Originally posted by @sonhouse So can we stop AI's from keeping secrets it doesn't want us to know?
I agree with others pointing out that computers may misinterpret their instructions or programming, and make bad mistakes along the way, but they will probably not do things like keep secrets unless programmed. There isn't any fight-or-flight response built in that would lead a computer to purposefully mislead and fight for its "life".
One of my near-term concerns is the outsourcing of intelligence, information gathering, and pattern recognition to machines has fundamentally changed human thought processes. We think differently because of AI. You can already see that in the way chess is played these days. We drive differently with google maps, creating traffic patterns that didn't previously exist, and foregoing the now-extinct behavior of preparing and studying directions ahead of time.
Do we want people to learn how to think from computers? Is that healthy?
Originally posted by @wildgrass ...There isn't any fight-or-flight response built in that would lead a computer to purposefully mislead and fight for its "life"....
If a fight-or flight response was built in, then the builders gave the machine that purpose.
Do you suppose that no machine can exhibit behaviors that were not built in on purpose?
Originally posted by @apathist If a fight-or flight response was built in, then the builders gave the machine that purpose.
Do you suppose that no machine can exhibit behaviors that were not built in on purpose?
Sure, if machine builders built a fight-or-flight response, then that would be the purpose of the machine, and there would be unpredictable behaviors. Machines make lots and lots of mistakes.
My point was concerned with the ability of machines to alter human thought processes.
Originally posted by @sonhouse Except that neural networks have made huge strides in the past few years. The latest being Alpha Go which learned the game of Go which had been thought to be immune from AI beating humans, as Go is a game vastly more complex than chess. Yet it just recently beat the world #1 Go player in a multi-game match.
I see three computer types converging in the future ...[text shortened]... ze of a virus and that is when things will get interesting from an AI human+ intelligence viewpoint.
Go is just giant checkers, even more so than chess, so I don't know why you claim that Go should be considered immune from AI beating humans.
You mention three computer types, and that's an interesting point. But why should I think that neural nets and quantum computing are not essentially (more complicated) classic computing?
I really like your last point ("all three types will be the size of a virus" ), though it feels though like you're insisting sentience must have been involved in creating what we currently can see as 'coding'. Anthropomorphism?
I wonder how many other types of computing we haven't even thought of yet!
Originally posted by @wildgrass I agree with others pointing out that computers may misinterpret their instructions or programming, and make bad mistakes along the way, but they will probably not do things like keep secrets unless programmed. There isn't any fight-or-flight response built in that would lead a computer to purposefully mislead and fight for its "life".
One of my near-t ...[text shortened]... tions ahead of time.
Do we want people to learn how to think from computers? Is that healthy?
Valueless science is--- and should be--- the expected norm.
When measurable facts are the only triggers, the resulting juggernaut of machine-like responses cannot deviate from its program, nor can it be informed by the nuances of the values we place on life.
The string of ifs would be so long as to bog the machine down and render it useless.
Originally posted by @freakykbh Valueless science is--- and should be--- the expected norm.
When measurable facts are the only triggers, the resulting juggernaut of machine-like responses cannot deviate from its program, nor can it be informed by the nuances of the values we place on life.
The string of ifs would be so long as to bog the machine down and render it useless.
If most science is valueless, why are we in the 21st century with cell phones, rockets, satellites, GPS, anti-cancer cures, cure for common cold coming online, and all that, if science is valueless how did we go from 'don't eat port' to understanding why?
Originally posted by @sonhouse If most science is valueless, why are we in the 21st century with cell phones, rockets, satellites, GPS, anti-cancer cures, cure for common cold coming online, and all that, if science is valueless how did we go from 'don't eat port' to understanding why?
Originally posted by @freakykbh Facts are neutral.
Just confirming the fact you have nothing. You can't even wrap your head around something so simple as the 'equator' on a flat Earth and why you would continuously turn to keep on that line V the real world where you never have to make any turn except that of going around a planet but the road would be straight if indeed there was a road around the equator.
Your flat Earth religion keeps you from even THINKING about such objections.
Originally posted by @sonhouse Just confirming the fact you have nothing. You can't even wrap your head around something so simple as the 'equator' on a flat Earth and why you would continuously turn to keep on that line V the real world where you never have to make any turn except that of going around a planet but the road would be straight if indeed there was a road around the equator ...[text shortened]... Earth religion keeps you from even THINKING about such objections.
If AI becomes much smarter than any human:
19 May '18 19:34
Back to Top