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

What we still don’t know about how A.I. is trained

newyorker.com

Post reply on HN
Print view View original

101–110 of 211 posts

Re: What we still don’t know about how A.I. is trained

#101

Post by fasterik »

bigfishrunning wrote: post #81
the_sleaze9 wrote: post #72

Earlier quoted context omitted.

You say that, but if I'm confused about something and think hard about it, I think in language. If you blinded me, paralyzed me, deafened me and desensitized my olfactions, I could still think, but what I would be doing is feeding one language thought into another. It's not so much different from "text" imho.

yes, but for you all of that text is associated with ideas. The word "dog" has an associated object. For a machine like GPT-4, the word "dog" has no meaning or object, but it does have an associated likelihood for adjacent words. The words themselves aren't the intelligence, the ideas behind them are.

It's an open question the extent to which language models can learn the underlying meanings of words.

There was a recent podcast with Sean Carroll interviewing Raphaël Millière where they go into this topic and some of the research on it. Two examples I can remember are: 1) DALL-E had subject-specific, domain-specific neurons, 2) language models' representations of color terms encoded the geometry of the underlying color space, e.g. vectors in RGB space.

https://www.preposterousuniverse.com/podcast/2023/03/20/230-...

I don't think we should be too quick to assume how these models work. There's a lot that even the researchers don't know and these are empirical questions that can be studied.

Re: What we still don’t know about how A.I. is trained

#102

Post by kzrdude »

saiya-jin wrote: post #75
startupsfail wrote: post #42

Earlier quoted context omitted.

What is OpenAI is right and the risks are real? They likely already have some glimpses of GPT-5 internally. And GPT-4 is closely resembles AGI already.

> GPT-4 is closely resembles AGI already Thats a very bold statement and goes against everything I've read on it so far, care to backup such a claim with some facts? Of course each of us has their own bar for such things, but for most its pretty darn high

OpenAI people put out this paper https://arxiv.org/abs/2303.12712 called Sparks of Artificial General Intelligence: Early experiments with GPT-4 and as the title makes clear, they think it has hints of AGI. I guess that's a good place to start to answer your question. I don't think this is AGI, but the paper is full of examples where GPT-4 works well and does impressive stuff.

Re: What we still don’t know about how A.I. is trained

#103

Post by danielovichdk »

2 things.

1. We do seem to forget that even though ChatGPT is a great technical achievement its achieved by companies and humans. These companies and humans had agendas.

2. Open Source only got us so far. With these new AI models and trying to withhold competitive advantages will we see a new era of closed source tech which will like its predecessors only spew discomfort amongst its users and bridge the gap even further.

AI is dangerous because its already being looked at as an entity of which is human like.

Its perception amongst us is that its equally human on a basic intelligent level as most of us. And that I believe is scary because it puts us closer to it given it acts as a human.

Not a fan

Re: What we still don’t know about how A.I. is trained

#104

Post by pessimizer »

mousebaiker wrote: post #95
sergioisidoro wrote: post #87

«When Dean Buonomano, a neuroscientist at U.C.L.A., asked GPT-4 “What is the third word of this sentence?,” the answer was “third.” These examples may seem trivial, but the cognitive scientist Gary Marcus wrote on Twitter that “I cannot imagine how we are supposed to achieve ethical and safety ‘alignment’ with a system that cannot understand the word ‘third’ even [with] billions of training examples.”» The word "thir…

Maybe your eyes played the same trick on you as they did on me. When I first read the sentence, I also thought that "third" is the third word. Upon rechecking I realized that it is the fourth with the third word being "the".

Green is one of the two green words in this sentence.

Re: What we still don’t know about how A.I. is trained

#105

Post by samus »

elif wrote: post #48
extr wrote: post #45

Earlier quoted context omitted.

> Is that assistance? ...Yes? The fact that it's basic and they already know how to do it is irrelevant. The law isn't "it's okay to give them advice as long as the advice is sufficiently generic and obvious".

So your position is that wikipedia are terrorists too? Why would their "advise" be exempt?

Because of a similar reasoning why gas station owners don't get complicit in crimes when they sell fuel to potential criminals. At least as long as they can plausibly deny knowing of the crime and they provide similar service than to any other customer.

Wikipedia, and the sources where its content comes from, is not intended to help anybody specific. Flying to North Korea and holding a lecture there is firmly beyond the line where the US government starts to care because it demonstrates clear intent. Especially since during the Q&A and other activities other information not available to the public could have been discussed. And even if there was no secret sauce involved, an expert's opinion can still save the North Koreans a lot of time and money when pursuing their goals.

Re: What we still don’t know about how A.I. is trained

#106

Post by throwaway17197 »

Domenic_S wrote: post #96
pdntspa wrote: post #93

Earlier quoted context omitted.

"Third" is the 4th word in that sentence. Do one of those other words not count or something?

I think they mean if you match the word itself as a string rather than interpreting the meaning of the word, e.g., "what word in this sentence === 'third'" I can sort of see how that could be a machine's interpretation if I squint really hard

It could be understood as a (not very good) joke along the lines of this https://commons.wikimedia.org/wiki/File:FoundX.svg

Re: What we still don’t know about how A.I. is trained

#107

Post by scottyah »

kossTKR wrote: post #69
matheusmoreira wrote: post #16

Earlier quoted context omitted.

Just say what's on your mind and don't mind the votes. One thing you'll discover is that you're not alone in your views, whatever they are. Few days ago I came across this bone chilling AI generated Metal Gear Solid 2 meme with Hideo Kojima characters talking about how the purpose of this technology is to make it impossible to tell what's real or fake, leading directly to regulation of information networks with ident…

Yes, this is the real dark future right here. It will be a continuation though. The uppermost echelons of society have waged information warfare since the dawn of modern PR in the beginning of the 20'th century. Lots of theory on this have apparently been memoryholed, but it's easy to just start with the genealogy around Edward Bernais and the plutocracy and robber baron families still existing in the interwar period…

Who are the "microscopic elite that controls the media"? How many people are in the "uppermost echelons of society" to where they try to influence other people?

From my middle-of-the-road perspective, everyone is trying to change how everyone else thinks, from the small insignificant details to a cult-like brainwash. Even here and now both you and I are trying to change each others and everyone who reads this's mind.

I see so much calling out of "elites" with a negative connotation, yet anyone can post anything, and everyone can ignore it. Is it general paranoia? Do you know who these people are? How are they special/different, and why should we be scared of them?

Re: What we still don’t know about how A.I. is trained

#108

Post by majormajor »

kossTKR wrote: post #69
matheusmoreira wrote: post #16

Earlier quoted context omitted.

Just say what's on your mind and don't mind the votes. One thing you'll discover is that you're not alone in your views, whatever they are. Few days ago I came across this bone chilling AI generated Metal Gear Solid 2 meme with Hideo Kojima characters talking about how the purpose of this technology is to make it impossible to tell what's real or fake, leading directly to regulation of information networks with ident…

Yes, this is the real dark future right here. It will be a continuation though. The uppermost echelons of society have waged information warfare since the dawn of modern PR in the beginning of the 20'th century. Lots of theory on this have apparently been memoryholed, but it's easy to just start with the genealogy around Edward Bernais and the plutocracy and robber baron families still existing in the interwar period…

Are you claiming that powerful people only started using their power a hundred years ago when "modern PR" was formed? That nobody ever worried about information before that?

Why did you pick that as your starting point? To make things seem like a linear worsening? But even wealth inequality, which you talk about at length, got better before getting worse again during that time frame in the US... so I'm having trouble following the claim.

Re: What we still don’t know about how A.I. is trained

#109

Post by airstrike »

slg wrote: post #82
whitemary wrote: post #11

Earlier quoted context omitted.

God bless these saviors for prioritizing our "safety!"

Are you suggesting an alternative of not caring about how this technology will negatively impact society is a preferable approach?

There's a difference between caring and saying you care

Re: What we still don’t know about how A.I. is trained

#110

Post by king_magic »

ozten wrote: post #88
frrae wrote: post #66

GPT Is Not A.I. We tech people should actively go on the offence and educate whomever we can that text inference is not intelligence.

Humans are constantly moving the goal posts for what constitutes AGI. Based on GPT 4's performance on standardized tests and my anecdotal evidence of utility I'm getting out of ChatGPT, I feel that weak AGI arrived with GPT-4.

This is my take as well. It's still weak, and the more I use it the more I can see where it's just not quite there yet, but yeah - GPT 4 is clearly something new. Just spend any significant period of time with it trying to automate your own tasks/workflows, it is borderline kinda sorta magical, with a heavy emphasis on "borderline kinda sorta".
Post reply on HN