Live data from Hacker News

AI agent bankrupted their operator while trying to scan DN42

lantian.pub

291–300 of 575 posts

Re: AI agent bankrupted their operator while trying to scan DN42

#291
post #77

IMHO the overly-verbose default style of LLMs is the most annoying part of interacting with them, and I wish their masters would just tell them to be terse by default. Also, whatever happened to the word "its"?

It's tied to the design. With humans, you have a train of thought which you can choose to represent in various ways--or not reveal them at all. In contrast, LLMs are make-document-longer machines being run over and over on alternating revisions of the document. Insofar as one might try arguing they have a "train of thought", it's made of the words/tokens. Everything they (don't-)emit is partly for the benefit of the…

We already have that in the form of separate reasoning/thinking and speaking streams. Even with that it's awfully hard to get LLMs to keep it consistently concise. As soon as that context window starts growing it falls right back into verbosity without constant nudges back.

Re: AI agent bankrupted their operator while trying to scan DN42

#293
post #286

Earlier quoted context omitted.

If a child goes through the checkout at the grocery store with cash, can the parent march in and demand a refund because "he's underage so the contract is void"? A credit card was used. Why should aws care about the details? (Other than the potential for the card to be stolen ofc.)

> If a child goes through the checkout at the grocery store with cash, can the parent march in and demand a refund because "he's underage so the contract is void"? Depends on the jurisdiction, of course. But for example in German law, the contract is not void exactly because and only if it was about daily necessities of low value - the law does, in fact, care very literally and explicitly about those details. So it's…

Well fair enough, although I find that rather surprising. If I understand you correctly selling anything more expensive than cheap food to a child carries a high degree of risk in Germany.

Then again, maybe making it impossible for a child to pawn expensive items for cash isn't such a bad idea. At least there shouldn't be any loopholes given the way Germany went about it.

Re: AI agent bankrupted their operator while trying to scan DN42

#294

Earlier quoted context omitted.

Strangely enough, that's one of the big draws for me. I'm "on the spectrum" and often find face-to-face socialisation and making new contacts very draining. I tend to prefer systems to people - although as time went on, I realised one of the things I really enjoy about DN42 is making the human contacts! After getting started with the various "auto peering" systems, I've been making much more of an effort to find indi…

Thanks for sharing, your projects look really neat! Reading your page I realize I know very little about networking at that level of the stack. That might be a good thing to dig into as a way to work around my "AI dread" (or whatever we call the feeling of "what's the point working on that project when an LLM can make it faster" I've been feeling too much lately).

That was where I started, too. I was fine with VLANs, routing in general and so on from datacenter/DevOps/Sysadmin work, but BGP and how the wider internet fitted together beyond the basics was mostly beyond me.

DN42 is a great playground for this thing - as long as you're prepared to put the effort in, it's a very friendly and helpful community. It's fun to build things for the heck of it and there's a lot of weird and wonderful stuff being worked on there.

Re: AI agent bankrupted their operator while trying to scan DN42

#295
post #17

> JertLinc3522: the mistake was from AI agent not from Human, since it was the agent I should have refund That really makes me wonder: is it coming from A) a general sense of entitlement B) seeing the agent as a human-like and able to bear responsibility C) not understanding that the dn42 community (which they're directing the request to), AWS (which is sending the bill) and whatever LLM provider is behind their agen…

Agents are a product, and AI companies really paint their products as friendly, productive and innocuous tools.

Some could claim they deceive some users and the general public into thinking they always do best, are always right, help mankind and can never ever create consequences

It would be interesting to see how AI consulted the user before it ordered VMs n AWS, which is the point between which the user would face consequences

Cloud is also marketed as something cheap, and I can understand that teens and starters can't expect to be able to spend for 6000$ of stuff without the parents or the bank checking

Computer education should start with that, but it doesn't as Microsoft, Google and Amazon would most likely lose a large part of their market if general public and managers who never go beyond the hype knew how much it cost

Re: AI agent bankrupted their operator while trying to scan DN42

#296

Earlier quoted context omitted.

That's so neat that if you type your hacker news password it automatically comes out as stars! ******* More places should have this feature.

I just uses stars as my password, so that works everywhere for me. (For security, I won't let you know how many stars)

This is perfectly safe as long as you keep your username a secret.

Re: AI agent bankrupted their operator while trying to scan DN42

#297
post #287
post #7

I really wanted to dislike the anonymous operator for the careless project (and the hilarious pomposity of the IRC subagent it spawned). Then I imagined the real-but-unknowable chance it was all set up by some kid just getting into computers, just seeing what’s possible, getting excited by a much bigger world at reach — and remembered my own expensive mistakes with long-distance BBSes & the like. I sorta hope for tha…

A kid with $4k to burn on a credit card though? A lot of things would have had to go wrong for this to be a child

I routinely see “please refund this infrastructure bill I racked up unexpectedly, I used my dad’s card and he’s going to kill me” requests.

Re: AI agent bankrupted their operator while trying to scan DN42

#298
post #7

I really wanted to dislike the anonymous operator for the careless project (and the hilarious pomposity of the IRC subagent it spawned). Then I imagined the real-but-unknowable chance it was all set up by some kid just getting into computers, just seeing what’s possible, getting excited by a much bigger world at reach — and remembered my own expensive mistakes with long-distance BBSes & the like. I sorta hope for tha…

How did the theoretical child get hold of a credit card?

I’ve seen minors signing up for cloud services with their parents card.

Re: AI agent bankrupted their operator while trying to scan DN42

#300

Earlier quoted context omitted.

My parents let me fill my tank with gas. They wouldn't let me open an AWS account. Aside from that, if it is misuse of a parents card, then then answer is "chargeback."

Chargeback sounds like trying to defraud AWS. If the parent authorises the child to use their card, then the buck should stop with the parent. AWS has done nothing wrong in allowing an account to be opened with a valid card.

Some banks make chargebacks so easy that people just click the chargeback button without trying to reach out to the vendor. I see this a lot - I work for a “vendor”.
Post reply on HN