didn't have confused people claiming calculators were human-like intelligences doing math.
86.
by jacobgold 3mo ago
> There are so many other sites. So many others. Why are you here? I've been here since 2007 when HN launched. You're confused about my objection. I don't like the term "AI" but I love the technology as much as
87.
by jacobgold 3mo ago
We could've slapped a chat interface on calculators and called them "AI" because they can do superhuman math instantly. Most technical people would've thought that was stupid. LLMs are the same kind of category mistake.
88.
by jacobgold 3mo ago
It's one example that points out a major (possibly fundamental) flaw. I can point to prompt injection as another example. There are tons more if you're interested. Are you actually claiming LLMs operate based on human-like intelli
89.
by jacobgold 3mo ago
The models are continually patched with training and post-training. All you have to do is find an area they haven't patched yet, and they'll be just as stupid. I run into deep technical examples every day where they fail in the mo
90.
by jacobgold 3mo ago
You can call your little doggy "AI" if it makes you happy. But when you call something "AI" and it tells you to walk instead of drive to the car wash, you're not talking about the "AI" science fiction auth

More ›