NextGen

[Rickety]

Try this for size...

http://www.dailymail.co.uk/sciencetech/article-1231158/Military-style-drones-set-patrol-coastline-spot-drug-smugglers-illegal-immigrants.html

There is stuff on youtube also.

Peace

Jim

Wow! I enjoyed reading that one! Very interesting. Do you think they'll start using them for other purposes? I think it would be a good idea as long as they kept the programing very simple and controlled by humans. But it does have the potential to wreak havoc if put into the wrong hands, for example: the enemies of the UK could get their hands on one and spy on them and they wouldn't know it because it would look just like the UK HERTI's. Hmm..... Something to think about huh? That was a very interesting article! Thank you for sharing that!

[Cadetak]

If super computers had access to networks and they were threatened I think they could become sentient in the sense that they would want to continue to exist. The servant becomes the master perhaps. Where is the harm in retaining control, particularly with weapon systems? If civilians are killed who is responsible, the engineer, the manufacturer? There needs to be accountability and armed drones don't allow for this.

Computers do have access to networks and have bin vulnerable to attack since well...the dawn of the internet. No sentience.

A computer does not know it exists unless it is sentient, so it has to have sentience before it can fear its death.

Armed drones are controlled remotely by human operators. We've have actually bin utilizing robotics and spy drones in the military for awhile.

[SlimJim22]

Computers do have access to networks and have bin vulnerable to attack since well...the dawn of the internet. No sentience.

A computer does not know it exists unless it is sentient, so it has to have sentience before it can fear its death.

Armed drones are controlled remotely by human operators. We've have actually bin utilizing robotics and spy drones in the military for awhile.

Isn't sentience just a complex strcuture for assessing probabilities? If a computer program could choose a course of action depending on the likelihood of success is this not a basic form of sentience. Using complex algorhytm or something couldn't a program begin to acquire info and potentially evolve beyond its initial programming?

[TheSearcher]

Isn't sentience just a complex strcuture for assessing probabilities? If a computer program could choose a course of action depending on the likelihood of success is this not a basic form of sentience. Using complex algorhytm or something couldn't a program begin to acquire info and potentially evolve beyond its initial programming?

An algorythm, as complex as you want, can still only work within the parameters it has been given or possibilites that have been programmed into it. For it to "evolve", like you say, beyond it's original programming is not possible as such.

Besides, the programm would need to have root access to rewrite itself, ever tried to get the root password from a sysadmin? The odds of getting a properly working OS from Microsoft are better. (I know, nerd joke alarm!!)

Like most coders will tell you, any software is only as good as it's worst coder.

Also I'm a bit uncomfortable with the word sentient, I would rather say self-aware, that is only my opinion though.

[SlimJim22]

An algorythm, as complex as you want, can still only work within the parameters it has been given or possibilites that have been programmed into it. For it to "evolve", like you say, beyond it's original programming is not possible as such.

Besides, the programm would need to have root access to rewrite itself, ever tried to get the root password from a sysadmin? The odds of getting a properly working OS from Microsoft are better. (I know, nerd joke alarm!!)

Like most coders will tell you, any software is only as good as it's worst coder.

Also I'm a bit uncomfortable with the word sentient, I would rather say self-aware, that is only my opinion though.

No you're right, self aware is a better phrase. I am not a computer programmer so am unqualified to give an accurate argument but there no doubt massive advances ahead in terms of AI development. The program becomes the decision maker but there would need an unlikely event for a program to become self aware and feel threatened. However, the internet is an infinte play area for programs to dissappear off grid in the future so we might be well advised to proceed with caution. I am still worried by the thought of drones being armed, as I have said their is a problem with accountability. If they have a controller that is fine but there are still chances of misidentification. Do the benefits outweigh the costs enough to take such a radical step. You'll tell me it's not radical but that is the idea behind skynet I guess. The next step... in our growth and security or our loss of control and possibly even destruction.

[TheSearcher]

No you're right, self aware is a better phrase. I am not a computer programmer so am unqualified to give an accurate argument but there no doubt massive advances ahead in terms of AI development. The program becomes the decision maker but there would need an unlikely event for a program to become self aware and feel threatened. However, the internet is an infinte play area for programs to dissappear off grid in the future so we might be well advised to proceed with caution. I am still worried by the thought of drones being armed, as I have said their is a problem with accountability. If they have a controller that is fine but there are still chances of misidentification. Do the benefits outweigh the costs enough to take such a radical step. You'll tell me it's not radical but that is the idea behind skynet I guess. The next step... in our growth and security or our loss of control and possibly even destruction.

Any software is in itself (on a very basic level) already a decision maker, as it has it's set of parameters, in the limits of which it can function and as such take "decisions".

If you want a full, self-aware AI, then you would have to go further than that. It would have to be able to learn and make decisions outside of it's established parameters (call it thinking outside of the box, lol, quite literally), and to a degree rewrite it's own parameters, like we humans do; rewriting it's own base code. That might seem quite straight forward, but that would be something quite complicated, as far as programming is concerned.

Certain programs can mimick decision making processes, like humans do it, but in terms of real self-awarness, we still have a long way to go. Again, this is my opinion only.

[SlimJim22]

Any software is in itself (on a very basic level) already a decision maker, as it has it's set of parameters, in the limits of which it can function and as such take "decisions".

If you want a full, self-aware AI, then you would have to go further than that. It would have to be able to learn and make decisions outside of it's established parameters (call it thinking outside of the box, lol, quite literally), and to a degree rewrite it's own parameters, like we humans do; rewriting it's own base code. That might seem quite straight forward, but that would be something quite complicated, as far as programming is concerned.

Certain programs can mimick decision making processes, like humans do it, but in terms of real self-awarness, we still have a long way to go. Again, this is my opinion only.

So, what you''re saying is we shouldn't teach programs to re-program themselves because that is when they 'could' become self aware and start making decisions for themselves? What do you think of the link Cameron has with these things? Terminator, Dark Angel (Wow!), and Avatar of course (Montauk chair/couch) Did the egyptians have a story about three couches or something. I will see if Kmt knows.

[Rickety]

Okay first of all, let's take a look at the acronym AI, what does it stand for? Artificial Intelligence, which means it has intelligence to a certain degree. Intelligence is what? The ability to think or the ability to reason and/or problem solve. So really what it should be is AS Artificial Sentience. Now I'm not saying that all AI is sentient, I'm saying that the word "intelligence" is almost directly tied TO sentience. Now it was "TheSearcher" I believe that said a computer program can not rewrite itself, that is wrong, if you have ever messed with BASIC coding with a program like BlitzBASIC or DarkBASIC, then you should know well about one of the tutorial programs that it gives you about saving information. Now, true enough it still goes on the programming that it's made of, but this program actually WRITES another script and saves it elsewhere on the computer. If that program has the kind of power to WRITE another file then what makes you think it can't rewrite itself? That program just created another file. Same basic principle with a virus, it creates and destroys files and with certain viruses, it has to rewrite itself based on your directories, for example: if you have a virus that forces a format of the common, or C:\ drive then it will do that right? Well what if somebody renamed their C:\ drive to say J:\ or X:\ well then it would have to rewrite itself to format the other name of the file system. So a program CAN rewrite itself, they've been doing it for years.

Quick question... Have you ever had a Trojan Horse Virus on your computer? Trickily little devils aren't they? You ran a virus scan I presume. Well what did that virus scan do? It went out and looked for any viruses on your computer. It conducted a search and picked out the files that were infected based on problem solving skills and recognizing anomalies, just like a doctor does with a diagnoses. So in conclusion, both virus scanning software and viruses are prime examples of a program that can snap into sentience at any given time. Now why haven't they? Basically, because they don't see us as a threat, a virus may but the scanning software doesn't because without us it wouldn't work.

And plus, what would a virus do with an ordinary computer? It has no weapons and Trojan Horse Viruses are easily spotted, even by the home computer so any weapons systems are gonna find easily.

[digitalartist]

Quick question... Have you ever had a Trojan Horse Virus on your computer? Trickily little devils aren't they? You ran a virus scan I presume. Well what did that virus scan do? It went out and looked for any viruses on your computer. It conducted a search and picked out the files that were infected based on problem solving skills and recognizing anomalies, just like a doctor does with a diagnoses. So in conclusion, both virus scanning software and viruses are prime examples of a program that can snap into sentience at any given time. Now why haven't they? Basically, because they don't see us as a threat, a virus may but the scanning software doesn't because without us it wouldn't work.

And plus, what would a virus do with an ordinary computer? It has no weapons and Trojan Horse Viruses are easily spotted, even by the home computer so any weapons systems are gonna find easily.

Virus scans look for what they call signatures that tell them that a virus is present and what the virus is. Humans take infected files then insert them into computers with no ties to any other systems. Then once the fresh files have been infected they dissect them, comparing them to what the files were and viola they have the virus code and from there create the signatures (code to look for) that make antivirus programs able to find the viruses. These new signatures are uploaded during an update. No sorry but no sentience there.

[Rickety]

Virus scans look for what they call signatures that tell them that a virus is present and what the virus is. Humans take infected files then insert them into computers with no ties to any other systems. Then once the fresh files have been infected they dissect them, comparing them to what the files were and viola they have the virus code and from there create the signatures (code to look for) that make antivirus programs able to find the viruses. These new signatures are uploaded during an update. No sorry but no sentience there.

Actually the virus scan software looks for things that are abnormal, it does use thinking skills to pick them apart. And yes viruses do adapt to different OS and different programs, I got a virus on Ubuntu once, Ubuntu doesn't typically get viruses and it certainly doesn't get Trojan Horses, yet when I scanned it was a Trojan Horse, so viruses are adaptable.

Plus I didn't say they WERE sentient, I said they had the capability.

Edited January 25, 2010 by AngelsShadow

[TheSearcher]

Okay first of all, let's take a look at the acronym AI, what does it stand for? Artificial Intelligence, which means it has intelligence to a certain degree. Intelligence is what? The ability to think or the ability to reason and/or problem solve. So really what it should be is AS Artificial Sentience. Now I'm not saying that all AI is sentient, I'm saying that the word "intelligence" is almost directly tied TO sentience. Now it was "TheSearcher" I believe that said a computer program can not rewrite itself, that is wrong, if you have ever messed with BASIC coding with a program like BlitzBASIC or DarkBASIC, then you should know well about one of the tutorial programs that it gives you about saving information. Now, true enough it still goes on the programming that it's made of, but this program actually WRITES another script and saves it elsewhere on the computer. If that program has the kind of power to WRITE another file then what makes you think it can't rewrite itself? That program just created another file. Same basic principle with a virus, it creates and destroys files and with certain viruses, it has to rewrite itself based on your directories, for example: if you have a virus that forces a format of the common, or C:\ drive then it will do that right? Well what if somebody renamed their C:\ drive to say J:\ or X:\ well then it would have to rewrite itself to format the other name of the file system. So a program CAN rewrite itself, they've been doing it for years.

Quick question... Have you ever had a Trojan Horse Virus on your computer? Trickily little devils aren't they? You ran a virus scan I presume. Well what did that virus scan do? It went out and looked for any viruses on your computer. It conducted a search and picked out the files that were infected based on problem solving skills and recognizing anomalies, just like a doctor does with a diagnoses. So in conclusion, both virus scanning software and viruses are prime examples of a program that can snap into sentience at any given time. Now why haven't they? Basically, because they don't see us as a threat, a virus may but the scanning software doesn't because without us it wouldn't work.

And plus, what would a virus do with an ordinary computer? It has no weapons and Trojan Horse Viruses are easily spotted, even by the home computer so any weapons systems are gonna find easily.

Think about the context I made that statement in. It can only do so in the parameters it has been given to begin with. Besides, it does NOT rewrite it's base code.

[TheResearcher]

Any software is in itself (on a very basic level) already a decision maker, as it has it's set of parameters, in the limits of which it can function and as such take "decisions".

If you want a full, self-aware AI, then you would have to go further than that. It would have to be able to learn and make decisions outside of it's established parameters (call it thinking outside of the box, lol, quite literally), and to a degree rewrite it's own parameters, like we humans do; rewriting it's own base code. That might seem quite straight forward, but that would be something quite complicated, as far as programming is concerned.

Certain programs can mimick decision making processes, like humans do it, but in terms of real self-awarness, we still have a long way to go. Again, this is my opinion only.

Not necessarily. As I mentioned in my previous posts on this topic, to some extent software can act outside the boundaries of it's programming. For example, ASIMO can be taught new movements, and how to recognise new objects. The movements are not programmed into the machine; the software is given the ability to make movements, and then it "decides" how to perform them. Of course though, this is far from close to human AI, but it does illustrate that a program does not have to be self-modifying to work outside it's usual parameters.

Okay first of all, let's take a look at the acronym AI, what does it stand for? Artificial Intelligence, which means it has intelligence to a certain degree. Intelligence is what? The ability to think or the ability to reason and/or problem solve. So really what it should be is AS Artificial Sentience. Now I'm not saying that all AI is sentient, I'm saying that the word "intelligence" is almost directly tied TO sentience. Now it was "TheSearcher" I believe that said a computer program can not rewrite itself, that is wrong, if you have ever messed with BASIC coding with a program like BlitzBASIC or DarkBASIC, then you should know well about one of the tutorial programs that it gives you about saving information. Now, true enough it still goes on the programming that it's made of, but this program actually WRITES another script and saves it elsewhere on the computer. If that program has the kind of power to WRITE another file then what makes you think it can't rewrite itself? That program just created another file. Same basic principle with a virus, it creates and destroys files and with certain viruses, it has to rewrite itself based on your directories, for example: if you have a virus that forces a format of the common, or C:\ drive then it will do that right? Well what if somebody renamed their C:\ drive to say J:\ or X:\ well then it would have to rewrite itself to format the other name of the file system. So a program CAN rewrite itself, they've been doing it for years.

It need not be so complex as you suggest. The BASIC language is pretty abstract from memory, however modifying the code of a program as it executes is a fairly simple chore. I haven't had much experience with BASIC, but it is probably easier in C++. All that has to be done is for you to open the process, access the next instruction to be executed, and overwrite the data. It can be done very easily - just as easily to another process as the current one executing (cheats such as "Aimbots" regularly do this to access the graphics routines of the game). You needn't mess around with the whole save a new file thing, but you were on the right lines.

Indeed there are already programs which use self-modifying code. For example, Just-In-Time compilers and polymorphic worms (to a lesser extent).

Quick question... Have you ever had a Trojan Horse Virus on your computer? Trickily little devils aren't they? You ran a virus scan I presume. Well what did that virus scan do? It went out and looked for any viruses on your computer. It conducted a search and picked out the files that were infected based on problem solving skills and recognizing anomalies, just like a doctor does with a diagnoses. So in conclusion, both virus scanning software and viruses are prime examples of a program that can snap into sentience at any given time. Now why haven't they? Basically, because they don't see us as a threat, a virus may but the scanning software doesn't because without us it wouldn't work.

And plus, what would a virus do with an ordinary computer? It has no weapons and Trojan Horse Viruses are easily spotted, even by the home computer so any weapons systems are gonna find easily.

I'm pretty sure a program not intended to develop self-awareness would not be able to do so on it's own. The reason trojans do not develop self-awareness is not down to the idea that they do not feel threatened (which is a presupposition in itself that the system could 'think' before it is self-aware), but instead because they do not have the capability. Trojans are fairly simple programs in essence; routines to hide detection, to open up a port to allow communication with the hacker, and perhaps other secondary tasks to keylog or what ever. These types of program are down to the intelligence of the creator rather than itself. Perhaps someday if hackers start incorporating AI methods into their programs, it might be possible, but it's highly doubtful.

P.S. Technically a virus is different from a trojan, but meh, technicalities.

[digitalartist]

Actually the virus scan software looks for things that are abnormal, it does use thinking skills to pick them apart. And yes viruses do adapt to different OS and different programs, I got a virus on Ubuntu once, Ubuntu doesn't typically get viruses and it certainly doesn't get Trojan Horses, yet when I scanned it was a Trojan Horse, so viruses are adaptable.

Plus I didn't say they WERE sentient, I said they had the capability.

You used the term "Snapping into sentience" indicates that at some point they become sentient which is more than just having the capability.

[Rickety]

You used the term "Snapping into sentience" indicates that at some point they become sentient which is more than just having the capability.

Yes, but again I said that they have the capability, I didn't say that they WERE so this is irrelevant. Plus somebody already started complaining about that so you're a little late.

[Stellar]

Im not even going to touch on Viruses and Anti Viruses being "capable" of sentience, because thats just so far gone I'm still having trouble believing someone proposed that, but I will touch on this:

Think about the context I made that statement in. It can only do so in the parameters it has been given to begin with. Besides, it does NOT rewrite it's base code

It can though. If you were to give me the right time, resources and motivation, I could write for you a program that would rewrite its own source code. It wouldnt even be that difficult. All that is required is for the code to be available to it. It could then look into the code, modify it to whatever end it wants, compile it, and replace the old executable with the newly modified one. In essence, it would have just rewritten its own source code.

[TheSearcher]

Im not even going to touch on Viruses and Anti Viruses being "capable" of sentience, because thats just so far gone I'm still having trouble believing someone proposed that, but I will touch on this:

It can though. If you were to give me the right time, resources and motivation, I could write for you a program that would rewrite its own source code. It wouldnt even be that difficult. All that is required is for the code to be available to it. It could then look into the code, modify it to whatever end it wants, compile it, and replace the old executable with the newly modified one. In essence, it would have just rewritten its own source code.

OK point taken, you're right.

[Rickety]

Im not even going to touch on Viruses and Anti Viruses being "capable" of sentience, because thats just so far gone I'm still having trouble believing someone proposed that, but I will touch on this:

It can though. If you were to give me the right time, resources and motivation, I could write for you a program that would rewrite its own source code. It wouldnt even be that difficult. All that is required is for the code to be available to it. It could then look into the code, modify it to whatever end it wants, compile it, and replace the old executable with the newly modified one. In essence, it would have just rewritten its own source code.

See the only reason I started talking about viruses is because they sometimes have to change their code to fit the name of the common drive on the computer so therefore are adaptive to computers giving them a small level of intelligence. They also infect and spread like a normal biological virus, so they have to have the capability to find different files without knowing their names and infect them. How does it infect those files without knowing their name? (I mean that quite literally lol I don't know a lot about viruses, I only ever knew how to make a very basic echo virus and that was it lol)

[WoIverine]

Future aviation will probably be something matrix based I would think. Sit down in a chair, plug in, link up to your plane, control it virtually which in essence controls a drone in the real world. That seems a logical next step due to the rapid advancements in neuroscience we've been seeing recently.

Edited January 28, 2010 by SpiderCyde

[Darkcore]

"NextGen is an aviation system being developed by the FAA. NextGen will take all aviation controls to a satellite based system rather than a ground based. It will allow airplanes to fly closer together "safely" and plot more direct courses from point A to point B. The satellite system of NextGen is being developed with the FAA the Military and NASA. Will the Military and Government want to use this new system for a weapons and defense system? The chances of that are extremely high."

Future aviation will probably be something matrix based I would think. Sit down in a chair, plug in, link up to your plane, control it virtually which in essence controls a drone in the real world. That seems a logical next step due to the rapid advancements in neuroscience we've been seeing recently.

I agree that this would be the next step. Seeing how supersonic fighter plane can manoeuvre, they are limited by the pilot that can only take so much G's before they pass out. Put the pilot "in link" with the plane trough the "NextGen" satellite system and they can make the fighter plane do trick's pilots can only dream of right now.

So I'm not getting into the computer going self-aware thing, but YES, they will want to use it as a weapon / defense system!