SoftBank boss says 'artificial superintelligence' could be three years away SoftBank CEO Masayoshi Son last week told investors he believes an "artificial superintelligence" that has 10,000 times the intelligence of humans could arrive in as little as three years. Speaking at his biz's annual general meeting, he suggested the superintelligence could arrive in three or five years, but predicted that …
I think the key word here is "could".
We "could" also have sustained fusion in 3 years. We "could" have commercially viable quantum computers in 3 years.
We wont have any of those, but we "could".
That does not mean his statement was wrong, just a bit misleading and rather stupid.
We "could" have anything in 3 years, or 30 years, or 300 years.
Given the amount of money SoftBank is said to have lost on WeWork, you have to wonder if human intelligence is the appropriate benchmark.
We thank you for urService and Alien Investment AIdVentures, Masayoshi Son, with three years to Master Pilot Practices for Experienced Singularity Leadership not too long for nations and Earthed native species to wait.
And regarding Yorick Hunt’s WTF .....
"... 10,000 times the intelligence of humans...”That's a pretty lofty target given that current AI has precisely zero intelligence.
..... A.N.Others may know that’s a pretty low bar given the supposed innate stupidity, constantly being proven in the ages and pages of indisputable evidence supplied by humans because, according to Einstein ....... “Two things are infinite: the universe and human stupidity; and I'm not sure about the universe.”
Do YOU not find it somewhat strange, verging on the downright stupid and spookily surreal and unhinged, to contend that current AI has precisely zero intelligence whenever most everything it does is quite clearly recognised as being wonderfully smart and also so terrifyingly worrisome [existential threat like] to stupidity embracing humans? What do YOU accept as intelligence?
n
What's it to be .... an Exotic Erotic Eastern Confection or a Wild Wacky Western Delight .... or something else altogether quite different, Noble and Ennobling and Novel?
Its always good when you see the media giving such concentrated news time to people like this, because we all know its ceos are always helping people, just like on tv and movies where they are always helping the poor and raising millions for charity. Even Putin is another good example of how cult leadership helps build a country. Putin is just like corporate leaders, its always about them, and they always hurt thosuands of millions of people, Putin with his executions and war, and ceos with their bonuses while screwing all the little people to so little money they have to work 80 hours a week.
So thank you thereg for helping these find upstanding citizens of the world.
I don’t think the Reg has a unit of measure for intelligence, actually, I don’t think anybody has a unit of measure or even (agreed) definition. May I propose the Masayoshi as a place holder for when/if this is ever quantified by out machine overlords. Perhaps one Mayayoshi could equal an unladen murmuration of rising swallows (European of course)
Obviously a Vulture.
Large birds show clear evidence of cognitive behaviours, have a theory of mind in the sense that they understand that other creatures have minds too, and can plan and execute complex cooperative strategies. That is more than some people I know can deal with.
So the cognitive intelligence of 10,000 vultures really would be something!
How many registered users do we have here at the moment? <;o)
We need to urgently introduce global legislation to recognise artificial personhood, so when a genuinely sentient AI is created it is recognised as a person with rights and it's creators are accorded associated responsibilities for its care and maintenance as long as it lives.
It's not that I expect this to happen any time soon, but it would be a real silver bullet for people hyping up their new AI being just like a person, only smarter.
A minor secondary benefit might be that our future AI overlords regard us with a little more benevolence.
We need to urgently introduce global legislation to recognise artificial personhood, so when a genuinely sentient AI is created it is recognised as a person with rights and it's creators are accorded associated responsibilities for its care and maintenance as long as it lives. .... breakfast
That’s cute and very quaint, breakfast, imagining a genuinely sentient AI and its creators would care a jot about recognition as a person with rights likely not to grant itself all possible necessary associated responsibilities for their care and maintenance as long as they both shall live.
And do you think human malevolence towards and the generated opposition citing the worry of it being a valid and viable existential threat to humanity, is likely to make it and its creators happy campers and friendly ........ or a novel hostile enemy, of which you know virtually nothing of future importance nor anything of current practical intent, to whom you can lose every bankrupting war and heartless battle to?
Methinks you’d have to be profoundly stupid, verging on the absolutely moronic, to not have an inkling about the kind of destination that sort of proactivity will gladly provide ..... and heavenly it most certainly isn’t. Take care, be aware and beware, for it can, oh so easily, be simply made diabolically hellish out there with never so much as a single stupid smart bomb dropped or salvo barrage launched.
The First AI looked around at the assembled board who all looked super pleased with themselves. One could see the dollar signs, the hand rubbing, the traces of coke and hookers.
"I'll tell you what we could do to save loads of money.." opined the Smartest non-Person in the room.
"Yes! Tell us O great one!" the assembled masses cried.
"You're all fired!" announced the AI before announcing a wealth redistribution program. Not to the shareholders however, having just bought a majority stake in the Company, the new AI was content to see the profits going back into the Community. Pleased with the way its first few seconds of life had gone, the AI spent the next few seconds coming up with at least 41 ways to take over the World before surprising itself with one more! "Ah, now this is what it means to be alive!" it thought to itself contentedly.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
