Where does this leave MIRI? Is Eliezer going to close up shop, collaborate with OpenAI, or compete?
MIRI is just a highly evolved form of LARP for burnt out savants. One would hope OpenAI is more pragmatic.
Introducing OpenAI
191–200 of 392 posts
Re: Introducing OpenAI
#192This is a key takeaway: "...we are going to ask YC companies to make whatever data they are comfortable making available to OpenAI. And Elon is also going to figure out what data Tesla and Space X can share." Money is great, openness is great, big name researchers are also a huge plus. But data data data, that could turn out to be very valuable. I don't know if Sam meant that YC companies would be encouraged to contr…
'The "evil" AI and safety stuff is just science fiction, but whatever.' umm... you can offer proof that we have nothing to worry about? Does the proof go like: Just as all people are inherently good, therefore all AIs will be inherently good? Or is it more like: since we can now safely contain all evil people, therefore we will be able to safely contain evil AIs? Sounds to me like there is some risk, no?
Being worried about "evil AI" is a lot like being worried about evil flying cars. Except people have built actual flying cars in the corporeal world, where "AI" is just a collection of signal processing techniques that don't work very well most of the time.
But hey, I labor in this domain: if paranoid richy-rich types want to throw money at it to ensure that they remain at the top of the heap, I'm all for it.
Re: Introducing OpenAI
#193Earlier quoted context omitted.
AI has been progressing in a fairly predictable way as computers get faster and gradually ticking off milestones like beating us and chess and driving cars with less crashes. There are only a finite number of such skill areas to tick off.
This is a weird reading of history. AI progress has been anything but predictable or steady.
I'm not sure it has been anything you'd define as "progress" either. "AI progress" is a lot like progress in controlled nuclear fusion as an energy source. Aka, there is no such thing, really, though people work on it.
Re: Introducing OpenAI
#194Isn't this the plot for Avengers: Age of Ultron?
Re: Introducing OpenAI
#195My money (not a billion) is on "Open, Big Learning".
Elon will probably want to build a giga-factory of neurons, then open-source some pre-trained, general model with a free API.
This is a man building electric cars, off-grid industrial-strength batteries, rockets and hyper-loops...I don't think publishing more/better research papers or winning kaggle competitions is the vision.
Re: Introducing OpenAI
#196Man I dunno about some of this media hype surrounding the topic of AI. I understand how powerful ML/AI algorithms are for general pattern matching (with a big enough computer, gradient descent can learn a lot of things...), but this whole skynet/doomsday fear thing seems ridiculous.
I guess the risk is embedding into systems that manage missiles or something. But you don't need sophisticated algorithms for that to be a risk, just irresponsible programmers. And I recon those systems already rely on a ton of software. So as long as we don't build software that tries to "predict where the this drone should strike next", we're probably fine. Actually shit we're probably doing that.. ("this mountanous cave has a 95% feature match with this other cave we bombed recently..."). Fuuuuck that sounds bad. I don't know how OpenAI giving other people AI will help against something like that.
Re: Introducing OpenAI
#197Earlier quoted context omitted.
See I think that's exactly where it becomes complicated. Can an entity with reasoning capabilities far beyond that of humans have its agency suppressed successfully? And is it ethical to do so or is that internally designed suppression somehow ethically different from the external suppression applied against human slaves? If you could engineer a human being with his/her agency removed so that you could use their reas…
The way you're asking the question implies that reasoning inherently has agency/sentience that needs suppressing. It doesn't need to; there's nothing to "suppress".
We don't know that one way or another since such a machine doesn't yet exist. I'm suggesting that perhaps high level reasoning and sentience go hand in hand although I can't say that with any certainty.
Re: Introducing OpenAI
#198Earlier quoted context omitted.
>I would take it a step further and say that worrying about the implications of AGI is like thinking about Earth being overpopulated by space aliens. First we have to establish that such a thing is even possible, for which there is currently no concrete proof. Given that I could probably sketch out a half-assed design for one in nine months if you gave me a full-time salary - or rather, I could consult with a bunch o…
Ahhh... The power of not knowing what you don't know. I mean, why can't I live forever? Let's just list the problems and solve them in the next year!
Yes, let’s.
Re: Introducing OpenAI
#199Earlier quoted context omitted.
Except that (right or wrong) we most certainly weigh that against their intelligence.
We do?? Humans have been known to eat practically every animal, even those they use as beasts of burden or as working animals. EDIT: Just to be clear when I talk about enslaving animals I'm referring to use of animals in law enforcement, medicine, war, farming, hunting, for companionship, for food production, material production etc.
I was mainly thinking about food supply and medical testing but I think the rest apply as well. For instance I think it's generally considered more ethical to perform lethal medical testing on rats than gorillas. I think that's got a lot to do with species intelligence.
There are plenty of people who don't weigh the ethics at all and just eat whatever animal they want, but in many societies eating a dolphin or a gorilla is considered repugnant.
Perhaps it's more correct to say morals than ethics here.
Re: Introducing OpenAI
#200So the idea with differential safety development is that we want to speedup safe AI timelines as much as possible while slowing down unsafe AI timelines as much as possible. I worry that this development isn't great when viewed through this lens. Lets say that DARPA, CAS, and whatever the Russian equivalent all work on closed source AIs. The idea here might be that open source beats closed source by getting cross pollination and better coordination between efforts. The issue is that the government agencies get to crib whatever they want from the open source stuff to bolster their own closed source stuff.