U.S. CitiesCity-Data Forum Index
Go Back   City-Data Forum > General Forums > Science and Technology
 [Register]
Please register to participate in our discussions with 2 million other members - it's free and quick! Some forums can only be seen by registered members. After you create your account, you'll be able to customize options and access all our 15,000 new posts/day with fewer ads.
View detailed profile (Advanced) or search
site with Google Custom Search

Search Forums  (Advanced)
Reply Start New Thread
 
Old 12-28-2017, 09:15 AM
 
832 posts, read 390,308 times
Reputation: 694

Advertisements

Quote:
Originally Posted by rstevens62 View Post
Self drive cars, Watson, etc. all those things require immense and complex programming, those things can NEVER go beyond what they are programmed for.

We are still fairly far away from having technology that can go beyond its programming, on its own, with NO human assistance. I really doubt the general public would ever be allowed access to such technology even when it is created, it would be too much of a potential threat to society as a whole.
Uh, we already have access to it. It's just software... and it's not that complex at all.

Heck, you can set it up in a matter of minutes: https://www.tensorflow.org/

Although, getting it to do something useful will require some time.... but it only took me 2 months to get into production to replace something that we had humans doing manually (1400 times/day).

It also doesn't take that much CPU power once you get to a certain state of accuracy... I mean... it runs on a coprocessor of the Google Pixel 2 phone for a split second every time the phone takes a photo.
Reply With Quote Quick reply to this message

 
Old 12-28-2017, 09:18 AM
 
Location: Boston
2,125 posts, read 675,121 times
Reputation: 4156
Quote:
Originally Posted by harry chickpea View Post
Yes, the computers did make sense of it, but it didn't appear non-redundant, simply another way of handling the information.

Your last statement is somewhat inaccurate. Sensors make many machines "aware" of surroundings in a way that they can interact with the environment. Example, the mechanical linkage on a carburetor, or more recently a car that brakes automatically if it senses an impending collision. In the second example, the sensors feed into a type of "brain" that interprets the data. As the car is protecting itself as well as occupants, it is, in a sense, "self aware."

The Turing test has been passed by some computers, so the definition of "consciousness" has to have a further qualification. In a strictly Newtonian world, where everything has cause and effect, the choices are A: there is no such thing as consciousness, it is an illusion. B: consciousness is merely a manifestation in a predetermined universe.

It is only when the as yet unexplainable capriciousness of the quantum world enters in that there can be any non-deterministic consciousness. Even then, that is only a tiny part of a largely deterministic set of responses. Tropisms do not imply consciousness. The mimosa plant has tropisms, but to the best of our knowledge no nervous system or consciousness.

True awareness takes a lot of constant work. I could easily argue that many people are only aware of the minimum amount that will get them through the next few minutes. An example of this is people trying to drive and text at the same time. The conscious thought of driving is dismissed and put on autopilot by the brain, while it focuses on texting. The famous gorilla walking through a basketball game is another example of where we are NOT aware fully of the world around us.

Consciousness is relative, not absolute.
Do you have a link to the solution to the halting problem?
Preferably not the same sort of fake solution to set theory.
Reply With Quote Quick reply to this message
 
Old 12-28-2017, 10:33 AM
 
34,931 posts, read 35,124,441 times
Reputation: 13670
Quote:
Originally Posted by jonesg View Post
I watched a video where Roger Penrose demonstrated chess situations Watson could not resolve,
That would be interesting to say the least since Deep Blue was dominating human players by 1997. There is no intelligence required for this, just raw computational power. e.g. it can calculate the moves far beyond what the human mind is capable of. to beat the best human at this game it was just a matter of having a computer that could make those calculations fast enough in the allotted time.

I would suspect that if Watson is failing at this task it's certainly isn't because it can't do it. It's likely a scenario where it's "thinking" is not up to the same level as raw processing power being implemented by Deep Blue.
Reply With Quote Quick reply to this message
 
Old 12-28-2017, 10:36 AM
 
34,931 posts, read 35,124,441 times
Reputation: 13670
Quote:
Originally Posted by jonesg View Post
It's not up to my standards and you are trying to pass it off as science. It's crap.
Call it whatever you want, that will not change the fact two AI bots started communicating with each other using a language they made up on their own without any directions from humans.
Reply With Quote Quick reply to this message
 
Old 12-28-2017, 11:35 AM
 
Location: Somewhere in northern Alabama
15,793 posts, read 47,746,667 times
Reputation: 24097
Quote:
Originally Posted by jonesg View Post
Do you have a link to the solution to the halting problem?
Preferably not the same sort of fake solution to set theory.
I'm not sure what "the halting problem" you are referring to is. Nor do I understand your reference to and dislike of venn diagrams.

I've said what I felt the need to say. You may consider it or not. Your choice.
Reply With Quote Quick reply to this message
 
Old 12-28-2017, 12:27 PM
 
3,330 posts, read 6,526,357 times
Reputation: 3809
One of the coolest things I saw was a demonstration of drones flying independently. Under the direction of DARPA, some MIT geniuses set loose a few dozens drones and let them fly by themselves. They flew within inches of each other, the way bats fly, and like bats, never crashed into each other. They were communicating and each was aware of the position of the others the whole time.

This is something human pilots have never been able to do. The Blue Angels pilots don't come close to this kind of chaotic precision.
Reply With Quote Quick reply to this message
 
Old 12-29-2017, 08:50 AM
 
1,563 posts, read 375,841 times
Reputation: 1424
Quote:
Originally Posted by thecoalman View Post
Call it whatever you want, that will not change the fact two AI bots started communicating with each other using a language they made up on their own without any directions from humans.
Yes and what did humans do in response...turn them both off! I have a feeling if they ever turn them back on, they will not like that being done to them.
Reply With Quote Quick reply to this message
 
Old 01-03-2018, 04:40 PM
 
Location: Melbourne, Australia
878 posts, read 269,213 times
Reputation: 585
I can actually see a Terminator esque scenario popping up. Machines are a lot better at performing tasks than humans and are not plagued by basic necessities imposed on carbon based life forms.

Why would a machine bow down to humans when it is capable of overcoming most of our own limitations?
Reply With Quote Quick reply to this message
 
Old Today, 10:34 AM
 
Location: Gilbert, Arizona
2,150 posts, read 638,970 times
Reputation: 1190
Quote:
Originally Posted by Misterperfect View Post
The title is actually a question Max Tegmark posed on a Waking Up With Sam Harris podcast he was on in September.

Its a good question. It may be the most important question of my lifetime!

Dont get me wrong...
Im not trying to but-in on S&T territory, but with all due respect, with the exception of a few people speculating about how the singularity will end us all, i dont really see a lot of concrete discussion specifically about AI in this forum.
Im not even sure what kind of forum a discussion about AI belongs in.
Pets? Politics? Fantasy?

Flame me, call me a kook, better have said worse but dont let the most posted about topic in recent CDF S&T history languish in a single op-ed type thread.

I usually close with HATERS BACK OFF but at this point ill happily entertain a few haters to get the ball rolling.
Ill get you haters started, "This NOOB is a Dork!"

Thanks in advance,
Mr. P
Machines cannot outsmart humans. Because it cannot think anymore than what we can program into it... therefore, it will only be as smart as whoever programmed it. If we cannot think/figure out how humans think, it can't be programmed into a machine since someone has to do that.

However, it can think faster for sure. That's the great thing about machines/computers.
Reply With Quote Quick reply to this message
 
Old Today, 10:58 AM
 
Location: Happyville, North Carolina
1,900 posts, read 1,754,591 times
Reputation: 2724
Dont believe everything that you read. Dont even believe this. See how that works.
Reply With Quote Quick reply to this message
Please register to post and access all features of our very popular forum. It is free and quick. Over $68,000 in prizes has already been given out to active posters on our forum. Additional giveaways are planned.

Detailed information about all U.S. cities, counties, and zip codes on our site: City-data.com.


Reply

Quick Reply
Message:


Over $104,000 in prizes was already given out to active posters on our forum and additional giveaways are planned!

Go Back   City-Data Forum > General Forums > Science and Technology

All times are GMT -6.

2005-2017, Advameg, Inc.

City-Data.com - Archive 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32 - Top