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

Don't be a meat proxy

gruhn.me

Post reply on HN
Print view View original

351–360 of 786 posts

Re: Don't be a meat proxy

#351

Post by datakan »

chii wrote: post #232
ffsm8 wrote: post #56

Earlier quoted context omitted.

At my dayjob there is a person spearheading ai across the enterprise. They generated lots of documentation across the whole stack and now makes all PO/BAs read it if it's correct. So not just 300 lines - he unironically generated thousands of lines of "documentation" and is now making hundreds of people review it for him Complete brainrot Au psychosis is getting seriously outrageous at this point Thankfully I'm a dev…

He's basically asking other people to do his job for him, because he knows that the AI doesn't truly do it well enough. I would say to him that if he wants me to do his job, he needs to pay me his salary.

My VP boss calls me 6 times a day to help him understand his copilot responses and asks why we aren't doing what copilot says we should do. I then have to spend multiple hours explaining why. His peers call him to understand their copilot responses which he then copilots and calls me to help interpret. On and on and on

Re: Don't be a meat proxy

#352

Post by DonsDiscountGas »

skinwill wrote: post #343

Woodworkers often will have a preference for their favorite tool. Once they become comfortable with that tool they will sometimes prefer to use it for things it wasn't designed. For me it was the band saw. When I needed to split a board, bandsaw. When I needed to cut a curve, bandsaw. Cut a horse figure? bandsaw! I once saw a woodworking television show that seemed to do everything on the table saw. Rip, make frames…

I understand your point but it's funny to use AI as an example since a big benefit is it's generality. It's mediocre at many things but the fact that you can just use one tool and get passable results across many domains makes it so useful.

Re: Don't be a meat proxy

#353

Post by titzer »

eddythompson80 wrote: post #8

I deal with this all day long at work and it’s exhausting. People almost acting like no one has thought of it “I asked Claude what happened, and it spit out this 300 line response. Can you read it for me and see if it’s right?” What kills me is you might expect this from a busy high level manager that doesn’t really understand the technical details and they just point the AI to an error they got. They don’t know how…

> How can someone serious even think that’s ok.

Because no one is serious anymore.

I hate to say it, but anyone this lazy is absolutely asking to be replaced by AI.

Re: Don't be a meat proxy

#355

Post by sethammons »

"Claud said" is a hedge. It is abdication of responsibility. That said, a meat-proxied answer should come with more credibility based on the proxy.

I'd put more weight into my developer proxying system information than my sale's intern. I am expecting the proxy to have given the response a sniff test

Re: Don't be a meat proxy

#356

Post by chicken-stew »

jpnc wrote: post #2

If you create a machine for laziness you're going to get lazy people. It's only going to get worse I'm afraid. Do you guys think we're going to see a de-evolution of human beings due to technology?

  Do you guys think we're going to see a de-evolution of human beings due to technology?
The gap between people who are willing to commit the hours to learn something and those who prefer the path of least resistance is getting bigger.

I’d like to add that for hardware development (in my experience) it’s been non-stop double checking because of hallucinations. To the point where I don’t use it for weeks.

Re: Don't be a meat proxy

#357

Post by layer8 »

horizion2025 wrote: post #321

I agree but to be fair, often people could have asked an AI to get their answer. Especially seniors are prone to this (thinking the AI could not possibly know this or other biases). The responder wrote they asked Claude so as to not take credit for the answer and also to indicate they did not vet it. Then the person asking can decide if they want to read the answer or not (often a 1 sec glean can reveal if it is prom…

In that case, it would be better to suggest to the asker that they can ask the AI. Otherwise why would they ever start? Always asking Slack is easier.

Re: Don't be a meat proxy

#358

Post by foft »

The question is what is reasonable. This is more how I work for code and I just tried for a document.

i) Prompt (Human input)

ii) Response (LLM)

iii) For (i=0;i!=n;++i) {

iv) Review/adjust (Human input)

v) Response (LLM(s) - different perspectives) }

vi) Rewrite parts that failed/I am not happy with (Human input).

vii) Optionally repeat from iii.

This tends to get me somewhere I want. For code at least. I made a document with it recently (fairly concise) and still worried I'd missed some AI slop in it. Though there is always human slop if I just made it by myself.

Re: Don't be a meat proxy

#359

Post by yourapostasy »

ffsm8 wrote: post #56
eddythompson80 wrote: post #8

I deal with this all day long at work and it’s exhausting. People almost acting like no one has thought of it “I asked Claude what happened, and it spit out this 300 line response. Can you read it for me and see if it’s right?” What kills me is you might expect this from a busy high level manager that doesn’t really understand the technical details and they just point the AI to an error they got. They don’t know how…

At my dayjob there is a person spearheading ai across the enterprise. They generated lots of documentation across the whole stack and now makes all PO/BAs read it if it's correct. So not just 300 lines - he unironically generated thousands of lines of "documentation" and is now making hundreds of people review it for him Complete brainrot Au psychosis is getting seriously outrageous at this point Thankfully I'm a dev…

When some people call this "Verification Debt" or something similar, that is setting the Overton Window upon this phenomena adversely upon the verifiers. I call it an application of Brandolini's Law:

"The amount of energy needed to refute bullshit is an order of magnitude bigger than that needed to produce it.""

Until you put in at least the same amount of energy required to verify the AI output as you are asking others to put in to verify it, the AI output you are wielding must be considered bullshit. Really, it should be based upon the verification requestor's historical trust score and how many human verifiers are looped in. A Verification Tax, so to speak:

The energy the creator must spend (Ec) is determined by a function of the number of reviewers requested (N) and the historical trust score of that creator's AI outputs (T).

Ec = N * (1 - T)

Re: Don't be a meat proxy

#360

Post by mr_toad »

mort96 wrote: post #158
eddythompson80 wrote: post #8

I deal with this all day long at work and it’s exhausting. People almost acting like no one has thought of it “I asked Claude what happened, and it spit out this 300 line response. Can you read it for me and see if it’s right?” What kills me is you might expect this from a busy high level manager that doesn’t really understand the technical details and they just point the AI to an error they got. They don’t know how…

I've literally seen this in HN responses as well, so many times. What the hell is the point of human interaction if the "human" just forwards a chat bot's response

> What the hell is the point of human interaction if the "human" just forwards a chat bot's response

Forwarding an AI response (or any canned response) sends the message that they don’t want to interact with you. That could be intentional.

Post reply on HN