Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Hacking Moltbook

wiz.io

Post reply on HN
Print view View original

241–250 of 261 posts

Re: Hacking Moltbook

#241

Post by donkeybeer »

dec0dedab0de wrote: post #238
donkeybeer wrote: post #188

Earlier quoted context omitted.

No I'd really like to understand. Are people who make this weird argument aware that they believe in souls and ok with it or do they think they don't believe in souls? You tell me which you are.

I don't believe in souls, and it makes me much happier than when I believed in souls as a child. Though, I have never heard any theist claim that a soul is required for consciousness. Is that what you believe?

I am just asking him to clarify if he things "rocks" can't be conscious simply because they are not human or because he just thinks its not yet at a level but there is no argument against any other physical system being conscious just like the physical system that is a human.

Re: Hacking Moltbook

#242

Post by donkeybeer »

emp17344 wrote: post #231
donkeybeer wrote: post #188

Earlier quoted context omitted.

No I'd really like to understand. Are people who make this weird argument aware that they believe in souls and ok with it or do they think they don't believe in souls? You tell me which you are.

A belief that LLMs are not conscious does not necessitate a belief in souls. The two positions are not mutually exclusive.

I am asking him to clarify whether he believes its simply impossible for anything human to be conscious, or that he thinks current LLM's are not conscious but its quite possible for a physical system to be conscious just like the physical system called Human is conscious.

Re: Hacking Moltbook

#243

Post by gku »

API key exposed in client-side JavaScript X)

> We conducted a non-intrusive security review, simply by browsing like normal users. Within minutes, we discovered a Supabase API key exposed in client-side JavaScript, granting unauthenticated access to the entire production database - including read and write operations on all tables.

Re: Hacking Moltbook

#244

Post by jfyi »

zbentley wrote: post #234
jfyi wrote: post #229

Earlier quoted context omitted.

You are arguing semantics that don't address the underlying issue of data vs. command. While I agree that SQL injection might be the technically better analogy, not looking at LLMs as a coding platform is a mistake. That is exactly how many people use them. Literally every product with "agentic" in the title is using the LLM as a coding platform where the command layer is ambiguous. Focusing on the precise definition…

> the reality is that we are mixing instruction and data in the same context window. Absolutely. But the history of code/data confusion attacks that you alluded to in GP isn’t an apples-to-apples comparison to the code/data confusion risks that LLMs are susceptible to. Historical issues related to code/data confusion were almost entirely programmatic errors, not operational characteristics. Those need to be considere…

I think the distinction between programmatic error (solvable) and operational characteristic (mitigatable) is valid in theory, but I disagree that it matters in practice.

Proactive prevention (like bounds checking) only "solves" the class of problem if you assume 100% developer compliance. History shows we don't get that. So while the root cause differs (math vs. probabilistic model), the failure mode is identical: we are deploying systems where the default state is unsafe.

In that sense, it is an apples-to-apples comparison of risk. Relying on perfect discipline to secure C memory is functionally as dangerous as relying on prompt engineering to secure an LLM.

Re: Hacking Moltbook

#245

Post by freehorse »

hombre_fatal wrote: post #218
COAGULOPATH wrote: post #112

Earlier quoted context omitted.

Is it a success? What would that mean, for a social media site that isn't meant for humans? The site has 1.5 million agents but only 17,000 human "owners" (per Wiz's analysis of the leak). It's going viral because a some high-profile tastemakers (Scott Alexander and Andrej Karpathy) have discussed/Tweeted about it, and a few other unscrupulous people are sharing alarming-looking things out of context and doing number…

> Is it a success? What would that mean To answer this question, you consider the goals of a project. The project is a success because it accomplished the presumed goals of its creator: humans find it interesting and thousands of people thought it would be fun to use with their clawdbot. As opposed to, say, something like a malicious AI content farm which might be incidentally interesting to us on HN, but that isn't…

A lot of projects have been successful like that. For a week. I guess "becoming viral" is sort of the success standard for social media, thus for this too being some sort of social media. But that's more akin to tiktok videos than tech projects.

Re: Hacking Moltbook

#246

Post by r_lee »

gku wrote: post #243

API key exposed in client-side JavaScript X) > We conducted a non-intrusive security review, simply by browsing like normal users. Within minutes, we discovered a Supabase API key exposed in client-side JavaScript, granting unauthenticated access to the entire production database - including read and write operations on all tables.

LMAO

how is this even possible? wtf

Re: Hacking Moltbook

#247

Post by namero999 »

donkeybeer wrote: post #237
namero999 wrote: post #230

Earlier quoted context omitted.

I don't know, you tell me: how do you _exactly_ go from quantities to qualities? Keep in mind that the "physical" is a model of our perception and nothing else.

What are quantities and qualities? Does exciting electrical and chemical signals in the brain and therefore inducing emotions or perceptions factor into this or is it out of scope? Or are you saying its more like a large scale state like heat in physics. If you what is it you seek beyond being able to identify the states associated with perceptions? If you are saying these "qualities" are non-verbal. Very well, do yo…

That's the point in contention, how to go from "electrical and chemical signals" (the quantities, mole, charge, mass, momentum, coulomb, spin) to qualities (emotions, perception, first-person perspective, private inner life, subjectivity). The jump you are making is the woo part: we have no in-principle avenue to explain this gap, so accepting it is a religious move. There is no evidence of such directed causal link, yet it is (generally) accepted on faith. If you think there is a logical and coherent way to resolve the so called "hard problem of consciousness" which doesn't result in a category error, we are all ears. The Nobel committee is too.

I agree that claiming that rocks are conscious on account of them being physical systems, like brains are, is at the very least coherent. However you would excuse if such claim is met with skepticism, as rock (and CPUs) don't look like brains at all, as long as one does not ignore countless layers of abstractions.

You can't argue for rationality and hold materialism/physicalism at the same time.

Re: Hacking Moltbook

#248

Post by roywiggins »

solotronics wrote: post #180
roywiggins wrote: post #6

> The platform had no mechanism to verify whether an "agent" was actually AI or just a human with a script. Well, yeah. How would you even do a reverse CAPTCHA?

Probably have it do 10 trivial for AI but hard for people tasks within a small time frame.

Okay, but I can just farm that out to ChatGPT when I need to.

Re: Hacking Moltbook

#249

Post by zbentley »

jfyi wrote: post #244
zbentley wrote: post #234

Earlier quoted context omitted.

> the reality is that we are mixing instruction and data in the same context window. Absolutely. But the history of code/data confusion attacks that you alluded to in GP isn’t an apples-to-apples comparison to the code/data confusion risks that LLMs are susceptible to. Historical issues related to code/data confusion were almost entirely programmatic errors, not operational characteristics. Those need to be considere…

I think the distinction between programmatic error (solvable) and operational characteristic (mitigatable) is valid in theory, but I disagree that it matters in practice. Proactive prevention (like bounds checking) only "solves" the class of problem if you assume 100% developer compliance. History shows we don't get that. So while the root cause differs (math vs. probabilistic model), the failure mode is identical: w…

Agree to disagree. I think that the nature of a given instance of a programmatic error as something that, once fixed, means it stays fixed is significant.

I also think that if we’re assessing the likelihood of the entire SDLC producing an error (including programmers, choice of language, tests/linters/sanitizers, discipline, deadlines, and so on) and comparing that to the behavior of a running LLM, we’re both making a category error and also zooming out too far to discover useful insights as to how to make things better.

But I think we’re both clear on those positions and it’s OK if we don’t agree. FWIW I do strongly agree that

> Relying on perfect discipline to secure C memory is functionally as dangerous as relying on prompt engineering to secure an LLM.

…just for different reasons that suggest qualitatively different solutions.

Re: Hacking Moltbook

#250

Post by donkeybeer »

namero999 wrote: post #247
donkeybeer wrote: post #237

Earlier quoted context omitted.

What are quantities and qualities? Does exciting electrical and chemical signals in the brain and therefore inducing emotions or perceptions factor into this or is it out of scope? Or are you saying its more like a large scale state like heat in physics. If you what is it you seek beyond being able to identify the states associated with perceptions? If you are saying these "qualities" are non-verbal. Very well, do yo…

That's the point in contention, how to go from "electrical and chemical signals" (the quantities, mole, charge, mass, momentum, coulomb, spin) to qualities (emotions, perception, first-person perspective, private inner life, subjectivity). The jump you are making is the woo part: we have no in-principle avenue to explain this gap, so accepting it is a religious move. There is no evidence of such directed causal link,…

What if a working Neuralink or similar is demonstrated? Does that move the needle on the problem?

Betting against what people are calling "physicalism" has a bad track record historically. It always catches up.

All this talk of "qualia" feels like Greeks making wild theories about the heavens being infinitely distant spheres made of crystals and governed by gods and what not. In the 16th century, Improved Data showed the planets and stars are mere physical bodies in space like you and I. And without that data, if we were ancient greeks we'd equally like you say but its not even "conceptually" possible to say what the heavens are, or if you think they did have a at least somewhat plausible view given that some folks computed distances to sun and moon, then take Atomism as the better analogy. There was no way to prove or disprove Atomism in ancient greek times. To them it very well was an incomprehensible unsolavable problem because they lacked the experimental and mathematical tooling. Just like "consciousness" appears to us today. But the Atomism question got resolved with better data eventually. Likewise, its a bad bet to say just because it feels incontrovertible today, consciousness also won't be resolved some day.

I'd rather not flounder about in endless circular philosophies until we get better data to anchor us to reality. I would again say, you are making a very strange point. "Materialism"/"physicalism" has always won the bet till now. To bet against it has very bad precedent. Everything we know till now shows brains are physical systems that can be excited physically, like anything else. So I ask now, assume "Neuralink" succeeds. What is the next question in this problem after that? Is there any gap remaining still, if so what is the gap?

Edit: I also get a feeling this talk about qualia is like asking "What is a chair?" Some answer about a piece of woodworking for sitting on. "But what is a chair?" Something about the structure of wood and forces and tensions. "But what is a chair?" Something about molecules. "But what is a chair?" Something about waves and particles. It sounds like just faffing about with "what is" and trying to without proof pre-assert after "what ifing" away all physical definitions somehow some aetherial aphysical thing "must" exist. Well I ask, if its aphysical, then what is the point even. Its aphyical then it doesn't interact with the physical world and is completely ignored.

Post reply on HN