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

I know you didn't write this

ammil.industries

Post reply on HN
Print view View original

121–130 of 145 posts

Re: I know you didn't write this

#121

Post by tediousgraffit1 »

GMoromisato wrote: post #84
tediousgraffit1 wrote: post #69

Nowhere in here does it indicate that the generated plan was wrong or broken. I dont care if you use ai to write. I care if you write well. If the author trusted the other person, then it shouldn't matter. If the author didn't trust the other person, then they'd have to validate their output anyway. Granted the tech allows people I dont trust to generate a lot more bs, a lot faster. But i just reject and move on with…

It's just that AI gives fools more power. It used to be that a well-written document was a proof-of-work that the author thought things through (or at least spent some time thinking about it). I'm all for AI--I use it all the time. But I think our current style of work needs to change to adapt to both the strengths and weaknesses of AI.

> It used to be that a well-written document was a proof-of-work that the author thought things through (or at least spent some time thinking about it).

I think you hit the nail on the head here. The problem isn't so much that people can do bad work faster than ever now, its that we can no longer rely on the same heuristics for quickly assessing a given piece of work. I dont have a great answer. But I do still think it has something to do with trust and how we build relationships with each other.

Re: I know you didn't write this

#122

Post by jiggawatts »

karaterobot wrote: post #47

I'm sometimes asked to produce meaningless 30-page documents that nobody ever reads. I mean literally nobody, since I can see the history of who has accessed it. Me and a proof-reader, and occasionally someone will open it up to check that it exists. But nobody reads them, let alone reads them closely . Not the distant funder who added it as a line-item requirement to their grant (their job is adding line items to gr…

This is the same argument as “why is software X so bloated when nobody uses more than 10% of the features?”

Because everyone uses a different 10%.

I write these documents too and I’ve watched people “read” them. They all do the same thing: flip to the conclusions and then if there is a need they will skim the section that’s relevant to their role.

The project manager cares only about the risks, costs, and time estimates.

The architect just wants to see the diagram and maybe check that the naming conventions have been followed.

Sysops just wants to know what they’re on the hook for after go-live.

None of them read the whole document, but the whole document ends up being read.

PS: I’ve found I have to take care of distributing documents myself. All organisations big and small are shockingly bad at disseminating information. Help them!

Re: I know you didn't write this

#123

Post by throw310822 »

jvanderbot wrote: post #18

If I discover you fed me AI output, directly from AI, it really makes me wonder what you are doing here. What did you add to this equation when I could have done it myself? At least a "Generated by AI, reviewed and edited by xyz" tag would be some indicator of effort and accountability. It may not be wrong to use AI to generate things whole cloth, but it definitely sidesteps something important and calls into questio…

Sorry, but if you don't trust your coworker to review and shape the AI's output, why would you trust them with actually writing the whole document?

And if you think that at this point you could have done it yourself, then why don't you? The only important thing is that the document is fine, if it takes you too much to verify it then you need to trust your colleague, that was their job.

Re: I know you didn't write this

#124

Post by etothepii »

I'm amazed by how few people appreciate the quote attributed to Abraham Lincoln, "If I had more time I'd write a shorter letter."

Why aren't people using LLM to shorten rather than lengthen their plans? You know what you meant so can validate whether the shorter version still hits the points you care about. Whereas if I use an LLM to shorten your email there is always a risk I've now missed your main point.

Cleaning up grammar, punctuation spelling etc is a good thing worth doing but adding padding is exclusively irritating.

Re: I know you didn't write this

#125

Post by EGreg »

zephen wrote: post #96
EGreg wrote: post #65

I think that we should have revision control for intermediate stages - for code, documents, even paintings. So we can at least have some idea of provenance, how it's made. Until AI is used to fake that, too.

There are some issues with this. 1) For things made with LLMs: 1a) The fact that older versions aren't online forever. You literally might never be able to put the original prompt in and get the same result. 1b) A minor change in input prompt can result in a huge output change, rendering the original prompt practically meaningless, especially if modifications were required for the output of the LLM. 2) For things mad…

I meant only for 2.

And I don’t care if it’s boring, it has to be available. Crime scene details or forgery details are mundane and boring too, but for the investigators they are essential.

Re: I know you didn't write this

#126

Post by SkyeCA »

GaryBluto wrote: post #3
messe wrote: post #2

> Suspicions aroused, I clicked on the “Document History” button in the top right and saw a clean history of empty document – and then wham – fully-formed plan, as if it had just spilled out of someone’s brain, straight onto the screen, ready to share. This isn't always a great indicator. I can't stand Google Docs as an interface to write with, so use VIM and the copy/paste the completed document into it.

It's bizarre to me that this didn't occur even slightly to the post author.

A world full of AI generated content and a world where we trust what we see seems to be mutually exclusive. I expect the default for a number of people going forward (myself included) will be extreme suspicion when presented with new images/videos/documents.

Re: I know you didn't write this

#127

Post by zephen »

EGreg wrote: post #125
zephen wrote: post #96

Earlier quoted context omitted.

There are some issues with this. 1) For things made with LLMs: 1a) The fact that older versions aren't online forever. You literally might never be able to put the original prompt in and get the same result. 1b) A minor change in input prompt can result in a huge output change, rendering the original prompt practically meaningless, especially if modifications were required for the output of the LLM. 2) For things mad…

I meant only for 2. And I don’t care if it’s boring, it has to be available. Crime scene details or forgery details are mundane and boring too, but for the investigators they are essential.

> And I don’t care if it’s boring, it has to be available.

Strong language, strong nope.

Demand to see shit I didn't even think was important when I was busy building stuff? Sucks to be you.

Re: I know you didn't write this

#128

Post by QuercusMax »

mlhpdx wrote: post #114
QuercusMax wrote: post #39

Earlier quoted context omitted.

Those two things aren't even comparable. Both of those are using technology to physically imprint letters onto the page, but in both cases those are still your own ideas in your own words.

But not the appearance. It’s not the same, but it rhymes. Edit: to clarify, people were judged by the clarity of their handwriting in the past and these tools made that impossible. Similarly, LLMs spackle over higher level language issues.

Dictation has existed for millennia; alternatively, hiring someone to neatly write out your letters after making a messy draft has also existed for a very long time. My mom paid half her way through college in the 60s by typing people's papers for them who didn't know how to type properly.

These things are not remotely comparable.

Re: I know you didn't write this

#130

Post by uhhhd »

Nowhere in the article does he explain why the use of AI is inherently problematic or why it necessitates rewriting the project plan. Work product is either good or bad, and responsibility for its quality rests with the person delivering it. The tools used to produce the work are irrelevant. In fact, for those who prioritize execution speed, the use of labor-saving tools should be encouraged wherever feasible.

This comment was generated by chatgpt (inspired by me).

Post reply on HN