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

AI coding is gambling

notes.visaint.space

Post reply on HN
Print view View original

451–460 of 471 posts

Re: AI coding is gambling

#451

Post by dolebirchwood »

duskdozer wrote: post #439
dolebirchwood wrote: post #192

Earlier quoted context omitted.

I have three side projects that revolve around taking public access data from shitty, barely usable local government websites, and then using that data to build more intuitive and useful UIs around them. They're portfolio pieces, but also a public service. I already know how to build all of these systems manually, but I have better things to do. So, hell yeah I'm just going to prompt my way to output. If the code wor…

>If the code works, I don't care how it was written This doesn't concern you for something you're presenting as a portfolio piece?

The code is still clean and organized. When I say I don't care "how" the code is written, I mean I don't care how it was generated (i.e., by prompting or by hand -- and I don't do any of it manually anymore). I still care about how it's structured, as a well-structured codebase helps the agents do their job better.

Re: AI coding is gambling

#452

Post by krupan »

lcampbell wrote: post #309
krupan wrote: post #158

Earlier quoted context omitted.

I ssk an AI to play hangman with me and looked at it's reasoning. It didn't just pick a secret word and play a straightforward game of hangman. It continually adjusted the secret word based on the letters I guessed, providing me the "perfect" game of hangman. Not too many of my guesses were "right" and not too many "wrong" and I after a little struggle and almost losing, I won in the end. It wasn't a real game of han…

The reasoning generally isn't kept in the context, so after choosing the secret word in the first reasoning block, the LLM will have completely forgotten it in the second and subsequent requests. So, it technically didn't change the secret word so much as it was trying to infer what its own secret word might have been, based on your guesses.

Thanks for the technical details, but like, it still sucks that I wasn't actually playing hangman. The BS machine did a good job of fooling me before I viewed the reasoning though

Re: AI coding is gambling

#453

Post by rcxdude »

the_af wrote: post #447
rcxdude wrote: post #443

Earlier quoted context omitted.

It is hard to read the comment in question as anything other than 'the struggle is important to your satisfaction', which to me feels at the very least not unversally true (different people will find satisfaction from different things: for me personally I have far more ideas than I could ever make real, so 'having to come up with one' is not really something I expect to be a problem), at worst a kind of religious 'vi…

I didn't read it like that. I read it like... say, asking a genie to produce stuff for you out of thin air. "Make the most beautiful gem for me!". Then, "a marvelous painting that will impress my beloved!", "a meal worthy of a king!", etc. After the novelty wears off, those are empty achievements. They are not yours, you haven't chipped a single facet of the gem, nor found it yourself, nor lifted a paintbrush or unde…

Phrases like 'you haven't earned it' imply some kind of judgement, IMO. You have the thing now, if all you wanted was the thing, not the lessons you learned along the way, then job done. The prediction that this will ring hollow, IMO, strongly depends on your personality. And LLMs are not like the genie, anyhow. You need to put in some effort yourself for anything past trivial examples.

(To project the argument to absurdity, it would be like saying you didn't bake a cake because you didn't grind the flour. And this isn't a new phenomenon: every time something is made easier and more accessible there are people claiming 'well you didn't _really_ do it unless you did it the old fashioned way', which lasts maybe a generation until the new way is just how you do it.)

Re: AI coding is gambling

#454

Post by the_af »

rcxdude wrote: post #453
the_af wrote: post #447

Earlier quoted context omitted.

I didn't read it like that. I read it like... say, asking a genie to produce stuff for you out of thin air. "Make the most beautiful gem for me!". Then, "a marvelous painting that will impress my beloved!", "a meal worthy of a king!", etc. After the novelty wears off, those are empty achievements. They are not yours, you haven't chipped a single facet of the gem, nor found it yourself, nor lifted a paintbrush or unde…

Phrases like 'you haven't earned it' imply some kind of judgement, IMO. You have the thing now, if all you wanted was the thing, not the lessons you learned along the way, then job done. The prediction that this will ring hollow, IMO, strongly depends on your personality. And LLMs are not like the genie, anyhow. You need to put in some effort yourself for anything past trivial examples. (To project the argument to ab…

> Phrases like 'you haven't earned it' imply some kind of judgement, IMO.

But those are my words, not the original commenter's. This is what the original said:

> Yet, you didn't go on the journey to get there.

This is the comment to which the reply was "you're overreacting". Yet there was no judgment, only understanding:

> I think the Slot Machine question is where a lot of early adopters are now at in this journey, and once more of us are there, then we can start asking the hard questions. Right now too many of us are where you're at, and it's impossible to know where things will end up in a year or so.

It's really, really hard to make the claim that it's judgmental and not understanding. "Too many of us", "we can start asking the hard questions", all of that points to a shared understanding of the situation.

> You have the thing now, if all you wanted was the thing, not the lessons you learned along the way, then job done.

Fair enough, but if you only wanted "the thing", then you're a consumer, not an author/creator/builder of the thing. It's the difference between buying (or commissioning) a painting or painting it yourself. If you just "want the thing" (the painting) commission it. But you're not a painter, you're a buyer of paintings.

> To project the argument to absurdity, it would be like saying you didn't bake a cake because you didn't grind the flour.

No, it would be like selecting a cake from an online store, selecting frosting, fillings, etc, and then upon having it delivered to you, claiming you baked it.

Re: AI coding is gambling

#455

Post by rcxdude »

the_af wrote: post #454
rcxdude wrote: post #453

Earlier quoted context omitted.

Phrases like 'you haven't earned it' imply some kind of judgement, IMO. You have the thing now, if all you wanted was the thing, not the lessons you learned along the way, then job done. The prediction that this will ring hollow, IMO, strongly depends on your personality. And LLMs are not like the genie, anyhow. You need to put in some effort yourself for anything past trivial examples. (To project the argument to ab…

> Phrases like 'you haven't earned it' imply some kind of judgement, IMO. But those are my words, not the original commenter's. This is what the original said: > Yet, you didn't go on the journey to get there. This is the comment to which the reply was "you're overreacting". Yet there was no judgment, only understanding: > I think the Slot Machine question is where a lot of early adopters are now at in this journey,…

'understanding' is orthogonal to 'judging'. OP may be judging themselves as well, but it's still a judgement. It reads like "I understand why you don't think this is bad', which is a statement that 100% implies the premise that this _is_ bad.

(and yeah, the consumer vs creator set of values seems to be a large part of the divide in the attitude to AI. But you must understand that a lot people got into creation because they wanted the thing, not because they wanted to be making the thing)

Re: AI coding is gambling

#456

Post by the_af »

rcxdude wrote: post #455
the_af wrote: post #454

Earlier quoted context omitted.

> Phrases like 'you haven't earned it' imply some kind of judgement, IMO. But those are my words, not the original commenter's. This is what the original said: > Yet, you didn't go on the journey to get there. This is the comment to which the reply was "you're overreacting". Yet there was no judgment, only understanding: > I think the Slot Machine question is where a lot of early adopters are now at in this journey,…

'understanding' is orthogonal to 'judging'. OP may be judging themselves as well, but it's still a judgement. It reads like "I understand why you don't think this is bad', which is a statement that 100% implies the premise that this _is_ bad. (and yeah, the consumer vs creator set of values seems to be a large part of the divide in the attitude to AI. But you must understand that a lot people got into creation becaus…

> 'understanding' is orthogonal to 'judging'

Sort of. "Understanding" in the sense I mean it has an empathetic inflection.

Anyway, I have nothing to gain by defending someone else's statement, so I'll let it rest. They can defend themselves if they feel like it.

> (and yeah, the consumer vs creator set of values seems to be a large part of the divide in the attitude to AI. But you must understand that a lot people got into creation because they wanted the thing, not because they wanted to be making the thing)

This is way more interesting to discuss and yes, I agree a lot of the divide happens there. In particular, I don't think you can be a programmer if you mostly prompt an AI. You're something else, but not a programmer. Does it matter? I don't know. "Programming" as an occupation doesn't have a fundamental right to exist; maybe it'll go the way of the Dodo. I care more about things like AI in art and human communication, in that case I do have a strong stance: the journey is as important as "the thing".

In general I think there's a drive in modern society (not all of it, but powerful parts of it) that wants to turn us into consumers of things. I'm pushing back against that.

Re: AI coding is gambling

#457

Post by Terr_ »

cyanydeez wrote: post #298
interestpiqued wrote: post #200

Earlier quoted context omitted.

I think AI literally makes even being wrong feel like getting something done. And that is the addictive part for people.

"Near-Miss" effect: https://harprehab.com/blogs/the-psychology-of-risk-why-gambl... I believe that's the strongest pattern in LLM gambling. Was listening the Syntax and they described that "Even though theLLM did it wrong 4 times, that 5th time could be right, so why not just go!"; paraphrased of course. It also explains the meta-LLM business, where all these CEO types put in some question and because the LLM just kn…

Well-put. My "sunk costs" will be redeemed if I just persevere slightly longer, so walking away would mean an unnecessary loss.

Re: AI coding is gambling

#458

Post by stitched2gethr »

raizer88 wrote: post #56
thisisbrians wrote: post #41

It is and will always be about: 1) properly defining the spec 2) ensuring the implementation satisfies said spec

AI: "Yes, the specs are perfectly clear and architectural standards are fully respected." [Imports the completely fabricated library docker_quantum_telepathy.js and calls the resolve_all_bugs_and_make_coffee() method, magically compiling the code on an unplugged Raspberry Pi] AI: "Done! The production deployment was successful, zero errors in the logs, and the app works flawlessly on the first try!"

We're way past that.

Re: AI coding is gambling

#459

Post by stitched2gethr »

SV_BubbleTime wrote: post #67
ambicapter wrote: post #54

Earlier quoted context omitted.

Then pulling the lever until it works! You can also code up a little helper to continuously pull the lever until it works!

We have a monkeys and typewriters thing for this already. Just instead of hitting keys, they’re hitting words, and the words have probability links to each other. Who the hell thinks this is ready to make important decisions?

It's not ready to make important decisions. But that's not the same as making important contributions.
Post reply on HN