Page 5 of 7 [ 104 posts ]  Go to page Previous  1, 2, 3, 4, 5, 6, 7  Next

TallyMan
Veteran
Veteran

User avatar

Joined: 30 Mar 2008
Gender: Male
Posts: 40,061

25 Jul 2014, 5:06 am

Humanaut wrote:
What goals would motivate a computer?


At some point during the development of AI's a programmer would likely embed "prime directives" into a computer; for example seek out more knowledge with an aim to understanding the physical world. Asimov came up with his three laws of robotics which are also a good basis.

In humans those directives evolved as "Self preservation and reproduce" and all our other needs, wants and desires are subordinate to those two prime directives or instincts.

If scientists/programmers ever produce nanites with the ability to reproduce themselves and give them the same two prime directives / instincts then the lid will be well and truly off Pandora's box! Welcome to the world of grey goo.


_________________
I've left WP indefinitely.


Humanaut
Veteran
Veteran

User avatar

Joined: 17 Jul 2014
Age: 53
Gender: Male
Posts: 4,390
Location: Norway

25 Jul 2014, 5:16 am

TallyMan wrote:
Welcome to the world of grey goo.

Here it is:

[youtube]http://www.youtube.com/watch?v=_0dYPnui3rM[/youtube]



TallyMan
Veteran
Veteran

User avatar

Joined: 30 Mar 2008
Gender: Male
Posts: 40,061

25 Jul 2014, 5:28 am

^ Excellent video. Funny but also a stark warning about what could really happen at some point in the not too distant future. Can you imagine what would happen with that technology in the hands of Islamic extremists or psychotic dictators? Even a kid with a lab in his parents basement... oops, didn't mean to do that... too late. There is also the possibility of accidents of labs in what are supposedly well regulated environments as evidenced recently by accidental leaks of dangerous man-made viruses (thankfully subsequently contained).


_________________
I've left WP indefinitely.


Humanaut
Veteran
Veteran

User avatar

Joined: 17 Jul 2014
Age: 53
Gender: Male
Posts: 4,390
Location: Norway

25 Jul 2014, 5:32 am

TallyMan wrote:
Can you imagine what would happen with that technology in the hands of Islamic extremists or psychotic dictators?

Yes.

Interesting subject.



TallyMan
Veteran
Veteran

User avatar

Joined: 30 Mar 2008
Gender: Male
Posts: 40,061

25 Jul 2014, 6:03 am

Humanaut wrote:
TallyMan wrote:
Can you imagine what would happen with that technology in the hands of Islamic extremists or psychotic dictators?

Yes.

Interesting subject.


It is interesting to speculate what could happen after release of such self replicating nanites. I wonder if it would mirror the evolution of organic life on earth from simple self-replicating chemicals. In other words, some of the nanites could evolve minor changes due to mutations/faults in their reproduction process or internal blue-plan for their own design. On this basis a slightly more "evolved" nanite could start "eating" other nanites and thus the whole arms race of evolution begins. I wonder if this could lead to the natural evolution of truly intelligent robots e.g. the mechanical equivalents of humans while the rest of the planet has millions of varieties of nanites taking advantage of each type of environmental niche e.g. within the oceans, on the surface taking advantage of sunlight for energy, below the surface using other forms of energy. A whole planet with mechanical "plants", "animals", "bacteria" and so on. I wonder if the intelligent robots would contemplate their origin and invent gods to explain it?


_________________
I've left WP indefinitely.


Humanaut
Veteran
Veteran

User avatar

Joined: 17 Jul 2014
Age: 53
Gender: Male
Posts: 4,390
Location: Norway

25 Jul 2014, 7:02 am

TallyMan wrote:
I wonder if the intelligent robots would contemplate their origin and invent gods to explain it?

This sounds a bit oxymoronic.



TallyMan
Veteran
Veteran

User avatar

Joined: 30 Mar 2008
Gender: Male
Posts: 40,061

25 Jul 2014, 7:19 am

Humanaut wrote:
TallyMan wrote:
I wonder if the intelligent robots would contemplate their origin and invent gods to explain it?

This sounds a bit oxymoronic.


The robots may not be ruled entirely by logic, but also by more primitive emotions related to their evolution, much in the same way mankind has multiple brain layers. Before the intelligent robots discover science or the scientific method they could have as little knowledge about the real world as did people in the middle ages. The nanites from which they originally evolved would typically have no more brain power than today's viruses.


_________________
I've left WP indefinitely.


Humanaut
Veteran
Veteran

User avatar

Joined: 17 Jul 2014
Age: 53
Gender: Male
Posts: 4,390
Location: Norway

25 Jul 2014, 7:38 am

Viruses are replicators, but I don't think they have managed to evolve into more complex forms. Are nanobots any different?



TallyMan
Veteran
Veteran

User avatar

Joined: 30 Mar 2008
Gender: Male
Posts: 40,061

25 Jul 2014, 7:49 am

Humanaut wrote:
Viruses are replicators, but I don't think they have managed to evolve into more complex forms. Are nanobots any different?
Nanobots are more like bacteria because they are self sufficient whereas viruses need a host to duplicate the virus RNA for them.


_________________
I've left WP indefinitely.


Humanaut
Veteran
Veteran

User avatar

Joined: 17 Jul 2014
Age: 53
Gender: Male
Posts: 4,390
Location: Norway

25 Jul 2014, 8:02 am

If they are anything like bacteria, well, I found this: http://www.npr.org/blogs/health/2013/11 ... ever-stops



TallyMan
Veteran
Veteran

User avatar

Joined: 30 Mar 2008
Gender: Male
Posts: 40,061

25 Jul 2014, 8:08 am

Humanaut wrote:
If they are anything like bacteria, well, I found this: http://www.npr.org/blogs/health/2013/11 ... ever-stops


Interesting article, thanks for posting the link.


_________________
I've left WP indefinitely.


ripped
Veteran
Veteran

User avatar

Joined: 5 Jan 2013
Gender: Male
Posts: 651

26 Jul 2014, 12:07 am

Self-replication needs DNA.

So the self replicator facility needs an automated form of genome construction, preferably atom by atom from the ground up.
But even that's not self-replication.
I think a workable self-replicating nanobot theory may still be some way off. :study:



Here
Deinonychus
Deinonychus

User avatar

Joined: 17 Jun 2012
Age: 61
Gender: Male
Posts: 379
Location: California

04 Aug 2014, 11:51 am

RELATED: LINK Why a Deep-Learning Genius Left Google & Joined Chinese Tech. Shop Baidu (interview) [Baidu is China's search-engine; that is becoming increasingly known outside of China]

(LINK) http://pagesay.com/why-a-deep-learning- ... interview/



Here
Deinonychus
Deinonychus

User avatar

Joined: 17 Jun 2012
Age: 61
Gender: Male
Posts: 379
Location: California

28 Oct 2014, 6:41 pm

Applied 'Artificial Intelligence' technologies have progressed lately. Here are a few examples:

Website development: http://www.webwire.com/ViewPressRel.asp ... FAkRldhCuJ

Search-Engine Optimization (SEO): http://www.matrixsearch.com/

Automated 'help-lines'/ troubleshooting (WHOA!): http://www.gizmag.com/amelia-artificial ... nce/34024/

The prospects of advanced Artificial Intelligence (AI) are viewed with caution: http://www.huffingtonpost.com/2014/10/2 ... 53804.html



QuantumChemist
Veteran
Veteran

User avatar

Joined: 18 Oct 2014
Gender: Male
Posts: 1,910
Location: Midwest

02 Nov 2014, 1:08 pm

I am limited on what I can discuss on this topic (for research reasons), but I can give you something to ponder on. This molecule was designed back in 2005 at Rice University to determine how buckminsterfullerene move over flat metallic surfaces:

http://www.edinformatics.com/nanotechnology/nanocar.htm

If you go to the bottom of the page, there is a link to the original research paper abstract on the molecule. Basically, it moves across a gold surface upon thermal excitation (ie. heating). The funny thing is that it actually does have potential applications, just not current commercial ones. You have to be able to think outside of the box to see some of them. (For example, it is a good starting point for building nanobot frameworks.) There are nanotrucks and nanorotors from the same research paper....



QuantumChemist
Veteran
Veteran

User avatar

Joined: 18 Oct 2014
Gender: Male
Posts: 1,910
Location: Midwest

02 Nov 2014, 1:09 pm

I am limited on what I can discuss on this topic (for research reasons), but I can give you something to ponder on. This molecule was designed back in 2005 at Rice University to determine how buckminsterfullerene move over flat metallic surfaces:

http://www.edinformatics.com/nanotechnology/nanocar.htm

If you go to the bottom of the page, there is a link to the original research paper abstract on the molecule. Basically, it moves across a gold surface upon thermal excitation (ie. heating). The funny thing is that it actually does have potential applications, just not current commercial ones. You have to be able to think outside of the box to see some of them. (For example, it is a good starting point for building nanobot frameworks.) There are nanotrucks and nanorotors from the same research paper....