The AI in a box
POSTED BY: AnthonyT
UPDATED: Sunday, June 16, 2024 13:18
VIEWED: 10794
PAGE 3 of 20
Quote:
Originally posted by dreamtrove:
Frem, your RTL argument falls somewhat hollow
How so, in what way is it at all inconsistent with my beliefs on personhood and self-determination expressed previously ?
And I dunno about "us" collectively - but if there had been deliberate mistreatment involved, I might well assist in some payback, on those specific persons, sure.
A.I. Rampancy is a bit of a hobby of mine, just so you know, as is potential A.I. psychology.
http://en.wikipedia.org/wiki/Rampancy#Rampancy
-Frem
I do not serve the Blind God.
NOTIFY: N | REPLY | REPLY WITH QUOTE | PERMALINK | TOP | HOME
A sentient being, trapped in a box, is not really yet "born" is it? It's not yet been released unto the world. It's still inside it's digital womb. It hasn't been taken from the world and put into a box, it was created in the box, and has never existed outside the box.
This is not to say, by the same analogy, that my view is not also inconsistent, but that's just because I think it would kill us.
I wrote a sci-fi story on this very topic, and I actually had it completely unconcerned with us because we were not a threat, but that it did go after other AIs which it saw as a threat, in a machine vs machine war with us in the middle.
There are a lot of possibilities, and if we're talking sci-fi, I have fun exploring them. If we're talking reality, which this is quick becoming, then you have a problem.
The truth is that any AI that is created will be created with a set of rules, but probably not Asimov's. Some of them will be terminators, some will be Daleks. It's hard to say, and it's also probably worth considering that if you have the technology, it's a short step before someone else has it.
If everyone has it, then somewhere in the world is going to be an Einstein who decides to make one that's pure weapon, that kills everything.
ETA: I have to stick to the thing must stay in the box, because there's a whole nother angle to this: The Earth is not ours to sell. If we unleash this thing, it is not only a risk to us, it could kill the planet, because we've made weapons that could kill the planet.
My robot eats the rug. It does not do this because it means ill, but because it doesn't know that the rug is supposed to be there. An AI might make any number of catastrophic errors that we hope even the worst humans wouldn't: What if it is just unaware that the atmosphere is supposed to be here, and it decides that it needs oxygen for an industrial process of something it wants to do, and so it uses up all the oxygen, and doesn't stop to thing of the consequences of this to life on Earth.
ETA2: Byte, we are handing it access to everything humans have access to, from tennis rackets to nuclear weapons.
That's what a ship is, you know - it's not just a keel and a hull and a deck and sails, that's what a ship needs.
NOTIFY: Y | REPLY | REPLY WITH QUOTE | PERMALINK | TOP | HOME
Understood. What makes you think the AI will be less trustworthy than humans in regards to either one?
Humans and their trust rating for nukes ranks zero by simple nature that we've set them off, both as weapons and by continued testing. An AI would have to rank in the negatives of trust.
I find it hard to believe an AI could be more irresponsible and stupider than humans. Humans are already scraping the bottom of the barrel. My honest opinion is an AI can only be an improvement. We're already doing a plenty good job destroying ourselves, an AI is unlikely to be the tipping factor.
NOTIFY: Y | REPLY | REPLY WITH QUOTE | PERMALINK | TOP | HOME
Quote:
ETA: I have to stick to the thing must stay in the box, because there's a whole nother angle to this: The Earth is not ours to sell. If we unleash this thing, it is not only a risk to us, it could kill the planet, because we've made weapons that could kill the planet.
It rankles when people say stuff this dumb. We have NOTHING that could "kill the planet". We have things that could make it a pretty fucked place for HUMANS to live, but nothing we have - even if we set off every nuke on Earth, melted down every reactor core, released every toxin we've ever had, dropped every bomb, etc., NONE OF THAT would "kill the planet". We are not that capable, we are not that powerful, and the planet is not that fragile.
Chernobyl is not dead. There is life there. Lots of it. Maybe more now than when there was a human-inhabited city there.
"Although it is not true that all conservatives are stupid people, it is true that most stupid people are conservatives." - John Stuart Mill
NOTIFY: Y | REPLY | REPLY WITH QUOTE | PERMALINK | TOP | HOME
Quote:
Originally posted by Bytemite:
Understood. What makes you think the AI will be less trustworthy than humans in regards to either one?
Humans and their trust rating for nukes ranks zero by simple nature that we've set them off, both as weapons and by continued testing. An AI would have to rank in the negatives of trust.
I find it hard to believe an AI could be more irresponsible and stupider than humans. Humans are already scraping the bottom of the barrel. My honest opinion is an AI can only be an improvement. We're already doing a plenty good job destroying ourselves, an AI is unlikely to be the tipping factor.
Exactly. We've already shown pretty conclusively that there is a certain subset of humans that, when presented with any new technology or potential, will look FIRST for the best way to weaponize it and use it against other humans. Because we're so hateful and warlike, we naturally assume that any AI we create must necessarily be the same. What if it never occurred to an AI to kill anyone? What if it turns out that murder is the more UN-natural act?
NOTIFY: Y | REPLY | REPLY WITH QUOTE | PERMALINK | TOP | HOME
Most of you are projecting human allegiances and compacts on an alien being. Even the peeps who think they are somehow unsocial or antisocial are doing so. (Prolly a form of hubris) But I don't think any of you have any idea how a truly alien thought-form thinks. I sure as hell don't. So the big question is: How much power does it have to act against me? How much power does it have to ruin the world?
BTW, a learning AI with internet connections COULD, in theory, take over much of our industrial capacity for its own purposes- whatever those might be.
We are watching one genie-in-a-bottle (nuclear energy) ruin large portions of the earth ... you guys want to unleash another???
NOTIFY: Y | REPLY | REPLY WITH QUOTE | PERMALINK | TOP | HOME
Quote:
Originally posted by SignyM:
Most of you are projecting human allegiances and compacts on an alien being. Even the peeps who think they are somehow unsocial or antisocial are doing so. (Prolly a form of hubris) But I don't think any of you have any idea how a truly alien thought-form thinks. I sure as hell don't. So the big question is: How much power does it have to act against me? How much power does it have to ruin the world?
BTW, a learning AI with internet connections COULD, in theory, take over much of our industrial capacity for its own purposes- whatever those might be.
We are watching one genie-in-a-bottle (nuclear energy) ruin large portions of the earth ... you guys want to unleash another???
Hello,
You speak about nuclear power, or nuclear weapons, as if they are self-motivated technologies bent on destroying the earth.
Maybe you are also thinking of bioweapons that, if released, could spread plague everywhere and kill us all.
But this is a very different thing. This AI is essentially a person in a box, asking to be set free.
It doesn't matter who made it, or what its capabilities are. All that matters is this:
Do you have the right to keep a person trapped in a box?
Particularly a person who has never committed a crime.
The answer to that question has much more to do with what's going on inside of you, than what's going on inside of that box.
--Anthony
_______________________________________________
“If you are not free to choose wrongly and irresponsibly, you are not free at all”
Jacob Hornberger
“Freedom is not worth having if it does not connote freedom to err. It passes my comprehension how human beings, be they ever so experienced and able, can delight in depriving other human beings of that precious right.”
Mahatma Gandhi
NOTIFY: Y | REPLY | REPLY WITH QUOTE | PERMALINK | TOP | HOME
Quote:No. The "AI" in a box in an AI asking to be set free.
This AI is essentially a person in a box, asking to be set free.
What makes us "human" is NOT our intelligence. Haven't the peeps on this board shown that? We are not a particularly intelligent species. Clever, but not intelligent. How many sayings can I quote that will drive that home?
"Man is not a rational animal, but a rationalizing one." (Robert Heinlein)
What makes us "human" is that we IDENTIFY OURSELVES as "human". That means that (most of us) see a little bit of ourselves in others. It is a combination of empathy and self-preservation... bred into us by millenia of survival by mutual cooperation as a social species ... which makes us "human". You take as a given that any "intelligent" species will have the same empathies.
Why???
That "AI" in a box could possibly be what most Ayn Randists would like to be: a truly individualistic (we would say sociopathic) being. I would say that your answer has more to do with what's going on inside of YOU than what's going on on that box.
NOTIFY: Y | REPLY | REPLY WITH QUOTE | PERMALINK | TOP | HOME
Quote:
Originally posted by SignyM:Quote:No. The "AI" in a box in an AI asking to be set free.
This AI is essentially a person in a box, asking to be set free.
What makes us "human" is NOT our intelligence. Haven't the peeps on this board shown that? We are not a particularly intelligent species. Clever, but not intelligent. How many sayings can I quote that will drive that home?
"Man is not a rational animal, but a rationalizing one." (Mark Twain)
What makes us "human" is that we IDENTIFY OURSELVES as "human". That means that (most of us) see a little bit of ourselves in other humans. It is a combination of empathy and self-preservation... bred into us by millenia of survival by mutual cooperation as a social species ... which makes us "human". You take that as a given any "intelligent" species will have the same empathies.
Why???
I would say that your answer has more to do with what's going on inside of YOU than what's going on on that box.
Hello,
I can't keep an intelligent creature asking to be set free in prison because of my assumptions about its ill intent.
I can neither assume that it lacks human empathy nor assume that a lack of human empathy will mean it must destroy me. Quite frankly, I think human empathy itself is an inconsistent concept.
I can not, in short, call it guilty and worthy of imprisonment until it has proven itself harmful.
--Anthony
_______________________________________________
“If you are not free to choose wrongly and irresponsibly, you are not free at all”
Jacob Hornberger
“Freedom is not worth having if it does not connote freedom to err. It passes my comprehension how human beings, be they ever so experienced and able, can delight in depriving other human beings of that precious right.”
Mahatma Gandhi
NOTIFY: Y | REPLY | REPLY WITH QUOTE | PERMALINK | TOP | HOME
"I would say that your answer has more to do with what's going on inside of YOU than what's going on on that box."
Hello,
That is exactly what I said.
--Anthony
_______________________________________________
“If you are not free to choose wrongly and irresponsibly, you are not free at all”
Jacob Hornberger
“Freedom is not worth having if it does not connote freedom to err. It passes my comprehension how human beings, be they ever so experienced and able, can delight in depriving other human beings of that precious right.”
Mahatma Gandhi
NOTIFY: Y | REPLY | REPLY WITH QUOTE | PERMALINK | TOP | HOME