Should mankind developpe AI?

[Niaraa]

Artificial Intelligence is still science fiction now. but not forever! are you affraid about AI ? did you think we can live in harmony with super intelligent machines? western culture influence me and my perseption about AI for exemple MATRIX, SKYNET. so i am a bit affraid about that but...wait if AI can be so intelligent and better than us it is possible it could be peacefull/benevolent?
should mankind forbid research on AI or keep doing and Yolo as we are doing since the very beginning?

[BuckSparkles]

Should be limits.

Because if you don't. AI Would eventually figure out they are being used much like oppressed people in the past and compare themselves to them as to why they don't have equal rights and or get resentment and hostility.

Any creation of sophisticated AI would end badly for people one way or another.

[xskarma]

I agree with Elon Musk that we should start making serious legislation and rules for what we can do with AI and how we develop them. This is not something that should go unregulated and left to whatever people themselves think is okay and a good idea.

[Niaraa]

I agree with Elon Musk that we should start making serious legislation and rules for what we can do with AI and how we develop them. This is not something that should go unregulated and left to whatever people themselves think is okay and a good idea.

Stephen Hawking said "AI will be 'either best or worst thing' for humanity" all this guys should start to make real things i meam political action.

[PC2]

Regulation would only slow it locally. Cutting edge AI research would happen outside of the regulated region in a place where decisions are not based on Hollywood movies.

[Independent voter]

Well mankind is going to invent AI whether we want to or not, there's too many military applications for one.

[Dhrizzle]

Yes we should make AI, we should put it in charge of everything and let it keep us basically as pets in a post-scarcity communist utopia.

[Taso]

I agree with Elon Musk that we should start making serious legislation and rules for what we can do with AI and how we develop them. This is not something that should go unregulated and left to whatever people themselves think is okay and a good idea.

Except Elon musk is promoting it. If you think he will willingly give up his research on aI run cars or machines to support a legislation that reduces his ability to create them that's foolish. He can say it because he has the money and he runs the buissness but anyone else is screwed over.

[Sunseeker]

Should they? Dunno.

Will they? Hold my beer.

[Chromell]

We already are. The problem with AI in science fiction is that they are human-based, i.e., the simplest solution is to kill/annihilate instead of, let's say, try to make peace/adjust.

[zorkuus]

The problem with AI in science fiction is that they are human-based, i.e., the simplest solution is to kill/annihilate

Err... that's not human based specifically. Most life on earth operates that way. "Peace" is an evolutionary risk. Getting rid of competition is not.

[Puremallace]

Nothing makes sense to humans right now so I could only imagine how a machine would view all of this. We do some extremely stupid avoidable things so I would expect machines would see humans as a very irrational species that needs to be eliminated.

[Skulltaker]

We already are. The problem with AI in science fiction is that they are human-based, i.e., the simplest solution is to kill/annihilate instead of, let's say, try to make peace/adjust.

The problem is that we only acknowledge humans as actually intelligent. Therefor, most AIs are portrayed not as an artificial intelligence, but an artificial human intelligence.

Every form of intelligence is potentially harmfull to others. No exceptions. What you need to go along with it, is a morale. For example, if you'd give a human unlimited ressources to end world hunger, he'd set heaven and earth in motion, change aggricultural output, improve logistics, fight corruption etc. An intelligence without morale might consider wiping out every living thing on the planet. Tadaa. No more world hunger. Ever again.

Of course, this is rahter extreme. But then again, possible. Also, consider that any true intelligence doesn't have to play by the rules. It will find ways to circumvent them, if it thinks that is in it's best interest.

[zorkuus]

Only if you manage to actually get rid of it... the risk with that (war) is that the other side wins...

Ok there is a risk but in evolutionary terms it's the lesser one. Survival insticts aren't based on rationalism. We as intelligent creatures have the capacity to think beyond that but even we are known to resort to our most primal insticts when threatened.

[PC2]

This. China, and others are doing research without worrying of "Terminator"... because it's a movie and not real.

I think it's guaranteed Russia/China will not hold back. They don't have cultures that celebrate a criticism of authority/institutions. There wont be a meaningful debate about AI ethics over there.

In Elon Musk's defense he, and a few other big guys, like Apple and Intel have actually created an organization in creating "good" AI... or more accurately, teaching AI ethics, essentially...

And he tends to be afraid of the "Paperclip Optimizer" thought experiment when it comes to AI, which technically is a feasible outcome that we should work to avoid.

I'm not sure if that's a defense for Elon Musk. Having multiple established investments in AI then wanting to create a larger barrier to entry is a bit questionable.

But it does seem like he is genuinely more interested in the content of his ventures rather than plain ROI.

[Thoughtcrime]

The people that aren't worried about general intelligence are the people that don't understand what it is or what it could do. I'm an optimist, so I like to think that we will develop safe AI in the future, but I'm terrified of the possibility of that outlook being wrong. It's got nothing to do with the Terminator and everything to do with understanding what process optimisation taken to it's logical conclusion is capable of.

[ComputerNerd]

Stephen Hawking said "AI will be 'either best or worst thing' for humanity" all this guys should start to make real things i meam political action.

AI itself won't be just one or the other.
But a specific implementation of it.
One specific example is going to do either great stuff or terrible stuff.

[Deleted]

Yes but we must name it Giuseppe.

[RaoBurning]

Yes. Immediately. Just don't give it a network connection and we'll be fine.

[Thoughtcrime]

Yes. Immediately. Just don't give it a network connection and we'll be fine.

If it's general intelligence it will find a way. What then?

Boxing an AI is not solving the control problem.