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

Douglas Hofstadter changes his mind on Deep Learning and AI risk

lesswrong.com

Post reply on HN
Print view View original

11–20 of 520 posts

Re: Douglas Hofstadter changes his mind on Deep Learning and AI risk

#11

Post by jurgenaut23 »

colechristensen wrote: post #4
hospitalJail wrote: post #3

Lets take someone, who is past their prime and interview them on a topic they have never worked on. Then we can mine it for quotes! >"I never imagined that computers would rival, let alone surpass, human intelligence. And in principle, I thought they could rival human intelligence. I didn't see any reason that they couldn't." Yeah, so he got fooled by LLMs and hasnt been burned by it failing to do the most basic logi…

People overly impressed by LLMs haven't spent a lot of time trying to make them actually useful. When you do, you learn that they're talented mimics but still quite limited.

Yes, I actually think this is true. See my above comment, which supports your claim.

Re: Douglas Hofstadter changes his mind on Deep Learning and AI risk

#12

Post by jurgenaut23 »

ethanbond wrote: post #10
mercurialsolo wrote: post #5

Do we really lack a good understanding of LLM's and deep nets that we need to be afraid of them? I would love to see this disproved with some open source work on the internals of these models and how they do inference and exactly reason. And why they could possibly never realize an AGI with the current stream of models. Being able to display human level intelligence and creative in confined spaces (be it Chess or Go…

No one has ever been convinced to fly an airplane into a building by a chess board. People have gotten convinced of that by words though. The “attack surface” on human sensibility is just enormous if you’re able to use believably human language.

> The “attack surface” on human sensibility is just enormous if you’re able to use believably human language.

That's a fantastic quote ;-)

Re: Douglas Hofstadter changes his mind on Deep Learning and AI risk

#13

Post by lr4444lr »

jackconsidine wrote: post #2

In GEB Hofstadter dismisses the idea that AI could understand / compose / feel music like a human. I thought about this a lot when I started using GPT, especially early on when it demonstrated an ability to explain why things were funny or sad, intrinsically human qualities hitherto insulated from machine

Why, because it figured out sentiment groupings of words and phrases? There's lots of humor and tragedy from writers of eras past that just don't really land well with modern audiences unless they've studied and acclimated themselves to the culture.

Re: Douglas Hofstadter changes his mind on Deep Learning and AI risk

#14

Post by hackeraccount »

colechristensen wrote: post #4
hospitalJail wrote: post #3

Lets take someone, who is past their prime and interview them on a topic they have never worked on. Then we can mine it for quotes! >"I never imagined that computers would rival, let alone surpass, human intelligence. And in principle, I thought they could rival human intelligence. I didn't see any reason that they couldn't." Yeah, so he got fooled by LLMs and hasnt been burned by it failing to do the most basic logi…

People overly impressed by LLMs haven't spent a lot of time trying to make them actually useful. When you do, you learn that they're talented mimics but still quite limited.

This. LLM's have a surface that suggests they're an incredibly useful UI. That usability is like the proverbial hand full of water though - when you start to really squeeze it, it just slips away.

I'm still not convinced that the problem isn't me though.

Re: Douglas Hofstadter changes his mind on Deep Learning and AI risk

#16

Post by hackeraccount »

nielsbot wrote: post #7
colechristensen wrote: post #4

Earlier quoted context omitted.

People overly impressed by LLMs haven't spent a lot of time trying to make them actually useful. When you do, you learn that they're talented mimics but still quite limited.

Part of me wonders, though, could we "just" connect up an inference engine and voila? We could really be on a cusp of general AI. (Or it could be a ways off) That's a bit frightening in several ways.

I kind of expected AI to be AI - and not a mirror.

Re: Douglas Hofstadter changes his mind on Deep Learning and AI risk

#17

Post by EA-3167 »

ethanbond wrote: post #10
mercurialsolo wrote: post #5

Do we really lack a good understanding of LLM's and deep nets that we need to be afraid of them? I would love to see this disproved with some open source work on the internals of these models and how they do inference and exactly reason. And why they could possibly never realize an AGI with the current stream of models. Being able to display human level intelligence and creative in confined spaces (be it Chess or Go…

No one has ever been convinced to fly an airplane into a building by a chess board. People have gotten convinced of that by words though. The “attack surface” on human sensibility is just enormous if you’re able to use believably human language.

Well... words and decades of circumstances. If you removed the circumstances (the religion, the conflict, the money, geography, etc) then the words would be absolutely hollow.

I think we tend to credit words where often circumstances are doing the heavy lifting. For example try to start a riot with words in Rodeo Drive. Now try to do it in Nanterre. Or better yet, try to start a riot in Nanterre before a 17 year was shot by police, vs. after.

You'll get a sense of just how valuable your words really are.

Re: Douglas Hofstadter changes his mind on Deep Learning and AI risk

#18

Post by gwright »

hospitalJail wrote: post #3

Lets take someone, who is past their prime and interview them on a topic they have never worked on. Then we can mine it for quotes! >"I never imagined that computers would rival, let alone surpass, human intelligence. And in principle, I thought they could rival human intelligence. I didn't see any reason that they couldn't." Yeah, so he got fooled by LLMs and hasnt been burned by it failing to do the most basic logi…

> Its a language model and it fooled DH. It hasnt gotten smarter than us yet. Its just faster at repeating what other humans verbally said.

One of my take aways from LLMs is that humans very often just repeating what other humans have said with only superficial understanding of what they are repeating.

I think there is more to general intelligence than pattern matching and mimicking but a disconcerting amount of our day to day human interactions might just be that.

Re: Douglas Hofstadter changes his mind on Deep Learning and AI risk

#19

Post by toxik »

gwright wrote: post #8

Hofstadter makes the claim that "these LLMs and other systems like them are all feed-forward". That doesn't sound right to me, but I'm only a casual observer of LLM tech. Is his assertion accurate? FWIW, ChatGPT doesn't think so. :-)

It depends on how you define fed forward, LLMs are typically auto regressive and so can take their own previous output into consideration when generating tokens.

Re: Douglas Hofstadter changes his mind on Deep Learning and AI risk

#20

Post by SpaceManNabs »

gwright wrote: post #8

Hofstadter makes the claim that "these LLMs and other systems like them are all feed-forward". That doesn't sound right to me, but I'm only a casual observer of LLM tech. Is his assertion accurate? FWIW, ChatGPT doesn't think so. :-)

They are not all feed-forward unless it is some other definition that i am not aware of. Convolutional layers, XL hidden states, and graphical networks (which transformers are a special case of) aren't considered feedforward.

Unless you consider the entire instance as a singular instance and don't use any hidden states, then I guess it could be considered feed-forward.

I don't know. Feedforward doesn't seem like a useful term tbh. Some people mean feedforward as information only goes one direction, but that depends on your arrow. Autoregressive seems more useful here.

Post reply on HN