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

The future of AI according to thousands of forecasters

metaculus.com

Post reply on HN
Print view View original

31–40 of 85 posts

Re: The future of AI according to thousands of forecasters

#31

Post by nfc »

sdwr wrote: post #17
davidmanheim wrote: post #7

Earlier quoted context omitted.

> Nowhere do they define "AGI" Ummm, maybe you should have looked? At the top of the very first prediction, here: https://www.metaculus.com/questions/5121/date-of-artificial-... We will thus define "an AI system" as a single unified software system that can satisfy the following criteria, all completable by at least some humans. Able to reliably pass a 2-hour, adversarial Turing test during which the participants can…

I think we're reaching a point where the Turing test is no longer useful. If you get into the nitty-gritty of it (instead of just handwaving "computer should act like person"), it's about roleplaying a fake identity. Which is a specific skill, not a general test of competence.

The Turing test seems to be a product of an era where the nature and capabilities of artificial intelligence were still in the realms of the unknown. Because of that it was difficult to conceive a specific test that could measure its abilities. So the test ended up focusing on human intelligence—the most advanced form of intelligence known at that time—as the benchmark for AI.

To illustrate, imagine if an extraterrestrial race created a Turing-style test, with their intelligence serving as the gold standard. Unless their cognitive processes closely mirrored ours, it's doubtful that humans would pass such an examination

Re: The future of AI according to thousands of forecasters

#32

Post by avionical »

zgluck wrote: post #26

"Forecaster" here means any rando who signs up to their service and answers a question. Sometimes a large polling number does not equal a more accurate answer.

Anyone can forecast, sure. But there's a large body of research on the accuracy of aggregated forecasts and on the ability of forecasters to become more accurate with practice. (Thinking here in particular of work by Mellers & Tetlock.)

Metaculus provides a transparent track record of community forecasts here: https://www.metaculus.com/questions/track-record/ It's very difficult for any one person to consistently beat the community.

Re: The future of AI according to thousands of forecasters

#33

Post by nologic01 »

Its depressing to go over all this ultra-shallow chit-chat that has short-circuited any intelligent discussion about the role and trajectory of information technology (let alone any more serious problem or opportunity of the current times).

Talking about AI (and AGI) as if its some xenomorph lurking somewhere in silicon, waiting for its inevitable escape from its human prison.

AI will not bootstrap itself with some emergent property if somebody spends gazillions of dollars and Watts to estimate petazillions of parameters.

Further progress is not going to come unless some very human brain and intelligence opens up completely new algorithmic vistas.

The future of AI is literally tied to the future development of human mental (mathematical) models around information, knowledge and its digital representation.

If not intuitively obvious, the history of mathematical thought development is crushing evidence that it follows its own dynamic over timescales that span centuries.

Re: The future of AI according to thousands of forecasters

#34

Post by _ea1k »

jiggywiggy wrote: post #14
ilaksh wrote: post #4

Nowhere do they define "AGI". I guarantee that is a big reason why the predictions have so much variance. For many people, what GPT-4 does qualified as AGI -- up until GPT-4 came out and then everyone seemed to decide that AGI meant ASI. I am guessing for many people answering this poll it means "a full emulation of a person". Or maybe it had to be "alive". The thing that irritates me so much is that there is this la…

Chatgpt 4 is amazing. Still cant comprehend they build this. However if you work with it a lot you realize it doesn't understand reality, it predicts language. It reminds me of YouTube videos of chess when human start to figure out chess bots and the tricks they use to draw out time. Understanding language is super impressive, but a far cry from general intelligence. Maybe from here open ai can build further. But I h…

Is it surprising that a system that was exclusively trained on emitted language will only do well on language?

I don't see how I can extrapolate to anything beyond that. I certainly can't extrapolate that it would be unable to learn other tasks.

Re: The future of AI according to thousands of forecasters

#35

Post by notahacker »

davidmanheim wrote: post #9
malfist wrote: post #8

> Metaculus predicts a 75% likelihood of an attempt to influence the 2024 US presidential election with AI-driven misinformation. Hasn't that already happened? Wasn't DeSantis caught running deepfaked voices in an ad? Or am I misremembering something

It's more specific than that once they actually operationalize the question: This question resolves as YES if, in Meta's 2024 Q4 Quarterly Adversarial Threat Report, Meta claims that there was at least one "coordinated inauthentic behavior" that specifically pertained to the 2024 US Presidential election, and Meta suspects was primarily conducted via AI.

The operationalized question seems to be more of a bet on what Meta admits happens on its platform than what happens!

We know that people aiming to influence the elections use bots to inflate particular sentiments, statistical techniques which are arguably 'AI' to target messaging and have the capability to effortlessly generate varations on content using LLMs... it'd almost be odd if people didn't do so in a coordinated manner for the US election.

Re: The future of AI according to thousands of forecasters

#36

Post by famouswaffles »

jiggywiggy wrote: post #14
ilaksh wrote: post #4

Nowhere do they define "AGI". I guarantee that is a big reason why the predictions have so much variance. For many people, what GPT-4 does qualified as AGI -- up until GPT-4 came out and then everyone seemed to decide that AGI meant ASI. I am guessing for many people answering this poll it means "a full emulation of a person". Or maybe it had to be "alive". The thing that irritates me so much is that there is this la…

Chatgpt 4 is amazing. Still cant comprehend they build this. However if you work with it a lot you realize it doesn't understand reality, it predicts language. It reminds me of YouTube videos of chess when human start to figure out chess bots and the tricks they use to draw out time. Understanding language is super impressive, but a far cry from general intelligence. Maybe from here open ai can build further. But I h…

This is the problem here.

A Claim is made. "GPT isn't general" or "GPT isn't Intelligent" or whatever but a testable definition is not made. That is, what is generality to you and what bar need be passed ? Or What Intelligence is and what competence level need be surpassed ?

Without clearly stating those things, Intelligence may well be anything or any goal and your posts could shift to anywhere.

I'm just going to tell you facts of the state we're in right now.

Any testable definition of AGI that GPT-4 fails would also be failed by a significant chunk of the human population.

Re: The future of AI according to thousands of forecasters

#37

Post by welshwelsh »

jiggywiggy wrote: post #14
ilaksh wrote: post #4

Nowhere do they define "AGI". I guarantee that is a big reason why the predictions have so much variance. For many people, what GPT-4 does qualified as AGI -- up until GPT-4 came out and then everyone seemed to decide that AGI meant ASI. I am guessing for many people answering this poll it means "a full emulation of a person". Or maybe it had to be "alive". The thing that irritates me so much is that there is this la…

Chatgpt 4 is amazing. Still cant comprehend they build this. However if you work with it a lot you realize it doesn't understand reality, it predicts language. It reminds me of YouTube videos of chess when human start to figure out chess bots and the tricks they use to draw out time. Understanding language is super impressive, but a far cry from general intelligence. Maybe from here open ai can build further. But I h…

Define "reality."

Human experience consists of more than language, for sure. There is also a visual, audible and and tactile component. But GPT-4 can also take visual inputs from my understanding, and it shouldn't be difficult to add the other senses.

Google's experiments with PaLM-e show that LLMs are also useful in an embodied context, such as helping robots solve problems in real-time.

Re: The future of AI according to thousands of forecasters

#38

Post by usaar333 »

sdwr wrote: post #17
davidmanheim wrote: post #7

Earlier quoted context omitted.

> Nowhere do they define "AGI" Ummm, maybe you should have looked? At the top of the very first prediction, here: https://www.metaculus.com/questions/5121/date-of-artificial-... We will thus define "an AI system" as a single unified software system that can satisfy the following criteria, all completable by at least some humans. Able to reliably pass a 2-hour, adversarial Turing test during which the participants can…

I think we're reaching a point where the Turing test is no longer useful. If you get into the nitty-gritty of it (instead of just handwaving "computer should act like person"), it's about roleplaying a fake identity. Which is a specific skill, not a general test of competence.

Correct, which is part of the reason the "weak" AGI is relatively out there. Will anyone bother dumbing down an AI to pass a Turing Test? "Oh a human can't write a poem that fast -- it's an AI!"

Re: The future of AI according to thousands of forecasters

#39

Post by usaar333 »

Imnimo wrote: post #24

Is there any already resolved question in which metaculus forecasters made a surprisingly accurate prediction about an AI advance?

https://www.metaculus.com/questions/track-record/

Only 13 resolved binary questions where you had longer prediction horizons (1+ year), the accuracy is zilch in AI category - Brier score of 0.25 which is akin to just guessing out 50% for all questions. Generally overconfident.

(Other categories much better - Brier of 0.14 1 year out)

Re: The future of AI according to thousands of forecasters

#40

Post by tersh »

yunohn wrote: post #29
Imnimo wrote: post #24

Is there any already resolved question in which metaculus forecasters made a surprisingly accurate prediction about an AI advance?

Anyone can signup and vote, so I assume it tends towards the majority option tbh. Seems like obviously noisy data.

The Metaculus prediction weights predictors by how accurate they've been.
Post reply on HN