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

OpenClaw is a security nightmare dressed up as a daydream

composio.dev

Post reply on HN
Print view View original

311–320 of 323 posts

Re: OpenClaw is a security nightmare dressed up as a daydream

#311

Post by bitwize »

never_inline wrote: post #257
bitwize wrote: post #46

Earlier quoted context omitted.

Been a middle-class IT drone much of my adult life. This is not my dream. In fact I just realized that one reason I don't like AI dev tools is because they turn me into the kind of dickhead manager I despise: one who doesn't understand the code or the nature of the work involved, just gives orders on what needs to be built and complains when it doesn't work.

I fix it by micromanaging it. Which class, method, function, module - I dictate the low level structure and features. I dump my all my hard earned coding opinions in a profoundly crafted markdown file.

What helps me adjust is what I call Thanos-coding: trying to get AI to do the thing with contexts and agents and SDD and whateverthefuck, and if it gets tangled on its own shoelace say "Fine. I'll do it myself." Alternatively: making its mistakes disappear with a snap of my fingers and starting over.

Re: OpenClaw is a security nightmare dressed up as a daydream

#312

Post by yencabulator »

kelnos wrote: post #223
davidw wrote: post #130

Earlier quoted context omitted.

Booking a flight is the kind of thing I want to dedicate my full attention to. It's expensive, and the timing and details matter a lot. I'm happy for the voice assistant to add stuff to my grocery list, though. The consequences are not serious if it screws up a letter or something.

To be fair, you can cancel flight reservations for a full refund within 24 hours, so if the LLM gets it wrong, you're not on the hook for anything. But in general I do agree: flight bookings are something I want to do myself, because even I don't fully know my preferences when it comes to timing and price until I see what's available. And in general I don't find it all that difficult to do. A couple days ago I booked…

Even worse, I will only know if the booking is "right" or "wrong" if I look at the alternatives, to see that it really is the best flight.

The best value of "AI" there is making a shortlist of best alternatives, which is just a Google Flights UX improvement. It's not like your own little agent would have good enough access to the underlying data, gatekeeping that is worth money.

Re: OpenClaw is a security nightmare dressed up as a daydream

#313

Post by yencabulator »

danpalmer wrote: post #184
ericd wrote: post #173

Earlier quoted context omitted.

Apparently I'm the only one here who finds it to be one of the worst things I ever have to do, I hate managing the combinatorial tab explosion by hand. Compounded by the adversarial nature of the price-setting algorithms that jack up the price on you if you show too much interest by researching too intensively. Just booked a flight for our family in two parts, and booking for one set of us made the price for the seco…

Do you think an agent is going to do all of that and get you the best price/time/comfort combination for your exact preferences. Or do you think it's going to pick the first that looks reasonable? Or do you think it's going to sacrifice one dimension too much? We already have agents for this if you really want to avoid it, they're called travel agents. They're pretty good at complex travel booking and not very expens…

Also, which one is easier to adversarially fool into bad decisions, an LLM or you?

Re: OpenClaw is a security nightmare dressed up as a daydream

#314

Post by justinhj »

habinero wrote: post #124
justinhj wrote: post #68

As a site for people curious about technology, where is the sense of adventure? People are inventing the future of human/ai interaction themselves because big tech could not do it within their own constraints. Don't get me wrong, those constraints are there for a reason, but the hacker mentality seems muted lately.

Hacker mentality means doing something new and clever, not reinventing IFTTT and related clones.

What? OpenClaw and the like are almost the polar opposite of IFTTT.

Re: OpenClaw is a security nightmare dressed up as a daydream

#315

Post by nickthegreek »

rvz wrote: post #299
nickthegreek wrote: post #109

Earlier quoted context omitted.

$10/month minimax using m2.7 and openai-codex oauth $20/month will allow you to mess around with this stuff for negligible cost.

But to do what? Other than being a hosting provider, how is using openclaw going to give someone a meaningful ROI?

The investment for me was an extra $10/month to mess around with some fun cutting edge stuff. This is interesting software tech that I mess with as a hobby. My ROI is understanding, exploring and entertainment. But it’s also doing some monitoring and logging for me on some various stuff that I wanted to keep an eye on with my network and actually proving some real value to my life. It’s genuinely exciting to mess with if you have the time.

Re: OpenClaw is a security nightmare dressed up as a daydream

#316

Post by philipallstar »

scuff3d wrote: post #271
philipallstar wrote: post #245

Earlier quoted context omitted.

We have governments to stop the obvious ones like tobacco too.

What point do you think you're making?

Almost every life lost is either directly or indirectly the fault of government. Why would corporate overlords be the more likely people to assume will be directing your dystopian future fiction when governments exist today?

Re: OpenClaw is a security nightmare dressed up as a daydream

#317

Post by mbesto »

threethirtytwo wrote: post #307
mbesto wrote: post #280

Earlier quoted context omitted.

> You need to think in terms of a probability of a successful hallucination or prompt injection. I would venture to say that an ACID compliant deterministic database has a 99.999999999999999999% chance of retrieving the correct information when asked by the correct SQL statement. An LLM on the other hand is more like 90%. LLMs by their innate code instruction are meant to hallucinate. I don't necessarily disagree wit…

Eh if you hire a programmer to program things for you, you won’t get a 99.9999999999%. Getting LLMs to have a reliability rate that is on par or superior to human performance is very very achievable.

> Getting LLMs to have a reliability rate that is on par or superior to human performance is very very achievable.

Source?

Re: OpenClaw is a security nightmare dressed up as a daydream

#318

Post by scuff3d »

philipallstar wrote: post #316
scuff3d wrote: post #271

Earlier quoted context omitted.

What point do you think you're making?

Almost every life lost is either directly or indirectly the fault of government. Why would corporate overlords be the more likely people to assume will be directing your dystopian future fiction when governments exist today?

... Because this thread is about a project that's a security nightmare, bought by a massive corporation currently ignoring the hugely problematic ethical issues surrounding their products. Corporations which btw, have more money and more power then most of the governments on the planet. Governments do terrible shit too, but it's less relevant to this conversation.

But I'm more interested in this framing. Are you saying that tobacco companies are somehow less responsible for their actions because the government didn't stop them from killing their customers? If the government just didn't exist, and tobacco companies could do whatever they wanted, do you think there would be less deaths from cigarettes?

Re: OpenClaw is a security nightmare dressed up as a daydream

#320

Post by salawat »

johnisgood wrote: post #150
stbtrax wrote: post #118

Earlier quoted context omitted.

the whole obsequious nature of how LLMs also amp them up thinking they're onto something incredible is throwing gas on this dumpster fire. "What a great idea! This will revolutionize linkedin commenting. Let's implement it together."

Oh for sure. When I present something to the LLM it always tells me how great it is until I make it "question" it, then it says it was overestimating this or that. Eh. Quite annoying.

You have to remember that LLM's don't have any persistent capacity to hold a "judgement". You ask for something, it provides an attempt at a completion for it. No fact checking, no reasoning, just a plausible looking output, tuned to hopefully get you to repeat the interaction.

Half the reason the dominant UX is a "Chat" is that's the only way to provide a facsimile of memory or persistence across requests. Append the last few turns, press go. Over time you can develop an eye for the model's tics/attractor topics.

Remember that they bill by token use, and suddenly, the entire UX/architecture starts making sense.

Post reply on HN