When you get a chance, you all should catch the free documentary on youtube:
AlphaGo https://www.youtube.com/watch?v=WXuK6gekU1Y
There is a point in the show when you start to really really feel for humanity. AI is kool but I never fathom the emotional and mental toll felt at the highest level of human intellect when AI was used against it.
I weeped several times... it was unexpected.
I fear for the future.
[UPDATE] Many insightfulness here. I just came off reading too much of Jung's work. ;-) And as one commenter stated; I'm just rambling :) No worries. Just wanted you all to know of a well done documentary. Stay safe!!
Calculators have existed for a long time and they crush humans at doing arithmetic. I wouldn't cry if I watched a documentray where a calculator beats the best human at 5-digit-number multiplication. AlphaGo is like a big expensive calculator that's specialised to doing the calculation of how to play Go.
However, having said that and we all know AI will have bugs and will very well take paths we couldn't have imagined. It isn't so comforting to think companies like Palantir making wrong profiles of you and autonomous driving running over people will all be possible. But may be we can delete racism, discrimination, prejudice? yeah.. probably not until humans are left out of the algo and let it self code itself up.
I'm just saying, the future will be interesting when emotions are put aside and people, places and things profiled in a manner totally unconventional today.
It's rather ironic that almost all of man's greatest discoveries were made accidently like vulcanization of rubber or post-it or even antibiotics.
Wondering what kind of future humanity will have where everything is done deliberately and unconditionally. No emotions, accidents or serenity at play.
Coincidently, my greatest finds in life were almost all accidently including the love of my life and my fav food. I would have never met her nor find my fav food if it wasn't for certain unforeseen accidents.
PS> It isn't possible like chess to precalculate Go strategies. The documentary proved that and the game itself dictates that.
My point is that at the current state of technology the difference is so huge that it's almost inappropriate to use the term intelligence to describe both how humans think and what computers do.
How fast you can beat eggs depends on how much transferable practice you got as a teenager.
I think that, on the contrary, finding new challengers with different styles pushes us to explore further, expand our envelope, and grow.
Imagine AlphaGo's impact on people who may not have playing partners. People could be able to enjoy the game whenever they want to and improve and come up with original moves when stressed, just as Lee Sedol came up with a move when stressed.
Clinging onto comfort zones seldom leads to progress or growth.
Well stated. Yes, I cannot agree more that you need challenges like what Sedol faced. He admitted playing AlphaGo made him a better player. I too feel same whenever I get challenged but there is difference of the challenge's virtue depending on the outcome. An emotionally strong deeper understanding stable person will possibly see growth, but the flipside is someone not ready/incapable/not mentally mature enough will probably get their souls crashed.
In martial arts, we always fight against a belt above as one cannot better one self being challenged by stagnant skill level.
One of the main reason why I like coding. Even if similar task, I try a different path to challenge myself.
Hasn't that been true since the dawn of time? I argue that this is how we grow. I often tell friends that I believe that the growth of people and societies is proportional to the number of difficult conversations they've had, either with themselves, others, or as a society.
Educators and societies that shield ego too much produce entitled dipshits in my opinion. Those who try to protect themselves from being hurt are holding onto one end of a rubber band whose other end is stretched further by Certainty, and Certainty will certainly let it go at some point. The longer you hold on to that end of the rubber, the more painful it will be when Certainty releases its end.
>In martial arts, we always fight against a belt above as one cannot better one self being challenged by stagnant skill level.
My Judo instructor did it differently. We'd all be jogging in different directions. He'd pick one randomly. We'd form a random queue from the configuration at the moment we stopped jogging. Champion/Challenger:
0: The person who was picked is (Champion)
1: (Champion) pops someone from the queue (Challenger).
2: (Champion) fights (Challenger)
3: If (Champion) wins then (Challenger) is eliminated
4: Go to 1
5: If (Challenger) wins then (Champion) is eliminated
6: (Challenger) becomes the new (Champion)
7: Go to 1
You end up matching with people of different belts, different compositions, weights, speeds, execution, order. You're exposed to all kinds of strange and at different energy levels.
My fear is that ALL new tech is eventually turned in to a weapon that we as humans use to dominate other humans. AI is used and will be used as a weapon and that scares me for the future.
although I guess one could "code" that ethics like aka Isaac's 3 rules of robotics but how would one do that to an autonomous killing drone? Or an autonomous car that needs to decide to kill one person or several depending on what? that one person is a father of 6 and does humanitarian work vs the other 6 are criminal low lifes? How does one "code" such algo?
Yes AI is a tool but then it wields a car, drone, finances that determines whether you get cancer treatment or not when you're 80.
Yeah, we're entering into a new phase of humanity.....
Funny, these are the exact same philosophical thoughts the Terminator movie series was addressing. Where the AI did become onto itself and became "self aware." Then did the only logical thing: get rid of humans which is the cancer to everything's existence.
Lots existentials to go through. Should we force AI to go through that? Maybe should be a prerequisite before we let it "live."
Isn't this literally how every killer AI is created in fiction? The only way to protect humans is to kill all humans or something along those lines.
If you're going to fear for the future, there are far grander, realistic, and imminent risks to be concerned with.
