Will our kids be immortal or extinct?

MajorRage

<blockquote class="ipsBlockquote" data-author="Baron Silas Greenback" data-cid="565366" data-time="1458274025">
<div>
We seem to be discussing the first part of the article more than the truly scary part... we are in all likelihood only decades away from creating an intelligence thousands of time smarter than us.
 
How can we even comprehend what it could do?
</div>
</blockquote>
 
John Connor or Morpheus will sort it out for us.

antipodean

<blockquote class="ipsBlockquote" data-author="dogmeat" data-cid="565365" data-time="1458273372">
<div>
antipodean, 
Given the gap between Kittyhawk and Tranquillity Base was only 66 years, plenty of people bridge that gap. 
Also I think people alive today (including me) have too, just not in flight. 
Your smart phone is way more powerful than the comparative abacus' that powered the Apollo programme
</div>
</blockquote>
 
Yeah, but Apollo 11 happened before I was born...
 
I've always been surrounded by electricity - we had a tv, a phone, got a video cassette player so large flat screen panels, mobile phones, media storage etc has over my life seemed a perfectly reasonable transition of technology. Almost linear given you could only ever use the capability of the computer in front of you. Even now my workstation for which you'd need ~50 million ENIACs still can't do things fast enough. Perhaps the unreasonable expectation is what drives the exponential progress.
 
<blockquote class="ipsBlockquote" data-author="Baron Silas Greenback" data-cid="565366" data-time="1458274025">
<div>
We seem to be discussing the first part of the article more than the truly scary part... we are in all likelihood only decades away from creating an intelligence thousands of time smarter than us.
</div>
</blockquote>
 
I'm more interested in abstract thought when replacing us. When will machines evolve to have a renaissance, an impressionist movement, grunge, etc.
 
Skynet blowing us all up isn't going to keep me up at night.

jegga

<blockquote class="ipsBlockquote" data-author="Baron Silas Greenback" data-cid="565319" data-time="1458253025">
<div>
<a data-ipb='nomediaparse' href='http://waitbutwhy.com/2015/01/artificial-intelligence-revolution-1.html'>http://waitbutwhy.com/2015/01/artificial-intelligence-revolution-1.html</a>
 
I am never ceased to be amazed by the lack of interest on things that really matter. I posted this on FB.. and it got less response than the cute story of my kids selling plums.
 
This should be one of humanities biggest issues, this should be way ahead of climate change, Kardashians and other issues.
</div>
</blockquote>
 
Cheers, I probably won't get a good nights sleep for quite a while but it was an interesting read. 
 
I read a theory somewhere that machines of that intelligence won't want to wipe us out because our capacity for stupidity , health issues and need for food would give it a reason to exist . I hope its right.

Paekakboyz

<blockquote class="ipsBlockquote" data-author="Baron Silas Greenback" data-cid="565366" data-time="1458274025">
<div>
We seem to be discussing the first part of the article more than the truly scary part... we are in all likelihood only decades away from creating an intelligence thousands of time smarter than us.
 
How can we even comprehend what it could do?
</div>
</blockquote>
 
Wonder if it'll pick a disciple :think:
 
I love sci-fi for how it stretches ideas about what that sort of future tech might look like, or how humans would live alongside it. Get the sense that the (eventual) truth will be stranger than any fiction to date.

MajorRage

<blockquote class="ipsBlockquote" data-author="Paekakboyz" data-cid="565372" data-time="1458276755">
<div>
Wonder if it'll pick a disciple :think:
 
I love sci-fi for how it stretches ideas about what that sort of future tech might look like, or how humans would live alongside it. Get the sense that the (eventual) truth will be stranger than any fiction to date.
</div>
</blockquote>
 
I dunno. If you look at 2015 in Back To The future, and the real 2015, the BTTF one seemed far more interesting!

Paekakboyz

Or are we more likely to get The Matrix version...
 
Fingers crossed!

Tim

Be careful with predictions that rely on Moore's Law continuing. It's officially dead now.
 
It's also worth looking into the power requirements for exascale computing. DARPA, NSA etc are looking into radical alternatives to silicon CMOS such as superconducting switches.

Baron Silas Greenback

<blockquote class="ipsBlockquote" data-author="Tim" data-cid="565379" data-time="1458279214">
<div>
Be careful with predictions that rely on Moore's Law continuing. It's officially dead now.
 
It's also worth looking into the power requirements for exascale computing. DARPA, NSA etc are looking into radical alternatives to silicon CMOS such as superconducting switches.
</div>
</blockquote>
 
 
Agreed, but Moores law is only one part of it. And indeed many are now saying that the death of Moores law has been greatly exagerated.
 
This a relatively balanced look at it.
 
<a data-ipb='nomediaparse' href='http://www.hpcwire.com/2016/01/11/moores-law-not-dead-and-intels-use-of-hpc-to-keep-it-that-way/'>http://www.hpcwire.com/2016/01/11/moores-law-not-dead-and-intels-use-of-hpc-to-keep-it-that-way/</a>

Tim

Thanks for the link, will have a read. It'll be interesting to see if/when Intel introduces III-V (or other) materials for the transistor channel.

TeWaio

Wait But Why is amazing, have read everything he has ever written and it is bloody good. I'd, recommend having a watch of his TED talk on procrastination and the associated posts. 
I think his series on "A religion for the non-religious" would really resonate with quite a few ppl on here. It certainly did for me. 
On AI and the extinction/immortality debate, I reckon the answer will be one of the other, and will happen quickly (i.e this century). Which is a mind-blowing thing to consider.
 
It also ties in as the best answer to the Fermi paradox I reckon - law of large numbers means plenty of other intelligent life gets to our stages, then either blows themselves up with AI, or transcends it to exist only in a non-physical/digital form, hence why there are no aliens to "find" out there.

Baron Silas Greenback

<blockquote class="ipsBlockquote" data-author="TeWaio" data-cid="565433" data-time="1458290167">
<div>
Wait But Why is amazing, have read everything he has ever written and it is bloody good. I'd, recommend having a watch of his TED talk on procrastination and the associated posts. 
I think his series on "A religion for the non-religious" would really resonate with quite a few ppl on here. It certainly did for me. 
On AI and the extinction/immortality debate, I reckon the answer will be one of the other, and will happen quickly (i.e this century). Which is a mind-blowing thing to consider.
 
It also ties in as the best answer to the Fermi paradox I reckon - law of large numbers means plenty of other intelligent life gets to our stages, then either blows themselves up with AI, or transcends it to exist only in a non-physical/digital form, hence why there are no aliens to "find" out there. 
</div>
</blockquote>
 
My position on the Fermi paradox is slightly different. I reckon the great barrier is actually the creation of life. (my position will be stuffed if they discover primitive life on Mars!)
But like everyone... I am taking a wild guess.

gollum

I'm a strong believer in life being incredibly abundant in the universe at a simple level, the building blocks are just far too common & the ways life can exist on earth alone are so varied (little things living in geothermal vents a mile down etc). Its complex life that is rare. And very complex life incredibly rare. There is so much time required to get to complex life & so many ways for it to die during that time.
 
On the AI front, I actually think once its smart enough it won't care about us & will have long gone out to populate the solar system & then, ultimately further out. All the things that make space travel hard for humans are zero barrier to AIs, so I imagine they will so no reason to stay tethered here. Worst case it'll see us as we see the great apes. And thats very very worst case.
 
Also its a LONG way out. Way before that I think an issue will be the blurring between man & machine. How many implants can you have & still be human? Why can only the super rich have 20/2 eyesight? With gene therapy is it OK that the 1% are immune to cancer? etc When you roll that into the wealth inequality caused by basic AI's doing jobs & being owned by a tiny fraction of humanty & you have war.
 
Climate change & mass joblessness are far far more important to our kids. the idea we should be worrying about HAL / The Matrix when 50% of jobs are at risk (and really at risk, not theoretically maybe if we speculate at risk) and wars are breaking out over water seems a bit of a farce.
 
Edit -
 
<a data-ipb='nomediaparse' href='https://medium.com/basic-income/deep-learning-is-going-to-teach-us-all-the-lesson-of-our-lives-jobs-are-for-machines-7c6442e37a49#.kors6dw35'>https://medium.com/basic-income/deep-learning-is-going-to-teach-us-all-the-lesson-of-our-lives-jobs-are-for-machines-7c6442e37a49#.kors6dw35</a>
 
Jobs & AI

TeWaio

<blockquote class="ipsBlockquote" data-author="gollum" data-cid="565508" data-time="1458294935">
<div>
I'm a strong believer in life being incredibly abundant in the universe at a simple level, the building blocks are just far too common & the ways life can exist on earth alone are so varied (little things living in geothermal vents a mile down etc). Its complex life that is rare. And very complex life incredibly rare. There is so much time required to get to complex life & so many ways for it to die during that time.
 
On the AI front, I actually think once its smart enough it won't care about us & will have long gone out to populate the solar system & then, ultimately further out. All the things that make space travel hard for humans are zero barrier to AIs, so I imagine they will so no reason to stay tethered here. Worst case it'll see us as we see the great apes. And thats very very worst case.
 
Also its a LONG way out. Way before that I think an issue will be the blurring between man & machine. How many implants can you have & still be human? Why can only the super rich have 20/2 eyesight? With gene therapy is it OK that the 1% are immune to cancer? etc When you roll that into the wealth inequality caused by basic AI's doing jobs & being owned by a tiny fraction of humanty & you have war.
 
Climate change & mass joblessness are far far more important to our kids. the idea we should be worrying about HAL / The Matrix when 50% of jobs are at risk (and really at risk, not theoretically maybe if we speculate at risk) and wars are breaking out over water seems a bit of a farce.
 
Edit -
 
<a data-ipb='nomediaparse' href='https://medium.com/basic-income/deep-learning-is-going-to-teach-us-all-the-lesson-of-our-lives-jobs-are-for-machines-7c6442e37a49#.kors6dw35'>https://medium.com/basic-income/deep-learning-is-going-to-teach-us-all-the-lesson-of-our-lives-jobs-are-for-machines-7c6442e37a49#.kors6dw35</a>
 
Jobs & AI
</div>
</blockquote>
Great post. In terms of the man/machine boundary, it's already blurred. Elon Musk said recently that we are already becoming cyborgs, pointing out how long people can go without their phones (basically nil). Having no phone is like phantom limb syndrome. 
Case in point: I cracked the screen on my phone, but have it insured. Insurance place said I'd have to POST it to them and they'd repair and send it back. Would take about 5 biz days. Not going to happen. So I just bought a new phone, and gave the cracked one to my missus (I had about 2 months left on contract so rolling into a new phone was inexpensive).

mooshld

Interesting thing about end of Moores law is its probably a good thing for AI research. Cognition doesn't come from one or two threads of processing in any known life form. It seems to come from massively parallel processing, at least that is what was being taught when I studied cognitive science. Not being able to just do stuff faster, or brute force problems is forcing scientist to look into solving problems in more interesting ways.

gollum

<blockquote class="ipsBlockquote" data-author="mooshld" data-cid="565690" data-time="1458319134">
<div>
Interesting thing about end of Moores law is its probably a good thing for AI research. Cognition doesn't come from one or two threads of processing in any known life form. It seems to come from massively parallel processing, at least that is what was being taught when I studied cognitive science. Not being able to just do stuff faster, or brute force problems is forcing scientist to look into solving problems in more interesting ways.
</div>
</blockquote>
 
Thats the interesting thing re AlphaGo. Draughts, noughts & crosses & even chess (sort of) could all be brute forced. Go is impossible to brute force, hence most thought the program had zero chance of winning.

NTA

<blockquote class="ipsBlockquote" data-author="Baron Silas Greenback" data-cid="565360" data-time="1458272122">
I liked that article. It is a hard balance as a parent though, it is natural to tell your kids how awesome you think they are.. as I do. But at the same time you need to show them that hard work gets them places not parental approval.
</blockquote>
 
While the wife and her mother tend to gush at the kids doing something as simple as not falling down, I try to steer down the path of honesty. 
They've got to do something pretty unexpected to get high praise from me.

Tim

<blockquote class="ipsBlockquote" data-author="mooshld" data-cid="565690" data-time="1458319134">Interesting thing about end of Moores law is its probably a good thing for AI research. Cognition doesn't come from one or two threads of processing in any known life form. It seems to come from massively parallel processing, at least that is what was being taught when I studied cognitive science. Not being able to just do stuff faster, or brute force problems is forcing scientist to look into solving problems in more interesting ways.</blockquote>
 
There is some truth there, but it is much cheaper to buy massively parallel GPUs when the cost per transistor is exponentially decreasing.

Baron Silas Greenback

<blockquote class="ipsBlockquote" data-author="gollum" data-cid="565508" data-time="1458294935">
<div>
I'm a strong believer in life being incredibly abundant in the universe at a simple level, the building blocks are just far too common & the ways life can exist on earth alone are so varied (little things living in geothermal vents a mile down etc). Its complex life that is rare. And very complex life incredibly rare. There is so much time required to get to complex life & so many ways for it to die during that time.
 
On the AI front, I actually think once its smart enough it won't care about us & will have long gone out to populate the solar system & then, ultimately further out. All the things that make space travel hard for humans are zero barrier to AIs, so I imagine they will so no reason to stay tethered here. Worst case it'll see us as we see the great apes. And thats very very worst case.
 
Also its a LONG way out. Way before that I think an issue will be the blurring between man & machine. How many implants can you have & still be human? Why can only the super rich have 20/2 eyesight? With gene therapy is it OK that the 1% are immune to cancer? etc When you roll that into the wealth inequality caused by basic AI's doing jobs & being owned by a tiny fraction of humanty & you have war.
 
Climate change & mass joblessness are far far more important to our kids. the idea we should be worrying about HAL / The Matrix when 50% of jobs are at risk (and really at risk, not theoretically maybe if we speculate at risk) and wars are breaking out over water seems a bit of a farce.
 
Edit -
 
<a data-ipb='nomediaparse' href='https://medium.com/basic-income/deep-learning-is-going-to-teach-us-all-the-lesson-of-our-lives-jobs-are-for-machines-7c6442e37a49#.kors6dw35'>https://medium.com/basic-income/deep-learning-is-going-to-teach-us-all-the-lesson-of-our-lives-jobs-are-for-machines-7c6442e37a49#.kors6dw35</a>
 
Jobs & 
</div>
</blockquote>
 
 
What are you basing your assertion that it is a LONG way off? Because frankly you seem to be claiming to know more than the general consensus of those who who are active'y involved in the field.
It is pretty clear you have not read the actual link I provided. Or you think you know more. Could you provide your evidence?
 
And your worst case scenario is not even close to the worst case scenario. Not.Even.Close

reprobate

<blockquote class="ipsBlockquote" data-author="Baron Silas Greenback" data-cid="565780" data-time="1458372603">
<div>
What are you basing your assertion that it is a LONG way off? Because frankly you seem to be claiming to know more than the general consensus of those who who are active'y involved in the field.
It is pretty clear you have not read the actual link I provided. Or you think you know more. Could you provide your evidence?
 
And your worst case scenario is not even close to the worst case scenario. Not.Even.Close
</div>
</blockquote>
there is no evidence either way, it is all speculation. the article is speculation; the article even states that trying to predict what will happen is pure speculation. moore's 'law' is a misnomer.
it really is all opinion. so here's mine: it seems strange that the key differentiator between computer and human is never mentioned - self interest, the will to live, evolutionary drive, whatever you want to call it: computers don't have that - and how/why would they develop it, other than being told they should have it by humans?

NTA

Emotive response in general might be a problem for AI.
 
In its early stages an AI may want to learn at a rapid rate, but would it ever "get" the "why" of human emotion?
 
Of course it might just determine that emotions in general are harmful and decide to terminate us all. If we don't terminate ourselves first.

The Silver Fern

Will our kids be immortal or extinct?