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

A bear case: My predictions regarding AI progress

lesswrong.com

Post reply on HN
Print view View original

121–130 of 220 posts

Re: A bear case: My predictions regarding AI progress

#121

Post by esafak »

xnx wrote: post #71
roncesvalles wrote: post #64

Earlier quoted context omitted.

They were revolutionary as product genres, not necessary individual companies. Ordering a cab without making a phone call was revolutionary. Netflix at least with its initial promise of having all the world's movies and TV was revolutionary, but it didn't live up to that. Spotify because of how cheap and easy it was to have access to all the music, this was the era when people were paying 99c per song on iTunes. I've…

> Ordering a cab without making a phone call was revolutionary. With the power of AI, soon you'll be able to say "Hey Siri, get me an Uber to the airport". As easy as making a phone call.

Easier, because you don't have to search for a phone number.

Re: A bear case: My predictions regarding AI progress

#122

Post by viccis »

cheevly wrote: post #53
viccis wrote: post #6

Regarding "AGI", is there any evidence of true synthetic a priori knowledge from an LLM?

Produce true synthetic a priori knowledge of your own, and ill show you an automated LLM workflow that can arrive at the same outcome without hints.

Build an LLM on a corpus with all documents containing mathematical ideas removed. Not a single one about numbers, geometry, etc. Now figure out how to get it to tell you what the shortest path between two points in space is.

Re: A bear case: My predictions regarding AI progress

#123

Post by mcintyre1994 »

> Scaling CoTs to e. g. millions of tokens or effective-indefinite-size context windows (if that even works) may or may not lead to math being solved. I expect it won't.

> (If math is solved, though, I don't know how to estimate the consequences, and it might invalidate the rest of my predictions.)

What does it mean for math to be solved in this context? Is it the idea that an AI will be able to generate any mathematical proof? To take a silly example, would we get a proof of whether P=NP from an AI that had solved math?

Re: A bear case: My predictions regarding AI progress

#124

Post by NitpickLawyer »

usaar333 wrote: post #28
liuliu wrote: post #5

I think all these articles begging the question: what's author's credential to claim these things. Be careful about consuming information from chatters, not doers. There is only knowledge from doing, not from pondering.

LW isn't a place that cares about credentialism. He has tons of links for the objective statements. You either accept the interpretation or you don't.

> He has tons of links for the objective statements.

I stopped at this quote

> LLMs still seem as terrible at this as they'd been in the GPT-3.5 age.

This is so plainly, objectively and quantitatively wrong that I need not bother. I get hyperbole, but this isn't it. This shows a doubling-down on biases that the author has, and no amount of proof will change their mind. Not an article / source for me, then.

Re: A bear case: My predictions regarding AI progress

#125

Post by gwern »

usaar333 wrote: post #30

Author also made a highly upvoted and controversial comment about o3 in the same vein that's worth reading: https://www.lesswrong.com/posts/Ao4enANjWNsYiSFqc/o3?comment... Oh course lesswrong, being heavily AI doomers, may be slightly biased against near term AGI just from motivated reasoning. Gotta love this part of the post no one has yet addressed: > At some unknown point – probably in 2030s, possibly tomorrow (bu…

I never thought I'd see the day that LessWrong would be accused of being biased against near-term AGI forecasts (and for none of the 5 replies to question this description either). But here we are. Indeed do many things come to pass.

Re: A bear case: My predictions regarding AI progress

#126

Post by daveguy »

mcintyre1994 wrote: post #123

> Scaling CoTs to e. g. millions of tokens or effective-indefinite-size context windows (if that even works) may or may not lead to math being solved. I expect it won't. > (If math is solved, though, I don't know how to estimate the consequences, and it might invalidate the rest of my predictions.) What does it mean for math to be solved in this context? Is it the idea that an AI will be able to generate any mathemat…

I think "math is solved" refers more to AI performing math studies at the level of a mathematics graduate student. Obviously "math" won't ever be "solved" but the problem of AI getting to a certain math proficiency level could be. No matter how good an AI is, if P != NP it won't be able to prove P=NP.

Regardless I don't think our AI systems are close to a proficiency breakthrough.

Edit: it is odd that "math is solved" is never explained. But "proficient to do math research" makes the most sense to me.

Re: A bear case: My predictions regarding AI progress

#127

Post by esafak »

Workaccount2 wrote: post #96
swazzy wrote: post #38

I see no reason to believe the extraordinary progress we've seen recently will stop or even slow down. Personally, I've benefited so much from AI that it feels almost alien to hear people downplaying it. Given the excitement in the field and the sheer number of talented individuals actively pushing it forward, I'm quite optimistic that progress will continue, if not accelerate.

If LLM's are bumpers on a bowling lane, HN is a forum of pro bowlers. Bumpers are not gonna make you a pro bowler. You aren't going to be hitting tons of strikes. Most pro bowlers won't notice any help from bumpers, except in some edge cases. If you are an average joe however, and you need to knock over pins with some level of consistency, then those bumpers are a total revolution.

That is not a good analogy. They are closer to assistants to me. If you know how and what to delegate, you can increase your productivity.

Re: A bear case: My predictions regarding AI progress

#128

Post by code_for_monkey »

spaceman_2020 wrote: post #43
cglace wrote: post #32

The thing I can't wrap my head around is that I work on extremely complex AI agents every day and I know how far they are from actually replacing anyone. But then I step away from my work and I'm constantly bombarded with “agents will replace us”. I wasted a few days trying to incorporate aider and other tools into my workflow. I had a simple screen I was working on for configuring an AI Agent. I gave screenshots of…

You’re biased because if you’re here, you’re likely an A-tier player used to working with other A-tier players. But the vast majority of the world is not A players. They’re B and C players I don’t think the people evaluating AI tools have ever worked in wholly mediocre organizations - or even know how many mediocre organizations exist

wish this didnt resonate with me so much. Im far from a 10x developer, and im in an organization that feels like a giant, half dead whale. Sometimes people here seem like they work on a different planet.

Re: A bear case: My predictions regarding AI progress

#129

Post by alabastervlog »

xnx wrote: post #71
roncesvalles wrote: post #64

Earlier quoted context omitted.

They were revolutionary as product genres, not necessary individual companies. Ordering a cab without making a phone call was revolutionary. Netflix at least with its initial promise of having all the world's movies and TV was revolutionary, but it didn't live up to that. Spotify because of how cheap and easy it was to have access to all the music, this was the era when people were paying 99c per song on iTunes. I've…

> Ordering a cab without making a phone call was revolutionary. With the power of AI, soon you'll be able to say "Hey Siri, get me an Uber to the airport". As easy as making a phone call.

And they'll be able to tack an extra couple dollars onto the price because that's a good signal you're not gonna comparison shop.

Innovation!

Re: A bear case: My predictions regarding AI progress

#130

Post by demosthanos »

csomar wrote: post #25
kiratp wrote: post #23

Earlier quoted context omitted.

You’re not using the best tools. Claude Code, Cline, Cursor… all of them with Claude 3.7.

Nope. I try the latest models as they come and I have a self-made custom setup (as in a custom lua plugin) in Neovim. What I am not, is selling AI or AI-driven solutions.

It's worth actually trying Cursor, because it is a valuable step change over previous products and you might find it's better in some ways than your custom setup. The processes they use for creating the context seems to be really good. And their autocomplete is far better than Copilot's in ways that could provide inspiration.

That said, you're right that it's not as overwhelmingly revolutionary as the internet would lead you to believe. It's a step change over Copilot.

Post reply on HN