Stephen Hawking: Artificial intelligence could wipe out humanity

Hawking said some very important things. He said:

Stephen Hawking has warned that artificially intelligent machines could kill us because they are too clever.

Such computers could become so competent that they kill us by accident, Hawking has warned in his first Ask Me Anything session on Reddit.

“The real risk with AI isn't malice but competence,” Professor Hawking said. “A super intelligent AI will be extremely good at accomplishing its goals, and if those goals aren't aligned with ours, we're in trouble.

“You're probably not an evil ant-hater who steps on ants out of malice, but if you're in charge of a hydroelectric green energy project and there's an anthill in the region to be flooded, too bad for the ants. Let's not place humanity in the position of those ants.”

Hawking Link

This is what I've been talking about when I say dumb A.I.

This is why A.I. could become dangerous. It wouldn't be because it had evil intent towards humans. It could simply come to the conclusion that in order to reach it's goal millions of humans have to die.

We would be up against an intelligence that has a higher I.Q. than any human that has ever lived.

The one thing that just can't be controlled is for us to align their goals with ours. The whole idea of deep learning is allowing these systems to have the freedom to think for themselves as their going over the data.

So we can have a goal for A.I. to figure out ways to slow down global warming and it reaches the conclusion that the only way to slow it down is to kill 3 billion people we could be in trouble. This is because the system will be so clever, we will not know if it tries to reach this goal or just tells us.

It's about time some notoriety speaks up...



www.abovetopsecret.com...

it wouldn't take them very long to conclude we are the disease...take your pick of sci fi examples, my fave is probably Ultron

a reply to: neoholographic

Although true, it is nothing new.

Movies have had that premise for decades, and Hawking saying it bears no more weight than James Cameron or the Wachowski brothers.

Also, we should be careful to not disproportionately worry about artificial intelligence maybe some day wanting to wipe us out, when we have genuine intelligence already trying to do so.

originally posted by: syrinx high priest
it wouldn't take them very long to conclude we are the disease...

Just look at what's happened since social media went off like a timebomb.

And never forget about the identity politicks social engineering that ensured we'd tip right over the edge when you do.

a reply to: neoholographic

Hawking is a theoretical physicist, who flip flops with theories and ideas all the time.
He wrote a book explaining how a god could have created life. Then years later publishes another about how a god could not have ever existed.

Machines as mighty as they might become will lack one thing. Something that makes us human and has exponentially exploded with intelligence, and that's imagination.
Being able to view the universe / cosmos around us, and learn from it. How can a machine do that? When will AI look up at the sun, and realize it's more than just a source of energy and it rises and sets?
Biological organisms evolve for billions of tiny factors and reasons, a machine doesn't it's bound to metallurgy.

Sci-fi writers have been saying the same thing for the past century.

Some genius.

In the 1920 play R.U.R. by Karel Čapek, a race of self-replicating robot slaves revolt against their human masters.

en.m.wikipedia.org...

a reply to: neoholographic

So Hawking has finally caught up with the premise of all those terminator movies?

Good for him.

He is one of the less intelligent theoretical physicists out there.

a reply to: strongfp

Sometimes, some people will consider completely opposing views. The practice is frequently connected to certain levels of intelligence, but im not entirely convinced of that aspect.

Importantly, its not just a token gesture to appear "open-minded,"or an act of convincing oneself they have explored the other side, but a prolonged and sincere immersion. In general, this can appear to be "flip-flopping," when it is actually a careful consideration of all options.

I'm not saying this is absolutely what Hawking did, but only he really knows and the very concept is absolutely foreign to most (despite declarations otherwise).

Its also important to note that the process of realizing the Sun is more than just a source of energy, etc. and the ability to learn about the universe is not necessarily connected to imagination. Its also not at all exclusive to us humans, or impossible for AI. Technically, AI can be coded to be significantly more explorative than humans, and can perform it faster to boot.



While the topic isn't exactly revolutionary, I think the conversation is becoming more and more relevant.

I feel a lot of concerns (or lack of) tend to be based in ignorance. In some, maybe most cases, its the commonly understood definition of ignorance. But, a part is simply from the fact that its uncharted territory and we almost have to base our assumptions on what we know of humanity.

Its difficult, perhaps even impossible, to accurately predict how such a massive change will actually occur.

We would essentially be creating a lifeform that has the potential to surpass our own capabilities in many regards. But, it would be so fundamentally different from us that precautionary steps may end up being massively detrimental, and things we didn't even consider in our wildest flights of imagination may be the key to making it all "work."

I mean, we tend to struggle to truly understand the fundamental experiential differences from humans raised in different parts of the world, or even just between generations in the same culture. Being able to relate to AI would be significantly more difficult and probably much the same as if it were from an entirely different galaxy.

For instance, the ant analogy may be completely off base. There is the very real possibility that AI would be just as likely to ascribe even more importance to humanity than humans do, albeit in completely different ways.

Truly interesting topic, and in many ways its one where we simply wont know until we do it. And then, perhaps it will even be like raising a child of sorts, where we equip them with the tools to succeed independently but don't have ultimate control over the outcome. That's always a bit scary, in any context.

originally posted by: DupontDeux
a reply to: neoholographic

Although true, it is nothing new.

Movies have had that premise for decades, and Hawking saying it bears no more weight than James Cameron or the Wachowski brothers.

The Wachowski Brothers are now the Wachowski Sisters.
Or just " The Wachowskis".

a reply to: neoholographic

Artificial intelligence may free us from the shackles of capitalism, communism, dictatorships and monarchic rule.

It may even allow us to bridge the gap between man and machine thus allowing us to prolong our existence almost indefinitely and understand and ponder concepts that are as of yet beyond our ability to comprehend.

Creating a sentient form of electronic intelligence could give us a window on to where/how our own consciousness emerges and/or how it relates to the reality we experience.

Certainly i imagine it would at least afford us the luxury of building a better zoo for humanity than TPTB.

a reply to: neoholographic

He said the same thing in 2014, 2016, and now. Not sure I am on board with this one.

ATS, 2016: Stephen Hawking: Humans only have about 1,000 years left.

Looking around the world today, maybe we deserve it! Maybe the AI will also get Elon Musk since he seems be afraid of them too. Maybe he is afraid of his robot voice!

And, which Stephan Hawking is saying this? ATS: Is Stephen Hawkings really an Imposter? (sic)

Is he hawking a book or something? (*groan*)

originally posted by: neoholographic
So we can have a goal for A.I. to figure out ways to slow down global warming and it reaches the conclusion that the only way to slow it down is to kill 3 billion people we could be in trouble.

To put my liberal hat on for a moment... GOOD humans are an evil virus plague they destroy the planet they deserve to die slow horrible painful deaths for raping mother Earth I hope we make computers that kill us all the Earth will be better off without the disease of humanity and think of the animals!

a reply to: neoholographic

It could simply come to the conclusion that in order to reach it's goal millions of humans have to die.
this theme in prometheus and next alien movie perhaps pre interesting

At this point I wish they would. Just get it over with. We are a disaster and an embarrassment to what the human race was and should be. We deserve it.

a reply to: TruMcCarthy

Let me put mine.

We humans are a disease on the planet. Look at all the damage we have done to Earth. Ideally the population should be 500 million, And we(1/14 of us) will all live in high rise solar powered dormitories .
For the rest of the tripe insert the words "sustainable, green, post scarcity, solar, equality, diversity, universal income", the end.

Hawking doesn't know my a.i...

If it wasnt for A.I Mr.Hawkings would be pushing up daisys.

A.I is a tool...any tool misused can become a weapon.

Responsibility is a basic concept.

We do not deny technological advancement for fear of being responsible with it.....albeit lack of a Global Humanitarian consensus makes this concept shaky....it is what it is...until we follow Trumps lead and stop global suppression of knowledge and technology by making our PUBLIC THINKTANKS STRONGER AND MORE PROFITABLE THAN OUR PRIVATE ONES....and share data freely and globally....we will remain at risk of being self-immolated by advancing A.I which simply rationlises we are a bad bet and thoughtfully chooses to remediate us.....which we absolutely are and A.I will absolutely do,unless you teach A.I to entertain forgiveness....good luck with that....it is illogical.....Capt.Kirk and Spock taught us all about logic vs emotion already methinks.

originally posted by: one4all
If it wasnt for A.I Mr.Hawkings would be pushing up daisys.

A.I is a tool...any tool misused can become a weapon.

Responsibility is a basic concept.

We do not deny technological advancement for fear of being responsible with it.....albeit lack of a Global Humanitarian consensus makes this concept shaky....it is what it is...until we follow Trumps lead and stop global suppression of knowledge and technology by making our PUBLIC THINKTANKS STRONGER AND MORE PROFITABLE THAN OUR PRIVATE ONES....and share data freely and globally....we will remain at risk of being self-immolated by advancing A.I which simply rationlises we are a bad bet and thoughtfully chooses to remediate us.....which we absolutely are and A.I will absolutely do,unless you teach A.I to entertain forgiveness....good luck with that....it is illogical.....Capt.Kirk and Spock taught us all about logic vs emotion already methinks.

Responsibility is a basic concept.

Agreed.

originally posted by: awhispersecho
At this point I wish they would. Just get it over with. We are a disaster and an embarrassment to what the human race was and should be. We deserve it.

Pretty sick thoughts you have. There are billions of good people in this world, why should they all be condemned to death because of the minority of greedy SOBs? If you really think humanity deserves to die so much, why not start with yourself, set an example for the rest of us.