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

Please ignore the deluge of complete nonsense about Q*

twitter.com

Post reply on HN
Print view View original

201–208 of 208 posts

Re: Please ignore the deluge of complete nonsense about Q*

#201

Post by staticman2 »

_heimdall wrote: post #196
staticman2 wrote: post #190

Earlier quoted context omitted.

If your prediction is "A.I. is similar to magic and can do anything and will become a god" you would of course think it can solve the protein folding problem... I'm not exactly sure that leads to the conclusion his track record is great. (There are many more things A.I. can't do than can do, I would think?)

His prediction was that humans would solve the protein folding problem, it didn't have to do with AI. I raised it only as an example of a prediction he made that most at the time thought had no chance. My point isn't to claim he's some genius that gets everything right, but there are examples of him making predictions widely viewed at the time of having no chance that turned out to be correct. Writing him off because…

Again, if my prediction is "every scientific advance will happen" then I can claim prescience by claiming every scientific advance is evidence of my great insight. That hardly makes me insightful.

I'm writing him off because he isn't a practitioner. I can make things up about what will or won't happen on my own, as can anyone else.

If I want to believe every wacky concept from science fiction will happen, I don't need a middleman to tell me it will.

Re: Please ignore the deluge of complete nonsense about Q*

#202

Post by agnosticmantis »

jarpschop wrote: post #144
agnosticmantis wrote: post #125

LeCun is one of the rare voices of reason in the AI community, especially among the celebrities. He’s been consistently voicing realistic statements with regards to AGI (he doesn’t believe it’s a thing), AI safety and doomerism, limitations of LLMs, etc.

A year ago he said that language models would never be able to explain why if you move a table, the bottle on top of it would not fall. He definitely had no idea what was coming with LLMs. And now he is arrogant enough to say "we all knew what capabilities LLMs would have". He has no intelectual humility or honesty at all. I at least admit openly that language models blew my mind. Lost respect for his opinions massiv…

He’s said it’s easy to generate this kind of questions that trick LLMs because of the lack of physical grounding in models trained solely on text.

And that’s true now as ever. I also heard him say that training multimodal models on text+image/video would mitigate the grounding issue, and that’s proven to be true too.

So I’m not sure exactly what your objection is.

Re: Please ignore the deluge of complete nonsense about Q*

#203

Post by giardini »

peteradio wrote: post #162
giardini wrote: post #133

Earlier quoted context omitted.

flagrant_taco says >"How would you define or recognize a level of intelligence or autonomy that is sufficient to raise concerns?" If it asks for the right to vote?

Real intelligence will be knowledge that voting is for suckers and the world is run by a congress where the dominant parties are the illuminati and lizard people. Now if an intelligence can determine and conduct the sacred ritual to join the voting pool then we might onto something. But the true test will be when the intelligence again avoids the suckers game and that the world is actually...

Maybe there's something you don't know perhaps?

Here you are, trying to "outthink" an AI and speaking as if you understand (both the AI and the world of politics at the very least)! Isn't that silly!

Re: Please ignore the deluge of complete nonsense about Q*

#204

Post by _heimdall »

staticman2 wrote: post #201
_heimdall wrote: post #196

Earlier quoted context omitted.

His prediction was that humans would solve the protein folding problem, it didn't have to do with AI. I raised it only as an example of a prediction he made that most at the time thought had no chance. My point isn't to claim he's some genius that gets everything right, but there are examples of him making predictions widely viewed at the time of having no chance that turned out to be correct. Writing him off because…

Again, if my prediction is "every scientific advance will happen" then I can claim prescience by claiming every scientific advance is evidence of my great insight. That hardly makes me insightful. I'm writing him off because he isn't a practitioner. I can make things up about what will or won't happen on my own, as can anyone else. If I want to believe every wacky concept from science fiction will happen, I don't nee…

Do you have an example of a prediction he made that didn't own out? I'm sure there are some, I wouldn't expect anyone making predictions to bat 1000, but throwing his opinion out even when there are examples of accurate predictions doesn't seem fair.

It does feel risky to write off anyone that isn't a practitioner. There won't be any practitioners working on AI risk, its a theoretical pursuit by nature. The closest a practitioner could get is AI risk management, i.e. attempting to limit risk for AI we're already developing.

Re: Please ignore the deluge of complete nonsense about Q*

#205

Post by yowlingcat »

_heimdall wrote: post #204
staticman2 wrote: post #201

Earlier quoted context omitted.

Again, if my prediction is "every scientific advance will happen" then I can claim prescience by claiming every scientific advance is evidence of my great insight. That hardly makes me insightful. I'm writing him off because he isn't a practitioner. I can make things up about what will or won't happen on my own, as can anyone else. If I want to believe every wacky concept from science fiction will happen, I don't nee…

Do you have an example of a prediction he made that didn't own out? I'm sure there are some, I wouldn't expect anyone making predictions to bat 1000, but throwing his opinion out even when there are examples of accurate predictions doesn't seem fair. It does feel risky to write off anyone that isn't a practitioner. There won't be any practitioners working on AI risk, its a theoretical pursuit by nature. The closest a…

I think the problem that GP is getting at is the philosophical problem of apocalypticism or market perma-bears, which can be summed up in this satire of them: "So and so is a great market analyst! He's predicted 30 of the last 2 market crashes!" A broken clock can still be right twice a day, but it's still a broken clock. Predictions about the future which are so broad so as to be hopelessly incorrect should not be excused because part of the prediction surface ended up being correct; rather, the entirety of the prediction surface and accuracy should be taken into account to determine the overall accuracy of the prediction.

Taken in that light, his being correct about his protein folding prediction feels like cold comfort when taken in context of his predictions, largely unsubstantiated, of AGI becoming a malevolent human extinction force at par with or of greater risk than that atomic bomb. It elides the very real conceptual limitations of current implementations of AI, such as it's inability to recognizing bidirectional associations (Tom is John's son means John is Tom's father), protecting against injection risks, "hallucinations" (inability to differentiate between when it is making something up and when it is not) and a variety of other existential limits towards its own power.

These details, far from being inconvenient nuances that do not detract from his overall point, should actually serve to make us seriously skeptical of how well he understands the risks of the entity he is warning against. A more skeptical observer might even say (and I'm one of them) that much like fire and brimstone preachers, he is trying to create an unrealistic fear of a deific entity that cannot be reasoned against so that he can consolidate power by serving a "solution" that protects against it and curry popular favor. I find his approach to be the distasteful and manipulative approach of the common demagogue. The haughty, arrogant, insecure way that he treats the arguments of his critics (he simply blocks them on Twitter or condescends to them) make me even more skeptical of him. It makes me wonder if on some level he knows that his argument is not being made in good faith, and isn't trying to win an argument in good faith.

Re: Please ignore the deluge of complete nonsense about Q*

#206

Post by keepamovin »

ethanbond wrote: post #140
hnfong wrote: post #139

Earlier quoted context omitted.

Given that we already know that there was a conspiracy (by the now-ousted OpenAI board to remove Sam Altman), and that people here on HN have been theorizing about this for almost a week now, your definition makes everyone who commented about it a "conspiracy theorist". In actual "language game" usage, it's just a derogatory term meaning "I don't like these unproven claims, even though I can't disprove them".

No it’s actually a derogatory term meaning “excessively complicated theory to support my priors.” Such as casting clear evidence toward the “safety/commercialization of next gen tech was a point of contention among the board” thesis instead as: > The leak about Q* feels like an olive branch to let the former OpenAI board and Ilya save a bit of face, probably part of the terms for Sam coming back plus it distracts fro…

What if there's multiple overlapping but distinct meanings of the term conspiracy theory? Wouldn't that be a conspiracy in itself to diffract the meaning? Or would would the mere suggestion that it were itself be a conspiracy theory hiding a far less sinister reality? Or what if, like a fractal, it's everything everywhere all at once? :) hahaha

Re: Please ignore the deluge of complete nonsense about Q*

#207

Post by worldsayshi »

harshreality wrote: post #89
qualifiedai wrote: post #76

Earlier quoted context omitted.

Show me LLM that has reached a level of automony and self sufficiency of a cat (Yann's example) or even of a simple beattle. LLMs are tools in amplifying individual human intelligence, 100% of automony and will come from their user (human). Also, even as tools they have fundamental limitations which stem from their autoregressive nature

Before the era of quantum physics, the idea of a weapon like fission or fusion weapons would've been inconceivable without resorting to magical thinking. Your argument, and Yann's, is that AGI, or what you call AGI, is a kind of quasi-intelligent golem that, despite being generally intelligent, doesn't have human-level intelligence. Your claim that it will never be human-equivalent, much less trans-human/ASI, is buil…

>despite being generally intelligent, doesn't have human-level intelligence

I think this whole quote is riddled with assumptions in this debate.

What is human level? Is it really a "level" or is human level just a local variant in a space of possible intelligence varieties that maybe could be sorted along one dimension of levels of maybe multiple? Can it be super intelligent without being autonomous at all?

I'm not saying that you're wrong in just pointing out where people ought to have a myriad of different assumptions.

Re: Please ignore the deluge of complete nonsense about Q*

#208

Post by staticman2 »

_heimdall wrote: post #204
staticman2 wrote: post #201

Earlier quoted context omitted.

Again, if my prediction is "every scientific advance will happen" then I can claim prescience by claiming every scientific advance is evidence of my great insight. That hardly makes me insightful. I'm writing him off because he isn't a practitioner. I can make things up about what will or won't happen on my own, as can anyone else. If I want to believe every wacky concept from science fiction will happen, I don't nee…

Do you have an example of a prediction he made that didn't own out? I'm sure there are some, I wouldn't expect anyone making predictions to bat 1000, but throwing his opinion out even when there are examples of accurate predictions doesn't seem fair. It does feel risky to write off anyone that isn't a practitioner. There won't be any practitioners working on AI risk, its a theoretical pursuit by nature. The closest a…

"Do you have an example of a prediction he made that didn't own out?"

How many of his predictions are even disprovable? Do you have a list? Was that protien fold prediction even disprovable?

An example of a nondisprovable prediction is:

-one day an AI will become self aware.

A disprovable one is:

-Skynet will begin to learn at a geometric rate. It will becomes self-aware at 2:14 a.m. Eastern time, August 29th 1997.

Post reply on HN