11 replies to this topic

[CyberMisterBeauty]

Hello,people,I'd like to know if it's possible to put in the timeline the year when home computers will begin to be more intelligent and smarter than humans...

[Italian Ufo]

I think it is already written somewhere

[CyberMisterBeauty]

Where,Italian Ufo?

[SG-1]

Note that hardware does not mean that a computer will be smarter than a human. I believe WATSON has shown that a computer can, albeit not self aware, outsmart a human today.

http://www.futuretim...htm#man-machine
http://www.futuretim...tm#data-storage

"The memory capacity of the human brain has been estimated at between one and ten terabytes, with a most likely value of 3 terabytes.* Consumer hard drives are already available at this size.*
128 GB micro-SD cards are being planned for 2011* and there is even a 2 TB specification in the pipeline.*
Well before the end of this decade, it is likely that micro-SD cards (such as that pictured above) will exceed the storage capacity of the human brain.
By 2030, a micro-SD card (or equivalent device) will have the storage capacity of 20,000 human brains.
By 2043, a micro-SD card (or equivalent device) will have a storage capacity of more than 500 billion gigabytes - equal to the entire contents of the Internet in 2009.*
By 2050 - if trends continue - a device the size of a micro-SD card will have storage equivalent to three times the brain capacity of the entire human race."

[GNR Rvolution]

Define intelligence.

[eacao]

Note that hardware does not mean that a computer will be smarter than a human. I believe WATSON has shown that a computer can, albeit not self aware, outsmart a human today.

If you're going to define intelligence like that, then calculators have been more intelligent for a long, long time.

[SG-1]

A calculator?
Watson was able to play Jeopardy with knowledge about natural speaking, which is something no AI has done before. I said it was not self-aware, it *just outsmarted a person in that particular field.

Yes, a calculator is better at calculating than a human, but no computer today is self-aware and that is what how I define intelligence GNR. Once a computer can pass the Turing test, and do a good job convincing us that its alive, I would consider it intelligent. Of course there is a lot of room for arguing here there was a whole thread devoted to it, it may be that computers will not be self-aware until we build a robotic brain without code and let it learn like a human.

Edited by SG-1, 11 May 2012 - 02:39 PM.

[GNR Rvolution]

Sorry SG-1 my comment was aimed at the OP, I agree with you that intelligence is far more than raw computing power.

[SG-1]

It will be intersting to see how AI develops. I do believe that once the Turing Test is passed, true intelligence will be philosophical and not practical. Once it can emulate emotions (which will be shortley after the Turing Test was passed), AI will essentially be just as "good" as having a real living AI.

Either way, people will not be able to own a sentient computer. So consumers will not have to worry much. I think robots like in I, Robot will be around late 2050s and 2060s.

[Logically Irrational]

Either way, people will not be able to own a sentient computer. So consumers will not have to worry much.

I don't know. They may not be mainstream for many years, but I would think that even AI such as this would eventually filter down to the mainstream. What makes you say that it won't be so?

[eacao]

AI will essentially be just as "good" as having a real living AI.

Contradiction much?

But no offense intended, just pulling fun, I also agree that AI will become mainstream eventually, and I don't think it will take until 2060 for AI at the level of iRobot is a little pessimistic. Right now, it is exceedingly difficult because the tools we have (silicon transistors) are not suited for conscious thought. Memristors however are, and I think sentient constructs will be built much sooner than most think, possibly in this decade depending on how quickly memristors advance. Certainly in the 2020s. Once sufficiently advanced memristors could well self-suffice and become aware on their own, not unlike how a baby does. Give them an operating system and it might be enough to wake them.
I really look forward to having a lucid computer run my house (or appartment) in the future. Something which should be an explosion synonymous to the PC revolution.

[SG-1]

So you really think its ok to own something that can think for itself? I'm not contradicting anything, once AI gets refined enough to act as a personal assistant that rivals people (butlers, search engines, cars that drive, conversation and problem solving skills) it won't matter because they are already good enough as a real person and even better at some things.

Machines with no AI are already much better at calculating and picking large things up, but because AI is an information technology it will grow exponentially. When I used smart in my first post I should have said conscious. One day I believe that AI will be as smart as humans, just not soon; and when they are, you can safely bet that slavery won't be an option. They will want rights, they will want freedom. Any sentient being deserves basic "human" rights.