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

LLMs are a 400-year-long confidence trick

tomrenner.com

Post reply on HN
Print view View original

71–80 of 202 posts

Re: LLMs are a 400-year-long confidence trick

#71

Post by runarberg »

BoxOfRain wrote: post #39
latexr wrote: post #26

Earlier quoted context omitted.

Those aren’t mutually exclusive; something can be both useful and a con. When a con man sells you a cheap watch for an high price, what you get is still useful—a watch that tells the time—but you were also still conned, because what you paid for is not what was advertised. You overpaid because you were tricked about what you were buying. LLMs are useful for many things, but they’re also not nearly as beneficial and p…

I think the following things can both be true at the same time: * LLMs are a useful tool in a variety of circumstances. * Sam Altman is personally incentivised to spout a great deal of hyped-up rubbish about both what LLMs are capable of, and can be capable of.

These are not independent hypotheses. If (b) is true it decreases the possibility that (a) is true and vice versa.

The dependency here is that if Sam Altman is indeed a con man, it is reasonable to assume that he has in fact conned many people who then report an over inflated metric on the usefulness of the stuff they just bought (people don’t like to believe they were conned; cognitive dissonance).

In other words, if Sam Altman is indeed a con man, it is very likely that most metrics of the usefulness of his product is heavily biased.

Re: LLMs are a 400-year-long confidence trick

#72

Post by falloutx »

krystofee wrote: post #32

I disagree with the "confidence trick" framing completely. My belief in this tech isn't based on marketing hype or someone telling me it's good – it's based on cold reality of what I'm shipping daily. The productivity gains I'm seeing right now are unprecedented. Even a year ago this wouldn't have been possible, it really feels like an inflection point. I'm seeing legitimate 10x gains because I'm not writing code any…

Are you actually reading the code? I have noticed most of the gains go away when you are reading the code outputted by the machine. And sometimes I do have to fix it by hand and then the agent is like "Oh you changed that file, let me fix it"

Re: LLMs are a 400-year-long confidence trick

#73

Post by self_awareness »

techpression wrote: post #40
ltbarcly3 wrote: post #3

I think anyone who thinks that LLMs are not intelligent in any sense is simply living in denial. They might not be intelligent in the same way a human is intelligent, they might make mistakes a person wouldn't make, but that's not the question. Any standard of intelligence devised before LLMs is passed by LLMs relatively easily. They do things that 10 years ago people would have said are impossible for a computer to…

I did that when I was 14 because I had no other choice, damn you SoundBlaster! I didn't get any menu but I got sound in the end. I don't think conflating intelligence with "what a computer can do" makes much sense though. I can't calculate the X digit of PI in less than Z, I'm still intelligent (or I pretend to be). But the question is not about intelligence, it's a red herring, it's just about utility and they (LLM'…

set BLASTER=A220 I5 D1

Re: LLMs are a 400-year-long confidence trick

#74

Post by pousada »

latexr wrote: post #46
BoxOfRain wrote: post #39

Earlier quoted context omitted.

I think the following things can both be true at the same time: * LLMs are a useful tool in a variety of circumstances. * Sam Altman is personally incentivised to spout a great deal of hyped-up rubbish about both what LLMs are capable of, and can be capable of.

Yes, that’s the point I’m making. In the scenario you’re describing, that would make Sam Altman a con man. Alternatively, he could simply be delusional and/or stupid. But given his history of deceit with Loopt and Worldcoin, there is precedent for the former.

It would make every marketing department and basically every startup founder conmen too. While I don’t completely disagree with that framing it’s not really helpful.

Re: LLMs are a 400-year-long confidence trick

#75

Post by ACCount37 »

latexr wrote: post #51
ACCount37 wrote: post #38

Earlier quoted context omitted.

LLMs of today advance in incremental improvements. There is a finite amount of incremental improvements left between the performance of today's LLMs and the limits of human performance. This alone should give you second thoughts on "AI doomerism".

That is not necessarily true. That would be like arguing there is a finite number of improvements between the rockets of today and Star Trek ships. To get warp technology you can’t simply improve combustion engines, eventually you need to switch to something else. That could also apply to LLMs, that there would be a hard wall that the current approach can’t breach.

If that's the case, then, what's the wall?

The "walls" that stopped AI decades ago stand no more. NLP and CSR were thought to be the "final bosses" of AI by many - until they fell to LLMs. There's no replacement.

The closest thing to a "hard wall" LLMs have is probably online learning? And even that isn't really a hard wall. Because LLMs are good at in-context learning, which does many of the same things, and can do things like set up fine-tuning runs on themselves using CLI.

Re: LLMs are a 400-year-long confidence trick

#76

Post by HWR_14 »

Traubenfuchs wrote: post #64
HWR_14 wrote: post #58

Earlier quoted context omitted.

Bad example. IKEA assembles better with a manual screwdriver.

You wouldn't say that anymore if you would have ever assembled PAX doors.

Maybe? I'm not familiar with every ikea product. But it looks like it take a dozen small screws into soft wood.

Re: LLMs are a 400-year-long confidence trick

#77

Post by falloutx »

bodge5000 wrote: post #69
schnitzelstoat wrote: post #11

I agree that all the AI doomerism is silly (by which I mean those that are concerned about some Terminator-style machine uprising, the economic issues are quite real). But it's clear the LLM's have some real value, even if we always need a human-in-the-loop to prevent hallucinations it can still massively reduce the amount of human labour required for many tasks. NFT's felt like a con, and in retrospect were a con. T…

> The LLM's are clearly useful for many things I don't think that's of any doubt. Even beyond programming, imo especially beyond programming, there are a great many things they're useful for. The question is; is that worth the enormous cost of running them? NFT's were cheap enough to produce and that didn't really scale depending on the "quality" of the NFT. With an LLM, if you want to produce something at the same s…

All of the professions its trying to replace are very much bottom end of the tree, like programmers, designers, artists, support, lawyers etc. While you can easily already replace management and execs with it already and save 50% of the costs, but no one is talking about that.

At this point the "trick" is to scare white collar knowledge workers into submission with low pay and high workload with the assumption that AI can do some of the work.

And do you know a better way to increase your output without giving OpenAI/Claude thousands of dollars? Its morale, improving morale would increase the output in a much more holistic way. Scare the workers and you end up with spaghetti of everyone merging their crappy LLM enhanced code.

Re: LLMs are a 400-year-long confidence trick

#78

Post by keyle »

krystofee wrote: post #32

I disagree with the "confidence trick" framing completely. My belief in this tech isn't based on marketing hype or someone telling me it's good – it's based on cold reality of what I'm shipping daily. The productivity gains I'm seeing right now are unprecedented. Even a year ago this wouldn't have been possible, it really feels like an inflection point. I'm seeing legitimate 10x gains because I'm not writing code any…

It's fine for a Django app that doesn't innovate and just follows the same patterns for the 100 solved problems that it solves.

The line becomes a lot blurrier when you work on non trivial issues.

A Django app is not particularly hard software, it's hardly software but a conduit from database to screens and vice-versa; which is basic software since the days of terminals. I'm not judging your job, if you get paid well for doing that, all power to you. I had a well paying Laravel job at some point.

What I'm raising though is the fact that AI is not that useful for applications that aren't solving what has been solved 100 times before. Maybe it will be, some day, reasoning that well that it will anticipate and solve problems that don't exist yet. But it will always be an inference on current problems solved.

Glad to hear you're enjoying it, personally, I enjoy solving problems, not the end result as much.

Re: LLMs are a 400-year-long confidence trick

#79

Post by latexr »

mono442 wrote: post #44

I don't think it's true. It is probably overhyped but it is legitimately useful. Current agents can do around 70% of coding stuff I do at work with light supervision.

> It is probably overhyped

That’s exactly what a con is: selling you something as being more than what it actually is. If you agree it’s overhyped by its sellers, you agree it’s a con.

> Current agents can do around 70% of coding stuff I do

LLMs are being sold as capable of significantly more than coding. Focusing on that singular aspect misses the point of the article.

Re: LLMs are a 400-year-long confidence trick

#80

Post by falcor84 »

> We should be afraid, they say, making very public comments about “P(Doom)” - the chance the technology somehow rises up and destroys us.

> This has, of course, not happened.

This is so incredibly shallow. I can't think of even a single doomer, who ever claimed that AI will destroy us by now. P(doom) is about the likelihood of it destroying us "eventually". And I haven't seen anything in this post or in any recent developments to make my reduce my own p(doom), which is not close to zero.

Here are some representative values: https://pauseai.info/pdoom

Post reply on HN