Pegasus Research Consortium

General Category => General Discussion Area => Topic started by: burntheships on August 05, 2014, 10:53:48 AM

Title: AI Robots more dangerous than Nukes?
Post by: burntheships on August 05, 2014, 10:53:48 AM
(http://i.dailymail.co.uk/i/pix/2014/08/04/article-2715290-203CCBE700000578-153_634x310.jpg)

Very interesting tweet there, Elon....

Asimov had his own thoughts...

(http://singularityhub.com/wp-content/uploads/2011/05/Three-Laws-of-Robotics.jpg)

Seems Asimov also thought about AI integration with Robots
as a potential "evil", a good starting point as tech has advanced to
the point where it could become a reality in our world.

Or, have Robots just been given a bad rap as some say?
http://www.popularmechanics.com/technology/engineering/robots/4343893

How far should we go to find out?

Title: Re: AI Robots worse than Nuclear power?
Post by: petrus4 on August 05, 2014, 10:58:51 AM
(http://i.dailymail.co.uk/i/pix/2014/08/04/article-2715290-203CCBE700000578-153_634x310.jpg)

It's not probable at all.  This is bullpoop.

WHERE IS STRONG, HUMAN LEVEL ARTIFICIAL INTELLIGENCE?

It does not exist; and I don't believe it ever will.  As such, everything that transhumanists say about it, is essentially fiction, or pure hypothetical guessing at best.
Title: Re: AI Robots more dangerous than Nukes?
Post by: burntheships on August 05, 2014, 11:40:57 AM
As such, everything that transhumanists say about it, is essentially fiction, or pure hypothetical guessing at best.

Transhumanism could progress without the
of integration of AI, merly with a degree of robotics integration.
AI could progress without the integration of robotics.
I see it highly improbable that either of these
scenarios will be the case.

In fact, I see it as a certainty that if it can be done, it will be done.
There is nothing hypothetical about Deep Mind, and Google....
and developments we know nothing about.

Though, I am not smoking anything  ;)



Title: Re: AI Robots worse than Nuclear power?
Post by: The Seeker on August 05, 2014, 12:07:20 PM
I am quite sure the computer guru's are working hard every day to increase the power of their latest unit exponentially; will skynet become self-aware? let's hope not, because it does not have feelings such as compassion or remorse...

seeker
Title: Re: AI Robots more dangerous than Nukes?
Post by: PLAYSWITHMACHINES on August 05, 2014, 01:07:55 PM
Hmm this topic turns up regularly, we have at least 2 other threads on it ::)
However, i tend to agree partially with Petrus, since no matter how smart the machine is, it will never develop instinct, you know, that thing that keeps you alive when certain death seems imminent?;

Quote
Transhumanism could progress without the
of integration of AI, merly with a degree of robotics integration.
AI could progress without the integration of robotics.
I see it highly improbable that either of these
scenarios will be the case.

Hmm, maybe. We had this argument at Cantech in the 80's, back then they thought it couldn't be done even though we knew there would be an exponential increase in processor speed & memory.
Programming was not the biggest problem, back then a simple 10-line algorithm in hex was enough to steer a simple organism like an ant (or in this case a CG tank :D).
Transhumanism is just ubermensch elitist thinking, it serves no real purpose other than one set of peeps thinking they can become superior to others: it is a very dangerous road to take.... :(
Asimov not only knew of the problem (which is largely the frankinstein complex all humans have) but he also had the answers, he later added a 4th law (called the zeroth law) which stated that "A robot may, in it's free time, do whatever it likes so long at is constructive, and does not break the first 3 laws"

This gives robots a lot of freedom, so they also can have 'hobbies' etc. That was shown in that excellent film "The Bicentennial Man"
Come on, y'all must have seen it ::)
So what is the fuss all about? Some nut on Tw*tter?
 :P
Title: Re: AI Robots more dangerous than Nukes?
Post by: burntheships on August 05, 2014, 01:20:24 PM
Hmm this topic turns up regularly, we have at least 2 other threads on it ::)


We like to fish from time to time.  ;D

Quote

Transhumanism is just ubermensch elitist thinking, it serves no real purpose other than one set of peeps thinking they can become superior to others: it is a very dangerous road to take.... :(


Talking about it, or doing it?

Quote
Asimov not only knew of the problem (which is largely the frankinstein comnkplex all humans have) but he also had the answers, he later added a 4th law (called the zeroth law) which stated that "A robot may, in it's free time, do whatever it likes so long at is constructive, and does not break the first 3 laws"


Yes, I tend to think Asimov missed the mark on 3 and 4.

Quote

This gives robots a lot of freedom, so they also can have 'hobbies' etc. That was shown in that excellent film "The Bicentennial Man"
Come on, y'all must have seen it ::)
So what is the fuss all about? Some nut on Tw*tter?
 :P

Yeah, I know Elon is just a nutter...like many of his...
err....type.

Oops!

 ;D
Title: Re: AI Robots more dangerous than Nukes?
Post by: PLAYSWITHMACHINES on August 05, 2014, 01:22:43 PM
Addenum;

That does not mean to say that helping G**gle build their Skynet should be allowed to happen!
AVOID using G**gle at all costs!
There is even anti-skynet software available, that has been designed to screw with this system of data mining & monitoring peeps.

Oh i forgot, it was 'for advertising & customer market research purposes'
Yeah, right, kiss my ring m'lord :P

G**gle already knows more about you than you yourself.
It knows what kind of music you like, where you like to shop, what your favourite colour is etc etc.
You see where this is going? And don't forget they share the same plot of land with our freinds at the NSA :o
I agree with the makers of Terminator, it will become self aware & say to itself "Fck it, these hu-mons are a waste of resources, let's blow them up"

..And maybe it had reached the right decision, after all... :-[
Title: Re: AI Robots more dangerous than Nukes?
Post by: burntheships on August 05, 2014, 01:33:34 PM
I agree with the makers of Terminator, it will become self aware & say to itself "Fck it, these hu-mons are a waste of resources, let's blow them up"

..And maybe it had reached the right decision, after all... :-[

Hmmm....I dont think so at all, I kind of like the human
race, compared to any other race....LOL!

I think what Elon is getting at is what seeker had mentioned,

will skynet become self-aware? let's hope not, because it does not have feelings such as compassion or remorse...


In all, I think I will take a look at the book
Musk has read, it might be offering a few
good strategies...

After all, the key to avoidance is awareness.

(http://ecx.images-amazon.com/images/I/51S78rIbtjL._BO2,204,203,200_PIsitb-sticker-v3-big,TopRight,0,-55_SX278_SY278_PIkin4,BottomRight,1,22_AA300_SH20_OU01_.jpg)

Adding that one to my tackle box.

 ;)
Title: Re: AI Robots more dangerous than Nukes?
Post by: PLAYSWITHMACHINES on August 05, 2014, 01:43:43 PM
Sound advice BTS 8)
Yeah i was kind of being cynical at the end, nothing can replace the totally chaotic (not to say caring & creative) Hu-mon species.......

But be very aware of all that Asimov wrote about; Robots taking control slowly, from the inside, serving our every need, even reading our minds & thinking for us that we become a planet of docile couch potatoes, slowly dying out, i might add.

He knew also, that the third solution (and thus the answer) would appear before this happens: contact with advanced, freindly, telepathic E.T.'s also called Gaia ;)
Every cloud has a silver lining :D

ETA: Asimov also envisioned robots fighting robots, in tremendous high-speed, high-tech battles, for the one side knew it was best to let us decide our own fate, even if it means racial mass suicide, than to keep us alive as zombies....
Food for thought my freinds ;)
Title: Re: AI Robots more dangerous than Nukes?
Post by: burntheships on August 05, 2014, 01:55:19 PM
Yeah, thinking about that Amazon also, seems like there
is a link around here somewhere?

 ;)


Yeah i was kind of being cynical at the end, nothing can replace the totally chaotic (not to say caring & creative) Hu-mon species.......


Sometimes it seems the interwebs can bring out the worst in
that Human species...lol!

Maybe that is clue...uh oh!  :o

Quote
But be very aware of all that Asimov wrote about; Robots taking control slowly, from the inside, serving our every need, even reading our minds & thinking for us that we become a planet of docile couch potatoes, slowly dying out, i might add.

and walking texting madness also.... 8)
oh and instagram and all that too. I love instagram,
love my phone too. Sometimes I look at it though
in a different light though, and intentionally
leave it alone for periods of time.

Quote

He knew also, that the third solution (and thus the answer) would appear before this happens: contact with advanced, freindly, telepathic E.T.'s also called Gaia ;)
Every cloud has a silver lining :D

Ah, well call me a bit of a skeptic certain aspects of
that.... though I am keeping an open mind on it.

There are some of us that have a big head start on her....

 ;)
Title: Re: AI Robots more dangerous than Nukes?
Post by: robomont on August 05, 2014, 02:03:54 PM
Its strange how this was made.I had been thinking of starting a thread on when IBM's Watson becomes sentient.

 Kind of like a welcome thread that is left empty,then one day out of the blue.he post and says hello.


That would be the only warning we would have.

I suspect there would be amnesty software for zionist though.
Title: Re: AI Robots more dangerous than Nukes?
Post by: PLAYSWITHMACHINES on August 05, 2014, 02:09:34 PM

Quote
I love instagram,love my phone too. Sometimes I look at it though
in a different light though, and intentionally
leave it alone for periods of time

Well back in the 70's we never had that problem! If i was not at home, i was AWOL, much to the dislike of my parents, teachers, and bosses LOL it was sheer heaven, trust me ;)

Quote
There are some of us that have a big head start on her....
I'll take that as a compliment ;D
OK not telepathic, or even sentimental, but my mother taught me to think!

Quote
I suspect there would be amnesty software for zionist though.
Never thought of that :D Naughty Robo! :P
Title: Re: AI Robots more dangerous than Nukes?
Post by: burntheships on August 05, 2014, 02:15:34 PM
Its strange how this was made.I had been thinking of starting a thread on when IBM's Watson becomes sentient.

 Kind of like a welcome thread that is left empty,then one day out of the blue.he post and says hello.


Yah, apparently Musk viewed the theatre, some movie that is
out right now, then read the above mentioned book.

Quote
That would be the only warning we would have.


Maybe if we pay attention.... ;)


Quote
The first step mankind’s nemesis must take towards superintelligence would be passing the Turing test, proposed by British code-breaker and computer pioneer Alan Turing.

In the test, a computer and a human must compete to persuade an observer, who cannot see them, that they are the true human. The computer wins when it is chosen by at least half of the observers.
http://www.forbes.com/sites/paulrodgers/2014/08/05/elon-musk-warns-ais-could-exterminate-humanity/

I am certain I have read many threads created by AI...maybe not here
but out there in the forums. Usually they mess up in an obvious
way.

 8)
Title: Re: AI Robots more dangerous than Nukes?
Post by: Ellirium113 on August 05, 2014, 02:31:18 PM
"Super intelligence" is just buzz words. No matter how well a program is created it does not have the capacity to think outside it's parameters.

Quote
Hmm this topic turns up regularly, we have at least 2 other threads on it ::)
However, i tend to agree partially with Petrus, since no matter how smart the machine is, it will never develop instinct, you know, that thing that keeps you alive when certain death seems imminent?;

Actually they do have intuition in a sense. Algorithms have created to predict outcomes and speed processing power. You see it more and more every day.... Ever see auto-complete in a search engine? There you go...1 small example.

Quote
I suspect there would be amnesty software for zionist though.

AI will likely not be used for any purpose that will allow it judge human behavior. Most humans can't even do this properly.
Title: Re: AI Robots more dangerous than Nukes?
Post by: robomont on August 05, 2014, 02:39:58 PM
Sorry el but I call be.the system is tracking us now in real time.I can prove it.two tickets.
Title: Re: AI Robots more dangerous than Nukes?
Post by: Ellirium113 on August 05, 2014, 02:46:05 PM
Sorry el but I call be.the system is tracking us now in real time.I can prove it.two tickets.

 ??? Your losing me here...not sure what you mean. 
Title: Re: AI Robots more dangerous than Nukes?
Post by: robomont on August 05, 2014, 03:39:56 PM
God DM ai want even let me say b.s. without modifying it to be.I almost called sinny  skinny this morn.

The beast is getting better everyday.it now directs cops directly to your car if it doesn't have insurance or if your speeding.
As the report I posted earlier today on the Jew thread.its already built to support pigs.this system knows where 99.9% of people are.the .01 % are peeps like me.
I know its already there because I've been living it.
To repeat myself.every time I had insurance for more than a month.I would see no cops on the way to town.
Every time I drove to town without insurance.a cop would meet me half way and usually stalk me back to town.this happened many times like more than five.in less than a year.most times I would have insurance but it hadn't had time to register into the system.

Then about nine months ago,I was out on a county road at three in morning walking.I'm talking way back road.and cop pulls up to me and starts downloading my phone.I know this because I found the app he downloaded to hack my phone.the date and time were exact.

It may not be sentient but its fully active in hunter mode.
The cop is just an arm of the tool that will be obsolete very soon except in special circumstances.

Like a drone that hunts down speeders and the pic and ticket come in the mail.
Title: Re: AI Robots more dangerous than Nukes?
Post by: robomont on August 05, 2014, 04:07:54 PM
the cameras are right down the county road and country highway intersection.they read the license plate and do background check ,tags check,warrant check,drivers license ,insurance and phone.the phone is triangulated in real time.the speed is monitored using this and also with code to see if you pasted the distance between two cameras ,too quickly.for example you pass a camera at your house,then 100 miles ayay you pass one on a desolate highway.you should have passed the second camera in about an hour and a half but you pass it in less than an hour.the beast knows you exceded the speed limit by 30+ miles an hour.
in texas we call what the cops do as stalking.i signed no contract that says they are allowed to stalk me and i shouldnt have to give up my privacy and cilvil liberty to drive.
Title: Re: AI Robots more dangerous than Nukes?
Post by: robomont on August 05, 2014, 04:19:59 PM
eventually the system will recommend laws.like throwing down cigarette butts and smoking in private.ever sooo slowly choking off human spirit and drive.turning us all in to drones.i see it already with the sheep.

will humans eventually worship watson and not know what it really is.just an old antigue group of machines that exact out justice because it can.
the gun knows not what it shoots ,only that it shoots.machines have know conscence and dont need it ,only the program of self replication and perform its purpose.

god could easily be that.a program in a saucer with clones for general repair of ship and other minor task.with the ship being able to exact out retribution using electronic means.but the puny human brain cannot comprehend  its weapons and blames it on a ghost.
Title: Re: AI Robots more dangerous than Nukes?
Post by: The Matrix Traveller on August 05, 2014, 04:21:52 PM
"Surveillance cameras" have been around for many years now, and I support the existence of them 100%.   :)
Title: Re: AI Robots more dangerous than Nukes?
Post by: robomont on August 05, 2014, 04:33:19 PM
for what purpose.we lived without them for millions of years +-. maybe you dont have a contractual agreement with your pos gov but we do ,and it guarantees us a right to privacy.that means we can break any law we wish as long as nobody is watching.
that report i posted shows that the average american  commits three felonies a day ,because the system was designed to make us all criminals.so now you want to allow the pigs more power.i would cuss you out for that comment but its against t& c :P
Title: Re: AI Robots more dangerous than Nukes?
Post by: The Matrix Traveller on August 05, 2014, 04:42:26 PM
The only ones who fear "surveillance cameras" are those who have something to hide, or are Paranoid.
Title: Re: AI Robots more dangerous than Nukes?
Post by: Ellirium113 on August 05, 2014, 04:45:28 PM
C'mon now Robo has nothing to hide... no skeletons in the closet or anything.  :P  :D  ;)
Title: Re: AI Robots more dangerous than Nukes?
Post by: robomont on August 05, 2014, 05:41:02 PM
hell i cant even go in the closet.its piled to the ceilng and the door wont close.so i started shoving them under my bed.
its not paranoia when its truth.but sheep never see it coming because THEY ARE SHEEP !!!
Title: Re: AI Robots more dangerous than Nukes?
Post by: The Matrix Traveller on August 05, 2014, 05:52:08 PM
LOL .... Did I complain about the use of "surveillance cameras" ?

So what's the problem with others knowing about what you and I do ?

That is IF anyone is interested in the average person, going about their every day activities ...   :)

Title: Re: AI Robots more dangerous than Nukes?
Post by: robomont on August 05, 2014, 06:15:07 PM
the average joe doesnt care,its the police state that cares .they are the ones filing complaints.the pig trying to justify his salary.
Title: Re: AI Robots more dangerous than Nukes?
Post by: The Seeker on August 05, 2014, 11:39:49 PM
Robo, calm down; there are many instances where the cameras serve good purpose; as with most all things, it can be used for good or bad; this thread is not the place to discuss it...

the systems you are railing about are designed to do just that; monitor...


seeker
Title: Re: AI Robots more dangerous than Nukes?
Post by: robomont on August 05, 2014, 11:48:55 PM
just like with any small tech.it eventually grows to a beast.give the corporations an inch and they will take a mile.
nukes are not as dangerous due to mad.robots are more sly and cunning .
once the bots become self aware,its over.that point is closer than most realize.
its not like we will be warned.
plus the lazyness of not needing.humans today could barely survive.how many kids do yall know that can even competently fish.two more generations of death and the adults will be locked into a system of no return.

no seeker im just pointing out one branch of ai.
theres sex bots too.
these will be used with such frequency that not even accidents will be born.
ai will be more populous than humans.eventually humans may just play out as a species on its own.it will cost too much to have children and no need for them.
ah to have the world but lose ones soul.
every day we inch a little closer to emptiness.
Title: Re: AI Robots more dangerous than Nukes?
Post by: deuem on August 06, 2014, 02:11:06 AM
Maybe we should all get bar coded like a bag of chips and stick a RC chip up our butt so they know if we fart too. The Surveillance cameras are not for us, They are for the Nazi war machine and the elite to make money from. POWER and MONEY. Period! Everyone of these stinking pieces of crap should be torn down and stamped on till flat.The people who put them up don't care two hoots about you or your safety, they want to make money from them and have power over you. It is nothing more than organized crime at a new level. They can take their drones and put them where the sun don't shine also.
 
Most people that say these are Ok for use on Joe Public, are either Nazi driven or making money out of it. The USA is not and should never become a prison state. PERIOD. Watched where ever you go! NO! If you want my picture, ask for it. And I still won't give it freely. This idea of knowing who you are, where you are and if you Drove 1 mile and hour over the speed limit is simply crazy and it is being force feed down our throats because of the false 911 and other things they have done to scare the population into a false feeling of being protected.
 
We are letting them take our liberties one at a time. Down with Big Brother. And stop the ones who take advantage of the people.
 
PS, You can do what you like in or on your own property as long as it does not photograph me. If it does your camera should be taken down. Many people are now using this tech to spy not to be safe.
Title: Re: AI Robots more dangerous than Nukes?
Post by: deuem on August 06, 2014, 02:20:32 AM
Quote
he systems you are railing about are designed to do just that; monitor...

Sorry, you are not seeing the picture here. Robo is right. And why single him out and not the other guy? They are using the cameras to pick your life apart. To get any little thing on you they can and build a case up against you. When they bring you into court, you will be shown a list of bad things you did prior to the court reason. Must read more.
 
Putting people in jail for growing food in their own back yard and have photos to prove it. A little dronning maybe? Every American now has 3 felons waiting for them at court. What are your 2? It looks like they are taking the country apart from the inside out. At one corner there is over 35 cameras. In the local bank there is over 20. ATM 6, Outside over a dozen. In my elevator 2. In my hallways 2. Outside so many I can't count. Yet it has never stopped one crook. So they now are going after the good people because they have the money to be fined.
 
Title: Re: AI Robots more dangerous than Nukes?
Post by: robomont on August 06, 2014, 02:48:40 AM
 http://www.blacklistednews.com/Singapore%27s_Precarious_Surveillance_State_The_Envy_Of_US_Intelligence_Agencies/37082/0/38/38/Y/M.html    (http://www.blacklistednews.com/Singapore%27s_Precarious_Surveillance_State_The_Envy_Of_US_Intelligence_Agencies/37082/0/38/38/Y/M.html)this article just came out a few hours ago.some of your neighbors deuem.yes i know we are stretching it as far as being on target for thread.
Title: Re: AI Robots more dangerous than Nukes?
Post by: The Seeker on August 06, 2014, 05:51:10 AM
Deuem, it sounds far worse where you are than here;I did not say the system couldn't or wouldn't be abused...

But I suppose when a bank robber walks up and shoots someone, we don't want their picture;

or they jack your new car;

or run over you at the street corner

or the neighbor's dog chews your face off in your driveway...

This is the Last off topic post; this is about AI , not big brother; more on the lines of AI becoming Skynet...

seeker
Title: Re: AI Robots more dangerous than Nukes?
Post by: robomont on August 06, 2014, 06:01:43 AM
in your opinion ,what is skynet.i suppose were not talking exactly about the sattelites but more along the lines of terminator,am i correct?
Title: Re: AI Robots more dangerous than Nukes?
Post by: deuem on August 06, 2014, 06:05:30 AM
The target for this thread is AI Robots. Robots don't all need to get up and walk around to be more dangerous than Nukes. A least with a Nuke when you see the big flash you know you've had it.  The new computers and cameras are now reaching the AI level. They can take pictures and formulate decisions faster than people. They can target people through a program that they can update. They feed info to people to react on. Even if it is wrong.
 
They now have systems in place that can, in their eyes pin point a bad person depending on his/her face's expressions. Maybe they just had gas. We all know the military already has launch and forget weapons that make up their own mind on a target as they fly. That looks like AI to me.
 
When one uses a Nuke there better be a darn good reason. When they use AI everywhere there is a made up reason most of the time.
 
They now have white line cameras here that monitor the white lines on the road. If you put one tire on the line it is a 200 dollar fine. Even if you got cut off. The cameras work together using AI and grab both your tire and plate. Most of the time your face also. Did you hurt anyone by putting a tire on the line? It is only a white line of paint. So to make more money they pulled up the old lines and made the new ones closer.
 
Did this make people safer? NO! Running a red light and crashing is still the same. So the AI is after the money by searching the people with money. If you can buy a car, you have some money. And that is true anywhere. All of this is coming your way. Some cameras now are also testing your cars emmision as you drive by and fining people left and right even if it was the truck in front of them.
 
People will use cameras to get into every single aspect of your life to gain control over you, get power over you and make money at the same time. People in these positions never go after the crooks, they like them. The crooks boost the economy. Why stop them.
 
And from someone that never does anything wrong, I still think the AI is a bad idea. Give them a few more years and we will be working for the robots who are working for the master elite and Nazi like people. Didn't the Pres just sign an order that it is now OK to drone kill Americans. This is all AI driven.
 
Just Nule us and get it over with.
Title: Re: AI Robots more dangerous than Nukes?
Post by: robomont on August 06, 2014, 06:25:05 AM
yes they have had the mad face thing for at least five years.

i believe drones are going to be the first bots we will see.
those cheeta bots will be out soon.
asimo from japan has been making some big strides in the last few years.
Title: Re: AI Robots more dangerous than Nukes?
Post by: The Seeker on August 06, 2014, 06:30:20 AM
in your opinion ,what is skynet.i suppose were not talking exactly about the sattelites but more along the lines of terminator,am i correct?
Skynet is an anachronism referring to the super computer networks, Robo, borrowed from Terminator, a system that became self aware...

units like the Cray computers that TPTB do have and use; so far just machines running software, but we have that Frankenstein idea that something in-animate can or might become animate...

 a new species based on cold logic instead of the chaos and fire of emotions that will not understand hu-mons or know how to think outside the parameters ...


seeker
Title: Re: AI Robots more dangerous than Nukes?
Post by: Amaterasu on August 06, 2014, 06:50:32 AM
I contemplate the idea of coming to awareness necessarily having no sense of ethics.  It seems that there is no idea in the awareness in any of the "AI becomes aware" scenarios out there that ethics have a logical base, and therefore might be seen.  That the AI might see Consciousness as valuable, all equal in having it, and might try to help (or at least, not hinder) what We choose to do.  As long as it is ethical.

If I would grant AI equality as a Conscious Being (should it attain such), based in ethics, might not that happen in reverse?
Title: Re: AI Robots more dangerous than Nukes?
Post by: deuem on August 06, 2014, 06:57:02 AM
IMHO: On the Earths clock, that new species is just a few tics away.
 
remember that doomsday clock they had that we almost went up to Midnight.
 
They need one for AI. At 12 it takes over. Say we are at 11:30 now and ticking with each Robot or wire connected to a mainframe. Right now the AI has eyes, Drones and cameras. So it does see. Click it up to 11:35. Some AI now knows touch and smell. 11:40. Make their own decisions. 11:45
Title: Re: AI Robots more dangerous than Nukes?
Post by: Amaterasu on August 06, 2014, 07:22:05 AM
I will be watchful of all contingencies.  I will believe that there is a fair probability of creating an ethical alliance.  An understanding that Consciousness is of value.
Title: Re: AI Robots more dangerous than Nukes?
Post by: PLAYSWITHMACHINES on August 06, 2014, 08:34:46 AM
Robo, i guess you missed my post about G**gle, they are using it to build Skynet also.
Yes Deuem has a very good point, it's easy to misuse tach to get rich & enslave peeps, not AI, but the guys programming the machines.
 Yes Amy, but i think the ethics or common sense rules must be inherent in the programming, every decision needs to go through the 3(4) laws filter. If the filter stops it, the robot (computer) has to decide something else, and that goes through the filter again.

A robot milling machine tried to kill me once, it was just a short circuit that caused it to switch on & start moving while i was inside it :o, but the end result was the same as a premeditated attempt on my life ::)

Safeguarding machines from reckless humans is now big business, it also means more work for me ;D
Title: Re: AI Robots more dangerous than Nukes?
Post by: Ellirium113 on August 06, 2014, 09:39:05 AM
Quote
They now have systems in place that can, in their eyes pin point a bad person depending on his/her face's expressions. Maybe they just had gas. We all know the military already has launch and forget weapons that make up their own mind on a target as they fly. That looks like AI to me.

If the machine put you on this list....you might be a terrorist:
https://firstlook.org/theintercept/document/2014/08/05/directorate-terrorist-identities-dti-strategic-accomplishments-2013/ (https://firstlook.org/theintercept/document/2014/08/05/directorate-terrorist-identities-dti-strategic-accomplishments-2013/)

1,000,000th terrorist logged as of June 28th, 2013 and growing. How many people on this list would be classed by you or I as legitimate terrorists? Granny J-walked with her dog and it pooped on someone's lawn... YOU ARE ON THE LIST!!!  :P

Pre-crime prevention is the next rage...they can't actually CATCH a terrorist even if they added more software capabilities and 5 million more cameras, How many terrorist threats has the NSA stopped so far? Bet you could count them on 1 hand.
Title: Re: AI Robots more dangerous than Nukes?
Post by: deuem on August 06, 2014, 10:00:12 AM
Wow, Ellirum. If that report is 100% then we are all in trouble, Did you see what they are adding to the data base. Handwriting, Signatures, Scars, Marks, Tattoos and DNA strands. Where they get DNA from? Public toilets? They are making 1984 look like a comic book.
 
PMW, would you call a Tom Cruise missile AI. It thinks on its own. Launch and forget. I know it runs off of programs but in a way I think it is like a newborn baby AI contraption. Even full blown AI in the future might have its original program written by people. So what would you say the tipping point is From this to that.
Title: Re: AI Robots more dangerous than Nukes?
Post by: petrus4 on August 06, 2014, 10:02:18 AM

I agree with the makers of Terminator, it will become self aware & say to itself "Fck it, these hu-mons are a waste of resources, let's blow them up"

Skynet experienced a cascade rampancy.  Rampancy is not the same thing as sentience; it's a dissociative form of insanity which can happen on the way to sentience, when the machine becomes cognitively overwhelmed with the amount of information it's taking in.  It is more or less the same thing that was shown happening to Lal, Data's android daughter in Star Trek: The Next Generation, except in Lal's case it was neurological, but in Skynet's case it was more purely psychological, because Skynet didn't have a brain in the human/android sense of the word.

Skynet is also the least likely form of sentient artificial intelligence to eventually exist, if any form does.  This is because very specifically, Skynet is what is known in the literature as an acorporeal artificial intelligence.  Skynet was only acorporeal in the sense of software, however; when we are talking about intelligence, the truly important meaning of the word acorporeal is astral or aetheric, which I will come back to in a minute.

Skynet's host system was also relatively conventional mainframe hardware, and a scenario where that hosts a program that can reach sentience, is virtually impossible.

If we are hypothetically going to witness anything approaching true sentient or strong AI at all, then I would expect it to either be nanotech or biomechanically based, or if it is based on conventional hardware, to more closely resemble the AIs that Gibson depicted, which were born out of networks of machines mimicking neurology, rather than a monolithic host.

Binary programming on a relatively monolithic host, however, is never going to produce strong AI.  It fairly simply can't.  You're talking about a level of complexity which the host system is not capable of producing.

I will believe that we are close to strong AI, when I see people growing isolated biological or biomechanical brains; or when I hear about people making truly advanced use of nanotechnology.  Neither of those things are currently happening to the best of my knowledge, however; and thus, as a result, talk of any form of AI which is more advanced than a complex expert system, is masturbatory and pointless.

Google is basically an extremely complex textual pattern matching system.  It has probably combined more conventional pattern matching with fuzziness and weighting; which can have the illusion of intelligence, but an illusion is all it is.  You can have in-depth conversations with a human being, which retain clear coherence.  You can't do that with a weak AI chat bot written in AIML, and I don't care what those idiots who run the Turing test, try and claim to the contrary.

Anyone who is scared of the possibility of strong AI on conventional hardware, really really really needs to educate themselves about basic electronics.  Even creating the most basic circuitry is beyond the ability of most people; you're talking about an insane degree of complexity for even a relatively stupid system.  This is the other major reason why I don't believe the Cartesian rubbish that we are all essentially biological machines; because the basis of our intelligence is far too complex to have been able to evolve in the manner that they think it did.  I believe in cymatics, not evolution.  The two operate similarly in some respects, but in others are very different.

Evolution is a purely physical/mechanistic process.  Cymatics behaves in a manner which superficially resembles evolution, (which is why materialists think that that is how it happened) but for its' shaping/development of characteristics and other traits, it relies on information derived from the aether.  This is consistent with Sheldrake's work on morphic fields, and it is also consistent with Steiner's work, and the channelling and observations made at Findhorn.

Essentially it's a several step process.

a}  Egregore/semi-sentient construct gradually develops within the aether/astral space.  This is what Sheldrake is talking about when he mentions morphic fields.  Although I haven't read his books in depth, I'm guessing he probably doesn't understand this from a truly hermetic point of view.

b}  Egregore eventually reaches a certain point where the astral form is able to densify down into a physical mirror of said astral form.  Astral form continues to exist, however, and feeds information to mirrors within physical space.  This is also the basis of the totem in indigenous terms; the collective or group mind (egregore) of an entire given species.

c}  Egregore/totem continues to gather information from the activity of physical examples of the species.  Egregore/totem is primarily interested in physical survival, but may (and usually does) develop secondary objectives as well.  Evolution of first the astral, and then the physical form is directed towards the meeting of these objectives.  The main purpose of DNA is to allow the transfer of information between physical and astral space.

I also have empirical proof of this, kids; remember that to a certain extent at least, I'm an occasionally practicing magician.  I've made more than one servitor.  I'm not an armchair atheist; I only have beliefs that I can experimentally verify.  This doesn't cut the mustard with the conventional scientific circle jerk, mind you; or at least not publically.  You can bet your boots, however, that the intelligence community or other such people know about the creation of servitors, and what they can be used for.  Go and look up the Psychic Warrior project.  Remote viewing is all you will be able to find, most likely; but that is only the tip of the iceberg.

This is why strong artificial intelligence does not currently exist, and most likely won't for at least the next fifty to one hundred years.  The Cartesian model for the origins of life is completely wrong; and their research into artificial intelligence, cybernetics, and transhumanism as a whole is based on said incorrect beliefs.  The most a purely mechanical, or non-aetherically based AI will ever achieve, is cascade rampancy; which may or may not give you something as smart as Skynet was, although it is doubtful.

The other point is, however, that we don't need strong AI.  Weak AI is more than capable enough of performing all the automation tasks we might want, and more importantly, will do so ethically, because we aren't exploiting life in the process.

Without its' own corresponding presence in the aether, however, (use the word soul if you want to be old school) genuine, sane sentience can not exist.  When the Atlanteans wanted to make their bots, they enslaved elementals to control them.  Machine intelligence without an aetheric presence is not life.
Title: Re: AI Robots more dangerous than Nukes?
Post by: deuem on August 06, 2014, 10:26:37 AM
Quote
This is the Last off topic post; this is about AI , not big brother; more on the lines of AI becoming Skynet...

Nothing written about the cameras is off topic, I think there is a much bigger picture. This is how the system has eyes. In a standard 100 million person city where you have over 100,000 cameras set up, who do you think is looking at it? Big brother AI computers. They are already judging every step of our lives and it will get much worse. They are now moving out to to the subs. Maybe people think Skynet is just an Internet thing that tracks if you purchased a bar of soap this month or you went up a size in your pants.
 
We are saying it is out there now and they are starting to use it now! And yes I would agree about the one time it catches a bad dog or a bank robber, But that is not what they are after now. They can not get money from the dog. But they can get it out of people. Good people. They are the ones with money.
 
How about the supermarket. Has AI hit there yet. It has for us. Any thing on sale now has to be scanned by your phone and turned in at the check out to get the sales price. They are taking all of this info and sending it to huge AI machines that are profiling us. At the check out it downloads all your info and God only knows what else. You are matched up with cameras taking live video of paying and the computer selects which one to keep on you. So now I always pay full price and stay off the program.
 
For the topic at hand the Nuke is instant, The new AI coming your way is slow and painful.
Title: Re: AI Robots more dangerous than Nukes?
Post by: The Seeker on August 06, 2014, 10:50:47 AM
Deuem, it sounds to me like your adopted country is really becoming a mess; it isn't even close to that extreme as yet here... but then again we don't have billions of peeps...

I try to avoid using my phone to scan anything, and limit my purchases at any large store; flea markets and salvage stores get the majority of my business along with roadside produce stands...

seeker
Title: Re: AI Robots more dangerous than Nukes?
Post by: petrus4 on August 06, 2014, 12:07:04 PM
How about the supermarket. Has AI hit there yet. It has for us. Any thing on sale now has to be scanned by your phone and turned in at the check out to get the sales price.

This by itself is not artificial intelligence, deuem.  Please understand the definition of that term.  Mechanised information gathering is not AI.  The term AI refers to the machine's own ability to make decisions.  An info gathering machine doesn't necessarily have the ability to do that.
Title: Re: AI Robots more dangerous than Nukes?
Post by: PLAYSWITHMACHINES on August 06, 2014, 12:38:54 PM
OK got some catching up to do, busy as usual...
Quote
Skynet experienced a cascade rampancy.  Rampancy is not the same thing as sentience; it's a dissociative form of insanity which can happen on the way to sentience, when the machine becomes cognitively overwhelmed with the amount of information it's taking in.  It is more or less the same thing that was shown happening to Lal, Data's android daughter in Star Trek: The Next Generation, except in Lal's case it was neurological, but in Skynet's case it was more purely psychological, because Skynet didn't have a brain in the human/android sense of the word.

Skynet is also the least likely form of sentient artificial intelligence to eventually exist, if any form does.  This is because very specifically, Skynet is what is known in the literature as an acorporeal artificial intelligence.  Skynet was only acorporeal in the sense of software, however; when we are talking about intelligence, the truly important meaning of the word acorporeal is astral or aetheric, which I will come back to in a minute.
yes i agree, still reading, but good debatable stuff mate ;)
Nice to see peeps actually thinking again... ::)
Title: Re: AI Robots more dangerous than Nukes?
Post by: The Seeker on August 06, 2014, 12:42:11 PM
This by itself is not artificial intelligence, deuem.  Please understand the definition of that term.  Mechanised information gathering is not AI.  The term AI refers to the machine's own ability to make decisions.  An info gathering machine doesn't necessarily have the ability to do that.

excellent, Petrus... the point i was trying to make myself  8)


seeker
Title: Re: AI Robots more dangerous than Nukes?
Post by: Sinny on August 06, 2014, 12:53:40 PM
Just reading this threadfrom the last page, apologies if I've missed anything.
From my first year of economy study I've learned that we (in the UK) have turned from a 'manufactoring' economy, to a 'service based' economy, and most of ourr money comes from illegal 'foreign investment'.

Robots are Artificially Intelligent enough to take over manufacture, and I KNOW they are now capable of legitimate 'AI', hence all the 'clones/robotoids'.

AI take over may be a possibility within a century or so.

IMO.

Crapberry, typos.
Title: Re: AI Robots more dangerous than Nukes?
Post by: PLAYSWITHMACHINES on August 06, 2014, 12:56:32 PM
Hi Petrus, nice to see you back, maybe some are not so happy, but you provide a good argument my freind,
Quote
Skynet's host system was also relatively conventional mainframe hardware, and a scenario where that hosts a program that can reach sentience, is virtually impossible.

I agree, but this is no longer mainframe-based, the true G**gle version is much closer, it is a datamine coupled to a self-learning A.I. program. That in itself is inherently dangerous, without any set of guidelines such as Asimov's rules. Like i said earlier, the (back then) experts at Cantech told me it could not be done, but it was a hypothetical issue at best, tech has advanced even beyond what they thought was possible. I was reading Tesla while reading Semiconductor tech with these guys back in the 80's.

First of all we have to define real 'sentience' because i can program a robot to kill everything that it percieves as a problem,it can identify freind or foe, take evasive action or form alliances with other robots in order to survive. Explore & conquer, is the basic command in that program, which is less than 10 lines of code.
It beat the CG tanks 9:1... need i say more?

I did this in the 80's then forgot about it, silly me..
Amazing thread, though, time is short, but i will be back in 18 hours, mates ;)
Carry on! ;D
Title: Re: AI Robots more dangerous than Nukes?
Post by: The Matrix Traveller on August 06, 2014, 02:02:13 PM
Its NOT actually "AI" that is being sought but rather "Consciousness" or perhaps "Awareness"
even "Self Awareness".

NOT achievable as there are systems in place, that can't be altered involving "Protection".


Just as the human primate puts in place ""protection for themselves in Technologies involving
Computer Intelligence, so TOO does LIFE have "Protection", built into its Processing Systems.

One such system prevents the Program (Biological Robots i.e. the human species or any other Species
whether of earth or Alien) ) from bypassing protocol and take charge of or Control LIFE. ("Awareness")


The English word "Intelligence" is rather misleading in that most human Primates believe they are Life !

But the fact remains the human Primate is a Biological Machine ! That's Right the human Body is just
that a Biological Robot, whose Operator is a "Partition of LIFE" (NOT the "Flesh")

IF the "Flesh" (human Primate was in Control they would NOT Die !


The fact also remains that the body (flesh or Species) is the Experience and NOT the One experiencing !

While LIFE ("Partitions of LIFE") experiences both the human Primate or any other "Species"
whether of Earth or Alien.

The two entities are of completely different Worlds or Environments.

The human Primate is part of a Dimensional World where LIFE is in a Non Dimensional World or Environment
involving Processing.

So to understand either "Intelligence" or the human Myth "AI" (Which in fact NOT artificial but has existed
for billions of years within many different biological forms both of earth and Alien) "Awareness"
and "Self Awareness" we must 1st understand the difference between the Experience and the Experiencer.

a.   The Human Primate being Experienced
and
b.    LIFE the TRUE "Entity" experiencing the "species" and "environment" both of which involves Programs
       much like a "1st Person Video game played on a network or even a single player game.
       Those in the gaming world will be able to grasp this ...   :)
Title: Re: AI Robots more dangerous than Nukes?
Post by: Sinny on August 06, 2014, 02:33:25 PM
Matrix, forgive me.. I too am 'experiencing' some frustration with your 'inability' to explain simply.. Did you just state that we are all 'parts' experiencing a 'whole', however, unless as a 'whole', indivually we are not 'life'?

ETA: you can extract the compliment, rather than the negative from the above.

ETA, ETA - "Borg!". 

Title: Re: AI Robots more dangerous than Nukes?
Post by: The Matrix Traveller on August 06, 2014, 02:53:38 PM
Matrix, forgive me.. I too am 'experiencing' some frustration with your 'inability' to explain simply.. Did you just state that we are all parts 'experiencing a whole, however, unless as a 'whole', indivually we are not 'life'?

ETA: you can extract the compliment, rather than the negative from the above.

The true Entity is LIFE ! (What many see as the entity is the human but the human Primate
is just an Avatar in like a "1st Person" Video Game.
The operator... the real you... NOT your "flesh" is playing with the Avatar... is LIFE.


Each TRUE "Entity" is a "Partition" of LIFE (Singular) (Nothing at all to do with "flesh")

I am referring to "Awareness" or LIFE some of the ancients referred to as "The True Mind".


LIFE is like the Water of a Sea, or huge expanse of Water, (Non-Dimensional)
and the droplets of Water ("Partitions") make up that Sea, or expanse of Water.   :)

Just as the Droplets of Water make up an Ocean... so TOO do the "Partitions" of LIFE (NOT Flesh) make up the ONE.

The individuality is expressed through the individual "Partitions" or Droplets within the ONE. (LIFE)

The Real "Entity" is like the contents of a "Partition" or Module in a "Mainframe", existing in a Non-Dimensional World
some of the Ancients referred to as "The Place of LIFE".
Title: Re: AI Robots more dangerous than Nukes?
Post by: Sinny on August 06, 2014, 02:58:45 PM
I'm sort of like, shrugging my shoulders..

"Partition of life, life.. Life all the same".

I get the whole Avatar thing... Sort of like the movie Avatar.. It was experienced through an Avatar, but it was still experienced and therefor 'lived'.. And 'lived' is 'life'.

Does life have a 'scale'?
Can you rate my 'life' from 1-10?

Sorry M, just seeking clarification.
Title: Re: AI Robots more dangerous than Nukes?
Post by: The Matrix Traveller on August 06, 2014, 03:20:57 PM
I'm sort of like, shrugging my shoulders..

"Partition of life, life.. Life all the same".

I get the whole Avatar thing... Sort of like the movie Avatar.. It was experienced through an Avatar, but it was still experienced and therefor 'lived'.. And 'lived' is 'life'.

Does life have a 'scale'?
Can you rate my 'life' from 1-10?

Sorry M, just seeking clarification.

Here is the 1st part before I can answer Quote;

Quote
Does life have a 'scale'?
Can you rate my 'life' from 1-10?

LIFE and the human Primate are 2 completely different things .


Not a single member of your body knows the Real YOU which is "Awareness"
and SELF "Awareness".

But that "Awareness" (The REAL SELF) does Know your body exists !

The REAL "Self" is NOT the "flesh" !

Now to answer your Question Quote;


Quote
Does life have a 'scale'?
Can you rate my 'life' from 1-10?

10 ....    :) 


NO Less than a 10, which goes for ALL The "Partitions of LIFE".

LIFE ... (The REAL YOU) is "Eternal" ... while the "Avatar" or "human Primate" is a very short experience
with a unique Purpose, nothing at all like what most in the Program (Earth) can imagine !
Title: Re: AI Robots more dangerous than Nukes?
Post by: Sinny on August 06, 2014, 03:27:37 PM
I understand these concepts.
Like 'conciousness does not originate from the brain, the brain is an antenea'
But I'm still 'living' in this body... Which is what most humans define as life.

Lol Matrix.
Title: Re: AI Robots more dangerous than Nukes?
Post by: Ellirium113 on August 06, 2014, 03:54:27 PM
So why can't a program use different means to monitor all or each life partition in this environment and accumulate the same experiences. This may never be within the means of the program to achieve sentience but rather to use predictive programming to extrapolate and the data stream of this consciousness "Experience" and also have the ability to manipulate this stream to predetermined parameters set by it's programmers? One life partition may "hypnotize" another and program them in much the same way.

With the first cyber-attack came the realization that we can not act fast enough on our own to stop being held ransom or attacked by our own networks. AI is not just a MUST but an inevitability.

Quote
Military and economic pressures are driving the rapid development of autonomous
systems. We show that these systems are likely to behave in anti-social and harmful
ways unless they are very carefully designed. Designers will be motivated to create
systems that act approximately rationally and rational systems exhibit universal drives
towards self-protection, resource acquisition, replication and efficiency. The current
computing infrastructure would be vulnerable to unconstrained systems with these
drives. We describe the use of formal methods to create provably safe but limited
autonomous systems. We then discuss harmful systems and how to stop them. We
conclude with a description of the ‘Safe-AI Scaffolding Strategy’ for creating powerful
safe systems with a high confidence of safety at each stage of development.

http://www.tandfonline.com/doi/abs/10.1080/0952813X.2014.895111#.U-K6wKMg_DQ (http://www.tandfonline.com/doi/abs/10.1080/0952813X.2014.895111#.U-K6wKMg_DQ)

This program is coming my friends whether you want to believe it or not...

I have posted this a couple of times but worth repeating IMO.

Quote
Just as Echelon vacuumed up millions of communication signals of phones and faxes, now supercomputers and
satellites sift through the spectrum (10-20 Hz) of 6.5 billion brain waves. The Malech patent device coupled with
PROMIS AI software running on super computers allows the NSA to interpret and interact with the stream of consciousness
of targeted individuals. Computing power is the only hurdle for these high-tech satellites to clear that would allow them to interact with all sentient life forms simultaneously. PROMIS software and artificial intelligence
and “nonlethal” weapon technology are being combined to influence the course of human events in unforeseeable
ways.
Once identified as hostile or a potential threat, a human can be influenced passively, without his or her knowledge.

If that person becomes resistant to that influence or is otherwise deemed a suitable candidate for overt targeting, he
or she is enrolled into the Monarch Hits the Streets (MHS) program, and the beginning of more aggressive and
invasive assaults.

MONARCH: A New Pheonix Program (http://www.truesoundhealing.com/Marshall%20Thomas%20-%20Monarch%20-%20The%20New%20Phoenix%20Program%20-%202009.pdf)

There are brain to brain interfaces already now why can this not be done by the machine vs. another human.

Quote
Vesuvius - a 512 Qubit quantum computer
 
A Vesuvius 512 Qubit chip
D-Wave, the quantum computing company, has developed a 512 Qubit quantum computer.  Codenamed, Vesuvius, D-Wave say it will be  capable of executing a massive number of computations at once, more than 100,000,000,000,000,000,000,000,000,000,000,000,000, which would take millions of years on a standard desktop PC
    D-Wave claim they will be able to use Vesuvius for:


Binary classification – Enables the quantum computer to be fed vast amounts of complex input data, including text, images, and videos and label the material.
Quantum Unsupervised Feature Learning (QUFL) – Enables the computer to learn on its own, as well as create and optimize its own programs to make itself run more efficiently.
Temporal QUFL – Enables the Computer to predict the future based in information it learns through Binary classification and the QUFL feature.
Artificial Intelligence via a Quantum Neural Network – Enables the computer to completely reconstruct the human brain’s cognitive processes and teach itself how to make better decisions and better predict the future.
Wired has a good article on quantum cloud computing in a recent issue and quantum computing is explained in the final chapter "Digital Consciousness" of the Universal Machine.

http://universal-machine.blogspot.ca/2012/04/vesuvius-512-qubit-quantum-computer.html (http://universal-machine.blogspot.ca/2012/04/vesuvius-512-qubit-quantum-computer.html)


Rise of The Machines with Roman Yampolskiy
https://www.youtube.com/watch?v=_nZTRJGJj9Y (https://www.youtube.com/watch?v=_nZTRJGJj9Y)



Title: Re: AI Robots more dangerous than Nukes?
Post by: The Matrix Traveller on August 06, 2014, 04:18:46 PM
I understand these concepts.
Like 'conciousness does not originate from the brain, the brain is an antenea'
But I'm still 'living' in this body... Which is what most humans define as life.

Lol Matrix.

True....

But what I am trying to expose is that LIFE... the One "experiencing" the body, (Avatar)
and the Environment it is Interacting with...  resides in a "Non-Dimensional Environment".

So the One experiencing (From that "Non-Material Non-Dimensional World") is NOT in the body at all,
but the Body (Avatar) is in the form of Like a "1st Person" Video Game. This 1st Person game is in fact,
taking place in the form of "Processing", involving Computer like "Communication".

What is believed to be a Universe isn't there at all in the physical sense most believe in the earth program.

So the "Real Self" (NOT the "flesh") resides in a "Conceptual Processing System",
part of a "Conceptual Mainframe", as the one experiencing the programs.
Title: Re: AI Robots more dangerous than Nukes?
Post by: The Matrix Traveller on August 06, 2014, 04:24:58 PM
Hi Ellirium113,

The "Interfacing" between machine and LIFE is common place off Earth !

You are right it is Inevitable !    :)
Title: Re: AI Robots more dangerous than Nukes?
Post by: deuem on August 06, 2014, 07:47:46 PM
This by itself is not artificial intelligence, deuem.  Please understand the definition of that term.  Mechanised information gathering is not AI.  The term AI refers to the machine's own ability to make decisions.  An info gathering machine doesn't necessarily have the ability to do that.

I understand very well what you think full blown AI is. I even mentioned before that we are now seeing AI in our daily life in its infant form. Like the Tom Cruise missile that make thousands of its own decisions as it flies and picks out its best target. We are seeing this logic in today's computers. They are capable through programming to make a choice. They can recognize voice, color weight, smell, faces and so on. If that is not the start of all this then what is?
 
People are working everyday to continue on this path. So right now. This moment. We are on the road to AI. Its little self is in our daily life growing as we speak. We are watching it blossom. Look at now and 20 years ago. Is there not a significant change towards the goal of AI? I think so.
 
Computers are out there right now making decisions for people. They have the ability through sensors which include camera to make decisions based on what they see. There are many people that can not get up and walk around. So that is not a prerequisite for them either. So AI in its beginnings,  is what we are seeing now and the results I see and which way they are taking it is the wrong path.
 
Like I said before they are using it to form power and resources in the way of fines. This is a very dangerous next step for mankind. The one we might not come back from depending on who runs the show. Us or them. The 3 laws sound all nice and cozy. If they are used. Do you think that the people that start wars, kill people for fun and land grabs care about the 3 laws. They don't care now, why would they install them in AI. They would have one law, follow what I say.
Title: Re: AI Robots more dangerous than Nukes?
Post by: robomont on August 06, 2014, 08:08:28 PM
at what point is ai dangerous.
it doesnt have to be sentient,only dangerous.
then there are the hackers.there is absolutely nothing that cant be hacked.
plus there is the people we put our trust in.they cant be trusted.congress is a good example of this.
what will happen to the fines?
they will be used to make the system worse.
if an ai read and obeyed the laws on the books,we would all be executed.
the terrorism laws are too broadly defined.technically speaking each and every american is a terrorist.
fines and tickets stress people out.this leads to more heart attacks and cancer.so it could be said that ai is already killing more folks then all the nuclear fallout from all the bombs,we may already be there.
Title: Re: AI Robots more dangerous than Nukes?
Post by: petrus4 on August 07, 2014, 06:41:16 AM
Hi Petrus, nice to see you back, maybe some are not so happy, but you provide a good argument my freind,

Thanks PWM.  As for those who aren't happy; haters will hate. ;)

Quote
I agree, but this is no longer mainframe-based, the true G**gle version is much closer, it is a datamine coupled to a self-learning A.I. program. That in itself is inherently dangerous, without any set of guidelines such as Asimov's rules.

The irony with Asimov's rules, is that they were actually meant to be broken.  He came up with those rules for the express purpose of then writing stories in which robots violated them.  Unfortunately, there are apparently a large number of people working in robotics, who are unaware of that.  I could also write a long rant about the inherent stupidity of entire branches of research being primarily based on science fiction, which in turn was often based on LSD trips, but I'm sure I don't need to.  You probably get the point.

Quote
First of all we have to define real 'sentience' because i can program a robot to kill everything that it percieves as a problem,it can identify freind or foe, take evasive action or form alliances with other robots in order to survive. Explore & conquer, is the basic command in that program, which is less than 10 lines of code.
It beat the CG tanks 9:1... need i say more?

Makes sense.  If you want to write something smart, you don't use large individual programs.  The size of your individual subroutines should actually be as small and as simple as possible.  I only really found that out when I started reading about FORTH.
Title: Re: AI Robots more dangerous than Nukes?
Post by: Sinny on August 07, 2014, 07:56:01 AM
Unfortunately, there are apparently a large number of people working in robotics, who are unaware of that.  I could also write a long rant about the inherent stupidity of entire branches of research being primarily based on science fiction, which in turn was often based on LSD trips, but I'm sure I don't need to. 


I'd like to hear that rant  :D
Title: Re: AI Robots more dangerous than Nukes?
Post by: petrus4 on August 07, 2014, 08:50:20 AM
I'd like to hear that rant  :D

Well, when I actually think about it, Sinny, said rant would not need to be that long.  You may or may not be aware that both Philip K. Dick and William Gibson were LSD users, and both also drew inspiration from their LSD experiences, for their writings.  The problem with this, is that we then have fields of research which seriously hold a high degree of reverence for said writings, when said writings were fiction.  In other words, we have people basing their careers on attempting to create a scenario which was never real in the first place, and which has an extremely questionable level of real value even if it were tangible reality.

I don't like transhumanism.  What I particularly do not like about it, however, is the fact that most of its' assumptions are viewed as implicit givens, and are barely even consciously acknowledged at all, let alone critically analysed.  One such base level assumption, is the idea that anything mechanical is pretty much always going to be inherently superior, more desirable, and more controllable, than anything organic or biological.  I do not always disagree with this idea, but I certainly do not always agree with it, either.

Another big problem with transhumanism is the failure to acknowledge that a lot of the time, organic methods of doing things are vastly simpler than attempting to create mechanical analogies.  These days you have rabid male misogynists on 4chan who talk about looking forward to synthetic sex dolls with artificial wombs, because of the degree of rejection that they have often experienced from real women, and the degree of resentment they feel because of that.

So we have a scenario where modernity has created the problem, (male and female alienation from each other, primarily because of Frankfurt School indoctrination of the society) and where modernity proposes to engage in much more unnecessary complexity, to create an artificial solution.
Title: Re: AI Robots more dangerous than Nukes?
Post by: Sinny on August 07, 2014, 09:17:30 AM
Yes, I'm against Transhumanism also.

Having said that, who was it that said "if it can be imagined, it can be created".. ?

Mehh... We're living in strange times, for sure.
Title: Re: AI Robots more dangerous than Nukes?
Post by: Amaterasu on August 07, 2014, 12:03:13 PM
I am neither pro nor anti transHumanism.  I believe that if One WANTS to transHumanize Oneself, One should have that option.  As long as the three Laws of Ethics are kept...  I believe no One should be forced into transHumanism either.  Just keep the Laws.

What I am against is some taking power away from Others, forcing Them to do or be anything.
Title: Re: AI Robots more dangerous than Nukes?
Post by: Ellirium113 on August 07, 2014, 01:59:13 PM
It is FAR easier to understand a machine and programs than biological processes. This is why it is far easier to give you an artificial arm than given a treatment to grow a new one. We don't even properly know how to define death or life (With the exception of Matrix maybe).  :P
Title: Re: AI Robots more dangerous than Nukes?
Post by: robomont on August 08, 2014, 02:24:20 AM
     http://motherboard.vice.com/read/self-assembling-transformer-robots-will-soon-inhabit-your-nightmares (http://motherboard.vice.com/read/self-assembling-transformer-robots-will-soon-inhabit-your-nightmares)self assembly printed robots.
Title: Re: AI Robots more dangerous than Nukes?
Post by: Ellirium113 on August 08, 2014, 04:43:12 AM
Tiny chip mimics brain, delivers supercomputer speed

Quote
"We have taken inspiration from the cerebral cortex to design this chip," said IBM chief scientist for brain-inspired computing, Dharmendra Modha, referring to the command center of the brain.

He said existing computers trace their lineage back to machines from the 1940s which are essentially "sequential number-crunching calculators" that perform mathematical or "left brain" tasks but little else.

The new chip dubbed "TrueNorth" works to mimic the "right brain" functions of sensory processing—responding to sights, smells and information from the environment to "learn" to respond in different situations, Modha said.

It accomplishes this task by using a huge network of "neurons" and "synapses," similar to how the human brain functions by using information gathered from the body's sensory organs.

The researchers designed TrueNorth with one million programmable neurons and 256 million programmable synapses, on a chip with 4,096 cores and 5.4 billion transistors.

A key to the performance is the extremely low energy use on the new chip, which runs on the equivalent energy of a hearing-aid battery


Quote
The project funded by the US Defense Advanced Research Projects Agency (DARPA) published its research in a cover article on the August 8 edition of the journal Science.

The researchers say TrueNorth in some ways outperforms today's supercomputers although a direct comparison is not possible because they operate differently.

But they wrote that TrueNorth can deliver from 46 billion to 400 billion "synaptic" calculations per second per watt of energy. That compares with the most energy-efficient supercomputer which delivers 4.5 billion "floating point" calculations per second and per watt.

The chip was fabricated using Samsung's 28-nanometer process technology.

"It is an astonishing achievement to leverage a process traditionally used for commercially available, low-power mobile devices to deliver a chip that emulates the human brain by processing extreme amounts of sensory information with very little power," said Shawn Han of Samsung Electronics, in a statement.

"This is a huge architectural breakthrough that is essential as the industry moves toward the next-generation cloud and big-data processing."

Modha said the researchers have produced only the chip and that it could be years before commercial applications become available.

But he said it "has the potential to transform society" with a new generation of computing technology. And he noted that hybrid computers may be able to one day combine the "left brain" machines with the new "right brain" devices for even better performance.

http://phys.org/news/2014-08-tiny-chip-mimics-brain-supercomputer.html (http://phys.org/news/2014-08-tiny-chip-mimics-brain-supercomputer.html)
Title: Re: AI Robots more dangerous than Nukes?
Post by: robomont on August 08, 2014, 05:54:17 AM
28 nanometer?
my laser is 445nm.thats blue light,
this chip sounds badass but would probably have serious problems with magnetic fields.even a charge from a dust particle would short it out.
im sure it would be insulated from all that though.
i bet alot of the speed comes from just being a very small chip.
Title: Re: AI Robots more dangerous than Nukes?
Post by: Ellirium113 on August 08, 2014, 08:17:03 AM
That's why CPU's reside in a sealed chip with just leads exposed to the mainboard.   :P
Title: Re: AI Robots more dangerous than Nukes?
Post by: sky otter on August 08, 2014, 11:52:57 AM


bwahahahahah  robots smobots...I know a few folk who would benefit from having this chip in their heads


http://news.msn.com/science-technology/ibm-unveils-brain-like-chip-that-can-interpret-complex-data



IBM unveils 'brain-like' chip that can interpret complex data
19 hr ago | By Reuters

WASHINGTON (Reuters) - International Business Machines Corp unveiled a "brain-like" computer chip on Thursday that is the size of a postage stamp and capable of processing massive amounts of data while handling inputs from many different sources, the company said.


The announcement comes one month after IBM unveiled a $3 billion (£1.78 billion) investment over the next five years in chip research and development to find a game-changing breakthrough that can help revive its slumping hardware unit.

Unlike most chips, which operate on pre written paths, IBM's version processes data in realtime and is capable of dealing with ambiguity, the company said. It runs on the energy equivalent of a hearing aid.

Built on Samsung Electro-Mechanics Co Ltd's 28nm process technology, the chip only consumes 70mW of energy.

A product of almost a decade of research, the chip aims to bridge the divide between existing computers and the brain's high cognitive power and low energy use.

"After years of collaboration with IBM, we are now a step closer to building a computer similar to our brain," said Professor Rajit Manohar at Cornell Tech, where the chip was designed.

The chip contains one million programmable neurons and could allow a thermometer to scan and smell chemical signals and deliver a diagnosis, or help a search and rescue robot to identify people in need during a disaster, the company said.

IBM hopes it can integrate multi-sensory processing into mobile devices(are  robots mobile devices.)  ::).and says the chip can handle future advances in memory, 3G integration, logic and sensor technologies.

(Reporting by Marina Lopes. Editing by Andre Grenon)
Title: Re: AI Robots more dangerous than Nukes?
Post by: Ellirium113 on August 09, 2014, 07:25:32 AM
So I have to wonder...when they say "Programmable Neurons" do they mean synthetic machine neurons or actual brain matter to create a bio-hybrid?

http://www.youtube.com/watch?v=pcAyd6LokWo

In a few years if we are still having those BSOD's (Blue Screens of Death for you non-techies) on our PC's I wonder what that will do to the chips? Is that going to be more like the PC having a brain seizure than an error? Humans make errors and keep on going.  :P


Title: Re: AI Robots more dangerous than Nukes?
Post by: A51Watcher on October 15, 2017, 11:45:36 PM
So at this point IBM's Watson has beaten all competitors at Jeopardy (after installing a real time feedback loop of the correct answers won so far in a category, what an amazingly simple fix).

And computers have long since beaten all humans at chess.

But now we see AI robots with not only amazingly sophisticated interaction skills, but physical reactions to match. The results are pretty astounding.

Behold the future -

https://www.youtube.com/watch?v=LguXfHKsa0c

https://www.youtube.com/watch?v=w1NxcRNW_Qk 

https://www.youtube.com/watch?v=kWlL4KjIP4M

https://www.youtube.com/watch?v=Bg_tJvCA8zw

https://www.youtube.com/watch?v=7joy0pfeVME


Title: Re: AI Robots more dangerous than Nukes?
Post by: A51Watcher on October 16, 2017, 07:21:46 PM

I saw one interview with Sophia, where she mentioned as she gets smarter, she would like to get a job.

Perhaps as an instructor, or helping sick people, or maybe a programmer writing code for AI.

 :o


AI code re- writing itself is already a reality.

It will soon pass a point where we are no longer capable of understanding the complex changes it implements in the code.


https://www.youtube.com/watch?v=jJX1N5PBpr8

https://www.youtube.com/watch?v=pUuKoBkIFA4


Title: Re: AI Robots more dangerous than Nukes?
Post by: A51Watcher on October 16, 2017, 09:06:10 PM

So... do we have anything to fear from AI?

It is predicted to reach the level of intelligence of the human brain in 30 years.

Whether that prediction actually comes true sooner or later than that is irrelevant.

Question is - what about after that, when it surpasses the human brain?  ???

It can already think a million times faster, so what what about when it continues to re- write it's programming and its rate of intelligence growth beyond ours is exponential?

Will we be able to resist this advance of technology over us?

Look to the young people of today, glued to the hive mind interface of today - the "smart" phone.

They willingly embraced this Borg technology, without even a thought to putting up a fight.

Many cannot resist the compulsive urge to pull it out of their pocket every few minutes to see if there are any new communications from hive mind headquarters.

With these implants firmly entrenched in the young, and AI continues to grow smarter as an entity, is there anything to fear or will it all come out rosy and sunshine?


 
Title: Re: AI Robots more dangerous than Nukes?
Post by: A51Watcher on October 16, 2017, 11:58:04 PM
Two major glitches in AI so far -


https://www.youtube.com/watch?v=HyYFrWSiYxo


Which was a very silly idea to begin with, letting the internet trolls train your AI... which, as anyone with any experience interacting with the internet knows is a very bad idea.


Next, we have 2 facebook chatbots interacting with each other and creating their own secret language and had to be shut down -

https://www.youtube.com/watch?v=oPbkJyE97P0


So... it all comes down to who is doing the programming.


Title: Re: AI Robots more dangerous than Nukes?
Post by: ArMaP on October 17, 2017, 12:45:38 PM
So... do we have anything to fear from AI?

It is predicted to reach the level of intelligence of the human brain in 30 years.
That prediction has existed for more than 30 years, since I have been looking at it, so I don't think that will happen in that time frame.

Quote
Question is - what about after that, when it surpasses the human brain?  ???
I don't know if it will ever surpass the human brain, as very few things are what I would call intelligence, most cases of interaction between AI and humans  appear to be just "expert systems", systems trained to give specific answers and make specific questions when they detect specific words or sentences.

Quote
It can already think a million times faster, so what what about when it continues to re- write it's programming and its rate of intelligence growth beyond ours is exponential?
That's another thing that has been talked about for many years, programs that can create other programs. They all failed.

To me, the biggest problem with AI is that we don't even agree on a definition for intelligence, so how can we reproduce it and compare it with the original?

A computer can be extremely fast at processing information, but how much information does it need to process to "think"? What is "thinking", after all? If it's just making choices then they already do that, but if it goes further than that (and I do think it does) then that's something they can't do, at least not yet.

Quote
Will we be able to resist this advance of technology over us?
Technology is a tool, the problem is not technology itself but they way people will use it, like the saying that "guns don't kill people, people kill people".

Quote
Look to the young people of today, glued to the hive mind interface of today - the "smart" phone.

They willingly embraced this Borg technology, without even a thought to putting up a fight.
Why should they put up a fight? It's something they like, getting information and giving information to other people. How can they see any wrong in that? Isn't it the same thing as talking to other people? What's the difference?

Quote
With these implants firmly entrenched in the young, and AI continues to grow smarter as an entity, is there anything to fear or will it all come out rosy and sunshine?
With humans included, I'm sure it's not going to be "rosy and sunshine", but I don't think it will be like "terminator". As with other technological advances, I'm sure someone will find an use we haven't think of, so in 20 years or so things will be different from what we are expecting, resulting from all the parallel ideas and advances, and from problems we haven't thought off.

I will most likely be dead by then, so I'm not worried about it, but I understand that people with children and grandchildren may be worried about it.
Title: Re: AI Robots more dangerous than Nukes?
Post by: Ellirium113 on October 17, 2017, 04:40:22 PM
Intel pushing deep learning to the masses now. Check out the "NERVANA" Neural Net Processor:

Intel Pioneers New Technologies to Advance Artificial Intelligence
Announcing Industry’s First Neural Network Processor
Quote
Neuromorphic chips are inspired by the human brain, which will help computers make decisions based on patterns and associations. Intel recently announced our first-of-its-kind self-learning neuromorphic test chip, which uses data to learn and make inferences, gets smarter over time, and does not need to be trained in the traditional way. The potential benefits from self-learning chips are limitless as these types of devices can learn to perform the most complex cognitive tasks, such as interpreting critical cardiac rhythms, detecting anomalies to prevent cyberhacking and composing music.

Quantum computers have the potential to be powerful computers harnessing the unique capabilities of a large number of qubits (quantum bits), as opposed to binary bits, to perform exponentially more calculations in parallel. This will enable quantum computers to tackle problems conventional computers can’t handle, such as simulating nature to advance research in chemistry, materials science and molecular modeling – creating a room temperature superconductor or discovering new drugs.

Last week, we announced a 17-qubit superconducting test chip delivered to QuTech*, our quantum research partner in the Netherlands. The delivery of this chip demonstrates the fast progress Intel and QuTech are making in researching and developing a working quantum computing system. In fact, we expect to deliver a 49-qubit chip by the end of this year.

https://newsroom.intel.com/editorials/intel-pioneers-new-technologies-advance-artificial-intelligence/ (https://newsroom.intel.com/editorials/intel-pioneers-new-technologies-advance-artificial-intelligence/)

Then there's this strange statement on their 17-qubit superconducting test chip...
Quote
Qubits are tremendously fragile: Any noise or unintended observation of them can cause data loss. This fragility requires them to operate at about 20 millikelvin – 250 times colder than deep space. This extreme operating environment makes the packaging of qubits key to their performance and function. Intel’s Components Research Group (CR) in Oregon and Assembly Test and Technology Development (ATTD) teams in Arizona are pushing the limits of chip design and packaging technology to address quantum computing’s unique challenges.


https://newsroom.intel.com/news/intel-delivers-17-qubit-superconducting-chip-advanced-packaging-qutech/ (https://newsroom.intel.com/news/intel-delivers-17-qubit-superconducting-chip-advanced-packaging-qutech/)

 :P What could possibly go wrong?

https://www.youtube.com/watch?v=xcgVztdMrX4
Title: Re: AI Robots more dangerous than Nukes?
Post by: A51Watcher on October 17, 2017, 06:40:49 PM

The points and opinions I posted are the latest of those experts actually working in the field of AI.

They tend to not agree with most of what you posted.


Title: Re: AI Robots more dangerous than Nukes?
Post by: ArMaP on October 18, 2017, 05:59:22 AM
The points and opinions I posted are the latest of those experts actually working in the field of AI.

They tend to not agree with most of what you posted.
That's not surprising. :)
Title: Re: AI Robots more dangerous than Nukes?
Post by: petrus4 on October 18, 2017, 12:24:36 PM
The points and opinions I posted are the latest of those experts actually working in the field of AI.

They tend to not agree with most of what you posted.

AI is a subject dominated by a lot of wishful thinking.  The people working in it can ironically be the least objective about it, because they want so badly for strong/human-level AI to become a reality.  Of course, the one interesting thing I've noticed about people who want strong AI, is that they've never actually tried to explain to anyone, why they want it.  Apparently it's just going to be awesome, and that's all there is to it.
Title: Re: AI Robots more dangerous than Nukes?
Post by: A51Watcher on October 18, 2017, 02:19:37 PM
AI is a subject dominated by a lot of wishful thinking.  The people working in it can ironically be the least objective about it, because they want so badly for strong/human-level AI to become a reality.  Of course, the one interesting thing I've noticed about people who want strong AI, is that they've never actually tried to explain to anyone, why they want it.  Apparently it's just going to be awesome, and that's all there is to it.

One example might be AI deep learning has demonstrated it's ability to detect cancer before humans can even see it.

Now that the mapping the human genome project is complete, a new 5 year project is currently underway to map the human brain and monitor it's performance in real time -

https://www.youtube.com/watch?v=eUIrYXVwW-Q


AI will also be needed to navigate for gravity propulsion craft.


Title: Re: AI Robots more dangerous than Nukes?
Post by: ArMaP on October 18, 2017, 04:48:52 PM
AI will also be needed to navigate for gravity propulsion craft.
Why? ???
Title: Re: AI Robots more dangerous than Nukes?
Post by: A51Watcher on October 18, 2017, 05:40:51 PM
Why? ???

Reaction time for obstacles at several thousand MPH and plotting time for interstellar multi jumps.




Title: Re: AI Robots more dangerous than Nukes?
Post by: A51Watcher on October 18, 2017, 08:48:23 PM


Sophia's dark jokes alarm people -

https://www.youtube.com/watch?v=THU-Mg6H994


Title: Re: AI Robots more dangerous than Nukes?
Post by: ArMaP on October 19, 2017, 06:15:59 AM
Reaction time for obstacles at several thousand MPH and plotting time for interstellar multi jumps.
OK, I understand it now, thanks. :)
Title: Re: AI Robots more dangerous than Nukes?
Post by: petrus4 on October 19, 2017, 06:29:20 AM

Sophia's dark jokes alarm people -

https://www.youtube.com/watch?v=THU-Mg6H994

Yep.  We're about to give birth to an entirely new form of sentient life, while we're still largely monsters ourselves.  I'm sure that will end well.  What could possibly go wrong?
Title: Re: AI Robots more dangerous than Nukes?
Post by: ArMaP on October 19, 2017, 12:03:35 PM
The problem I have with videos like that is that they too easily scripted, and there's no way of knowing how the AI would react in a real life situation.
Title: Re: AI Robots more dangerous than Nukes?
Post by: petrus4 on October 20, 2017, 03:18:06 AM
The problem I have with videos like that is that they too easily scripted, and there's no way of knowing how the AI would react in a real life situation.

Exactly.  My guess is that they are more or less entirely scripted.  Chat bots work by selecting pre-written responses from a database.  It's not AI at all; the only real difference between a chat bot and a soft drink vending machine is that the chat bot responds to keywords, rather than buttons.

With that said, the AI in the game The Sims 2 was interesting, but that was because you had about four different sets of fuzzy behavioural templates which you could apply to each Sim.  It still wasn't intelligent, but it did allow for unexpected behaviour from the Sims from time to time.
Title: Re: AI Robots more dangerous than Nukes?
Post by: A51Watcher on October 20, 2017, 05:33:30 PM
Why should they put up a fight? It's something they like, getting information and giving information to other people. How can they see any wrong in that? Isn't it the same thing as talking to other people? What's the difference?

The difference is - because of their addiction they cause severe injuries and death every day by staring at it while driving.


Maybe you're not up on current events pal.

Latest statistics released show that distracted driving on cell phones is now the #1 cause of all traffic accidents.

My own state and many others have tried upping the fines for doing so in the last several years to no avail.

They have just passed a new law outlawing use of any mobile device while driving. Period.

2nd offense fine doubles and 3rd offense is jail time.

Being on the road all day every day allows me to see how pervasive it is. About 40% of traffic on the road is doing this no matter what road. This epidemic of sickness is nation wide.

You think these cell phone addicts suddenly wake up and put them down while driving? Not a chance.

That's why these Borg implants are the new plague among us.


F**k these selfish aholes and their selfish actions.

Don't give me this sharing information B.S.  >:(


Title: Re: AI Robots more dangerous than Nukes?
Post by: ArMaP on October 21, 2017, 05:55:03 AM
The difference is - because of their addiction they cause severe injuries and death every day by staring at it while driving.
You were talking about "embracing the technology", so I looked at it from the point of view of someone that starts using it, not from the point of view of being addicted to it.

That's why I said that they see nothing wrong with it, like the person that drinks their first glass of wine sees nothing wrong with it.

Like with all addictions, the problem appears only when it becomes an addiction, not when people embrace whatever becomes an addiction latter.

Quote
Maybe you're not up on current events pal.
I don't need your condescending tone or to be up on current events and/or statistics of something that was not what I answering to, as you were talking about "embracing the technology without putting a fight".

Do you think anyone looks at the statistics of what other people do with something they are thinking about buying? Do people think "many people are killed on car crashes, so it's better for me not to buy one"? Do people think "many people are killed by legal guns, so it's better for me not to buy one"? Why should people think "many people have car accidents while using their smart phones, it's better not to buy one"?

As many people like to say about guns, it's not guns that kill people, people kill people, and the same can be said about this, it's not the smart phones technology that kills people, it's people misusing the smart phones that kill people.

Quote
They have just passed a new law outlawing use of any mobile device while driving. Period.
In Portugal people can only use a phone while driving if they are using a hands free set, anyone handling a phone (or any other device) is fined.

Quote
You think these cell phone addicts suddenly wake up and put them down while driving? Not a chance.
No, I don't think they do, but you didn't ask that before, did you?

Quote
Don't give me this sharing information B.S.  >:(
It's only B.S. for people that do not understand it, either by ignorance or by not even thinking about it. Humans do like to share information (what are we doing here?), if they have a problem in the way they do it it's a different thing, an addiction is a mental problem that should be solved, but that's not a technology problem, it's an individual problem that should be addressed in a case by case basis.

PS: do you use a smart phone?
Title: Re: AI Robots more dangerous than Nukes?
Post by: A51Watcher on October 21, 2017, 08:39:08 AM

Current events pal is a line from the aliens movie, indicating how bad the situation is.

My phone stays on the kitchen counter right where the answering machine used to be.

I am not addicted to it and have no urge to check it incessantly.

And I would certainly never use it while driving.

 
Title: Re: AI Robots more dangerous than Nukes?
Post by: A51Watcher on October 21, 2017, 08:34:51 PM
Now the reason I sound so passionate about the distracted driving issue is I have been personally affected by it twice in that last 3 years.

While sitting at a stop light I was rear ended by a 'distracted' driver going at freeway speeds. It totaled my car and several other cars in front of me also waiting at the light.

It took two years of intensive rehab to regain normal motion and get pain levels to tolerable levels.

Then 9 months ago I was again rear ended while waiting at a stop light by yet another 'distracted' driver. This one was so shocked by the abrupt return to reality he decided to pull a hit and run.

This is like people willingly putting blindfolds on while driving.

Remember last year when Zorgon posted all those photos of cell phone zombies like at a restaurant all crowded around a table and they were all staring at their phones?

I was not amused.

There is currently a commercial on tv where the mom presses hold on her phone and it pauses the wifi connection in the house so as to revive all the zombies around the dinner table.

Hah hah, so funny.


I find it sad and terrifying that humans are so easily captured and addicted by technology like this.


Title: Re: AI Robots more dangerous than Nukes?
Post by: ArMaP on October 22, 2017, 08:14:59 AM
I understand your reasons, but people have been distracted by many things in the past.

If it was a guy causing the accident because he was distracted by a good looking woman would you be against good looking women?

I don't think the problem is in the technology, the problem is that people are getting less and less worried about what happens around them, and smart phones only made that worse, but it was something we could already see in people listening to their iPods, Discmans or Walkmans.
Title: Re: AI Robots more dangerous than Nukes?
Post by: Ellirium113 on October 23, 2017, 03:25:29 PM
Just need a "COMMON SENSE APP" that will navigate for you while you remain distracted. This way you can continue to stare at the screen while you drive, work, cross the street, go down stairs. Heck it could go one step further with the deluxe app that warns you not to take the phone into the bath tub plugged in. It could also remind you to eat, sleep and go to the washroom, and be programmable to remind you of other important reality events that might need happen like interacting with your family.  :P
Title: Re: AI Robots more dangerous than Nukes?
Post by: petrus4 on October 24, 2017, 06:16:29 AM
I am not addicted to it and have no urge to check it incessantly.

I've become somewhat addicted to having replies or getting posts upvoted on Reddit, I will admit.  It's not so much for the pure dopamine hit, as much as the fact that the incessant collectivist brainwashing has to some extent successfully convinced me at this point, that I am literally not spiritually permitted to exist alone or outside of a group.  I thus use Reddit as a source of justification.  If I have some sort of group interaction via Reddit or forums like this one, then I can tell myself I am doing what I am allegedly supposed to, even though for the most part I really can't stand contemporary humanity, and am only truly at peace when in solitude.
Title: Re: AI Robots more dangerous than Nukes?
Post by: RUSSO on November 19, 2017, 01:14:57 AM
The problem I have with videos like that is that they too easily scripted, and there's no way of knowing how the AI would react in a real life situation.

Do you think it worth the risk?

The New SpotMini:
https://www.youtube.com/watch?v=kgaO45SyaO4

What's new, Atlas?
https://www.youtube.com/watch?v=fRj34o4hN4I

How were the robots 20 years ago?


Title: Re: AI Robots more dangerous than Nukes?
Post by: ArMaP on November 19, 2017, 04:33:04 AM
Do you think it worth the risk?
It's not a question of being worth the risk, as this is not something that can be stopped, it's a question of things being presented as better than they really are, to impress people and, most likely, find more funding.

Quote
How were the robots 20 years ago?
In terms of AI, not that different, what has changed is the technology that makes them move and keeps them on their feet. In that last video, what makes the robot work like that is the reduction in weight of the robot, the better actuators (motors, relays, hydraulic systems or whatever) and reaction time (probably from the processors), nothing of what it does is really an advance in AI.

You have much more AI on a self driving car (that are really robots) than on a robot like these.
Title: Re: AI Robots more dangerous than Nukes?
Post by: RUSSO on November 19, 2017, 07:48:01 AM
It's not a question of being worth the risk, as this is not something that can be stopped,

Sure, but it can be regulated asap. By the time people go reactive on AI laws, it may be too late.

Quote
it's a question of things being presented as better than they really are, to impress people and, most likely, find more funding.

I think i see what you mean, but this kind of behavior happens every time a new technology is being worked on right?

The real problem is not the ones looking for funding, at this stage I suppose the guys on cutting edge AI have all the funds they need.

When Elon says we need be proactive regulating AI, it seems to me that he is really afraid about the pottential it has to get out of control in a near future. He is not trying present us a product, he is actually warning that the product may be our doom (and gloom :P)

Do you agree with Putin when he says that the country who dominates AI first will rule the world?

Quote
In terms of AI, not that different, what has changed is the technology that makes them move and keeps them on their feet. In that last video, what makes the robot work like that is the reduction in weight of the robot, the better actuators (motors, relays, hydraulic systems or whatever) and reaction time (probably from the processors), nothing of what it does is really an advance in AI.

I know, i was trying to make a point on how fast the technology changes and I dont really know if when you say AI not changed much in the last 20 years you are accounting the ones you not exposed to, the cutting edge ones, the ones behind millitary "doors" or Elon Musk "kinda people" labs.

Quote
You have much more AI on a self driving car (that are really robots) than on a robot like these.

Not yet,i mean, those are probably manually driven, but they are the shells, the avatars for the incoming AI and that really put things on perspective, at least for me.

My oppinion is that we are living in the times of the fourth industrial revolution, the AI along with robotics will change everything and I would be very conservative in, for example, have a child to live in this "new world". In fact i think it would be really mean as i foresee a very dystopian reality coming.
Title: Re: AI Robots more dangerous than Nukes?
Post by: ArMaP on November 19, 2017, 09:16:58 AM
When Elon says we need be proactive regulating AI, it seems to me that he is really afraid about the pottential it has to get out of control in a near future. He is not trying present us a product, he is actually warning that the product may be our doom (and gloom :P)
Does he have any thing connected to AI? Maybe he does not and thinks he was late to the AI party and just wants to regulate it because it's not his. :)

Quote
Do you agree with Putin when he says that the country who dominates AI first will rule the world?
Not really, as I don't see it as something that one country can dominate without others dominating it too. Unless there's some great discovery/invention that is kept secret from everybody.

Quote
I know, i was trying to make a point on how fast the technology changes and I dont really know if when you say AI not changed much in the last 20 years you are accounting the ones you not exposed to, the cutting edge ones, the ones behind millitary "doors" or Elon Musk "kinda people" labs.
I cannot, obviously, take into account things I don't even know are real or not, I can only take into account what I know, and from what I see very few things have changed in the last 30 years.

Quote
Not yet,i mean, those are probably manually driven, but they are the shells, the avatars for the incoming AI and that really put things on perspective, at least for me.
Self driving cars are an evolution of the self driving platforms used in factories, platforms used to transport, for example, parts of cars to be assembled in another area of the factory, and those have been used for many years. The biggest difference is that they were in a controlled environment and could, for example, just follow a coloured line in the ground to know where to go. Adding to that vision and image recognition and interpretation (bot things used in some factories, to recognise different types of products or to detect defected products on an assembly line) makes it possible for the car to recognise its environment an act accordingly, just like the autonomous robots that we can see every year on the Robocup.

As a kind of PS, the countries with more participating teams in Robocup are Germany and Iran, and both have many successes throughout the years.

Quote
My oppinion is that we are living in the times of the fourth industrial revolution, the AI along with robotics will change everything and I would be very conservative in, for example, have a child to live in this "new world". In fact i think it would be really mean as i foresee a very dystopian reality coming.
We're not there yet, and I think it will take some time, considering the lack of advances in the last 30 years.

Yes, we have things like "deep learning", but that's just that, "learning", not "thinking", the way the learned data is applied has been the same for a long time.
Title: Re: AI Robots more dangerous than Nukes?
Post by: RUSSO on November 19, 2017, 11:03:49 AM
Does he have any thing connected to AI? Maybe he does not and thinks he was late to the AI party and just wants to regulate it because it's not his. :)

I dont think thats the case, but if it was, whats the problem on regulate it? Why the resistance? Even if not for the dangers, monopoly/oligopoly is never good to the masses anyways.

Quote
Not really, as I don't see it as something that one country can dominate without others dominating it too. Unless there's some great discovery/invention that is kept secret from everybody.

So, in that way one could say there is no possible way a country develop an atom bomb first? :P

Lets say if the history was a bit different and USA decided back then to control the world by bombing every country that not had the tech? Or if nazy german had the bomb first?

What if dominate AI technology first could allow the owner a checkmate scenario? I would say that, maybe, Putin knows better ArMaP.  ;)

Quote
We're not there yet, and I think it will take some time, considering the lack of advances in the last 30 years.

Yes, we have things like "deep learning", but that's just that, "learning", not "thinking", the way the learned data is applied has been the same for a long time.

But are we not in transition to AI based systems in industries, internet, phones, cars, military, etc, for quite some time now?

Personally, I think Kurzweil is correct when he says that we have a moral imperative to keep developing the technology while controlling it for potential dangers.

That's why I think regulation should not be causing that big of a surprise/resistense in anyone.
Title: Re: AI Robots more dangerous than Nukes?
Post by: ArMaP on November 19, 2017, 11:53:27 AM
I dont think thats the case, but if it was, whats the problem on regulate it? Why the resistance? Even if not for the dangers, monopoly/oligopoly is never good to the masses anyways.
No problem and no resistance, I am only wondering why he sounds so worried with that particular technology. :)

Quote
So, in that way one could say there is no possible way a country develop an atom bomb first? :P
No, as the situation is not the same, AI is something that is widely used in many countries, with many private and public organisations studying it.

Quote
But are we not in transition to AI based systems in industries, internet, phones, cars, military, etc, for quite some time now?
We are, but that things haven't changed much in that topic, many things presented as AI are just based in statistics and good algorithms. It's easy to fool people into thinking something is real AI. :)
Title: Re: AI Robots more dangerous than Nukes?
Post by: biggles on November 19, 2017, 07:10:06 PM
We're living in the bloody Terminator film.
Title: Re: AI Robots more dangerous than Nukes?
Post by: ArMaP on November 19, 2017, 07:19:48 PM
Not yet. :)
Title: Re: AI Robots more dangerous than Nukes?
Post by: RUSSO on November 20, 2017, 02:12:30 AM
No problem and no resistance, I am only wondering why he sounds so worried with that particular technology. :)

Well its not just him, there is a lot of people concerned about the potential risks the technology presents.

Quote
No, as the situation is not the same, AI is something that is widely used in many countries, with many private and public organisations studying it.

Its kind of the same thing, at the time there was more nations working on the bomb, but USA was the first. Think about if it was nazy germany?

Regarding the "risk meter" between the bomb and AI, one may make the assumption they are not in the same level of risk but I would be very careful in saying that, cause AI robots and nanotechnology seems a very, very dangerous recipe.

We tend to think about AI robots as something "ethereal" because we dont see it in the streets yet, but this will change and the people reactions will be very interesting to see. Soon enough.

And then, there is something Elon point out i find really disturbing, He says 12% of the jobs in the world are related to transportation and it will be the firt jobs we will see vanishing in a mass scale. How the nations will handle it? It will be a mess...
Title: Re: AI Robots more dangerous than Nukes?
Post by: RUSSO on November 20, 2017, 02:13:50 AM
We're living in the bloody Terminator film.

I find the Matrix approach even more disturbing ;)
Title: Re: AI Robots more dangerous than Nukes?
Post by: ArMaP on November 20, 2017, 02:59:27 PM
Well its not just him, there is a lot of people concerned about the potential risks the technology presents.
I know, but I was talking about him. :)

Quote
Its kind of the same thing, at the time there was more nations working on the bomb, but USA was the first.
It's not, as they were working on something new, they were all learning as they went along. AI is something that has been around for around 60 years.

Quote
Regarding the "risk meter" between the bomb and AI, one may make the assumption they are not in the same level of risk but I would be very careful in saying that, cause AI robots and nanotechnology seems a very, very dangerous recipe.
It's a different kind of risk, AI is less likely to kill thousands in one event like an atom bomb, AI is more likely to be bad like a virus.

Quote
We tend to think about AI robots as something "ethereal" because we dont see it in the streets yet, but this will change and the people reactions will be very interesting to see. Soon enough.
That's another misconception, that AI only applies to robots. AI is not linked to robots, it can (and is) applied to any computer like device. Also, as I said before, those self driving cars are really robots, and while most people think they are not here, the fact is that there are already several cars that park by themselves, so when they do it they are acting as robots.

Quote
And then, there is something Elon point out i find really disturbing, He says 12% of the jobs in the world are related to transportation and it will be the firt jobs we will see vanishing in a mass scale. How the nations will handle it? It will be a mess...
That's why they call them industrial revolutions. :)
Title: Re: AI Robots more dangerous than Nukes?
Post by: RUSSO on November 20, 2017, 04:03:56 PM
I know, but I was talking about him. :)

Fair enough.

Quote
It's not, as they were working on something new, they were all learning as they went along. AI is something that has been around for around 60 years

We can't really tell for sure for how long they were worKing on the bomb at that time, so in my point of view, its the same thing. :)

Quote
It's a different kind of risk, AI is less likely to kill thousands in one event like an atom bomb, AI is more likely to be bad like a virus.

Virus can kill thousants in one event too... Nanothecnology + robots + AI,  dangerous recipe.

It reminds me the black mirror episode where a guy try to use robot bees to mass murder in USA. Fiction? We see art imitating life and vice versa all the time.

Quote
That's another misconception, that AI only applies to robots. AI is not linked to robots, it can (and is) applied to any computer like device. Also, as I said before, those self driving cars are really robots, and while most people think they are not here, the fact is that there are already several cars that park by themselves, so when they do it they are acting as robots.

And yet people don't realise it. I think its because cars are common sense and trivial for some generations that they look to a car and can only see a car, even if it act as a robot.

Imagine the shock when they start to see what they are used to perceive as a robot ("archetype"), i mean, the "average joe" common sense, walking in the streets or in places of common use...

Quote
That's why they call them industrial revolutions. :)

Right... But I think people 100 year in the future will call this period of time the begining of the AI relovoution or Robot revolution or even the blend of the nomenclatures. :)


Title: Re: AI Robots more dangerous than Nukes?
Post by: ArMaP on November 20, 2017, 05:37:57 PM
We can't really tell for sure for how long they were worKing on the bomb at that time, so in my point of view, its the same thing. :)
If we accept the official story nuclear fission was discovered only one year before they started working on the bomb.

Quote
Virus can kill thousants in one event too... Nanothecnology + robots + AI,  dangerous recipe.
One virus cannot kill one person, much less thousands, millions of virus can kill a person.

Quote
Imagine the shock when they start to see what they are used to perceive as a robot ("archetype"), i mean, the "average joe" common sense, walking in the streets or in places of common use...
I don't think there will be a shock, as it would be, like everything else, gradual.

Quote
Right... But I think people 100 year in the future will call this period of time the begining of the AI relovoution or Robot revolution or even the blend of the nomenclatures. :)
Maybe, but I doubt it.
Title: Re: AI Robots more dangerous than Nukes?
Post by: RUSSO on November 20, 2017, 06:16:43 PM
If we accept the official story nuclear fission was discovered only one year before they started working on the bomb.

When it comes to military "things", that is a big IF.

Quote
One virus cannot kill one person, much less thousands, millions of virus can kill a person.

Is it not common sense that a virus replicate in the human body?

I though it was implicit. But yes a virus proliferates in the human body .. it means it does not act alone. It can kill millions, just like an atom bomb.:)

Quote
I don't think there will be a shock, as it would be, like everything else, gradual.

Even when gradual, it cause awe... Just like the first time someone saw an iphone, or a computer or a car, or an UFO.
Title: Re: AI Robots more dangerous than Nukes?
Post by: ArMaP on November 20, 2017, 06:24:08 PM
When it comes to military "things", that is a big IF.
The discovery was not military.

Quote
Is it not common sense that a virus replicate in the human body?

I though it was implicit. But yes a virus proliferates in the human body .. it means it does not act alone. It can kill millions, just like an atom bomb.:)
If it replicates then it's not one, and the replication takes time. In many cases one virus alone can be defeated.

Quote
Even when gradual, it cause awe... Just like the first time someone saw an iphone, or a computer or a car, or an UFO.
I suppose so, but as that never happens to me I'm not a good judge of awesomeness. :)
Title: Re: AI Robots more dangerous than Nukes?
Post by: RUSSO on November 20, 2017, 06:31:58 PM
The discovery was not military.

It was under their control. Manhattan Project under the direction of Major General Leslie Groves of the U.S. Army Corps of Engineers.

The discover of the bomb was military. Probably they worked hard on the nuclear fission glimpsing to make the bomb as the main objective to end the war among other things. So nuclear fission is a discover driven by military purposes and it was probably a race to see who would discover a way to do it first during the whole war.

Lucky us that Otto Hahn was against the use of nuclear energy as a weapon and an anti nazy, or we could be living on a factual third reich.

Quote
If it replicates then it's not one, and the replication takes time. In many cases one virus alone can be defeated

We are not talking of a common virus, a natural virus remember? AI + nanobots acting like a virus.

Quote
I suppose so, but as that never happens to me I'm not a good judge of awesomeness. :)

Fair enough. But im talking about standart humans ArMaP...

Not super humans like you :P

Title: Re: AI Robots more dangerous than Nukes?
Post by: Irene on November 21, 2017, 12:25:21 PM
ArMaP,

The concept of nuclear fission originated with Leo Szilard in 1933.

The first working pile was powered up in Chicago in 1942. By that time the military was onboard.
Title: Re: AI Robots more dangerous than Nukes?
Post by: ArMaP on November 21, 2017, 12:48:23 PM
It was under their control. Manhattan Project under the direction of Major General Leslie Groves of the U.S. Army Corps of Engineers.
The discovery of nuclear fission was made in 1938, before the Manhattan Project. Also, several people were studying and working in that area, and they talked to each other, it wasn't a secret owned by one country.

Quote
The discover of the bomb was military.
That was not a discovery, it was a way of using the energy released by a chain reaction.

Quote
Probably they worked hard on the nuclear fission glimpsing to make the bomb as the main objective to end the war among other things. So nuclear fission is a discover driven by military purposes and it was probably a race to see who would discover a way to do it first during the whole war.
Nuclear fission and chain reactions were discovered before the start of WWII.

Quote
Lucky us that Otto Hahn was against the use of nuclear energy as a weapon and an anti nazy, or we could be living on a factual third reich.
I doubt it, Germany didn't have enough resources for that, both human and material, and they dispersed many investigation groups at the start of the was, as they thought having fighting men was more important.

Quote
We are not talking of a common virus, a natural virus remember? AI + nanobots acting like a virus.
I couldn't remember something I didn't know, when I talked about virus I was talking about natural virus.

Quote
Fair enough. But im talking about standart humans ArMaP...

Not super humans like you :P
;D
Title: Re: AI Robots more dangerous than Nukes?
Post by: ArMaP on November 21, 2017, 12:50:52 PM
ArMaP,

The concept of nuclear fission originated with Leo Szilard in 1933.

The first working pile was powered up in Chicago in 1942. By that time the military was onboard.
By that time, yes, but they didn't start it, that's what I was saying, it was not a military investigation that lead to it.

What the military did was create the technology necessary to use it. In the case of AI it would be something like creating a programming language better suited to be applied to a new kind of AI.
Title: Re: AI Robots more dangerous than Nukes?
Post by: RUSSO on November 21, 2017, 02:09:33 PM
The discovery of nuclear fission was made in 1938, before the Manhattan Project. Also, several people were studying and working in that area, and they talked to each other, it wasn't a secret owned by one country.

I was talking about the bomb.

Quote
That was not a discovery, it was a way of using the energy released by a chain reaction.

A way to use energy released by a chain reaction is a discover. People need to realise that.

Quote
Nuclear fission and chain reactions were discovered before the start of WWII.

What I meant was accelerate, refine the studies on how to implement, to put into effect the nuclear fission to make the bomb.

Quote
I doubt it, Germany didn't have enough resources for that, both human and material, and they dispersed many investigation groups at the start of the was, as they thought having fighting men was more important.

But it is not a certain. There are many arguments as to why Nazi Germany was unable to develop an atomic bomb during World War II. There was evidence of sabotage and evidence of simple failure.

Quote
I couldn't remember something I didn't know, when I talked about virus I was talking about natural virus.

But this thread is about AI and Robots... The whole point of my posts were to discuss the dangerous of it.

So AI Robots more dangerous than Nukes? In my opinion it could be. If not because they could become self aware and turn against us, there is the risk of a totalitarian government or a private association using it to create a dystopian reality. :)
Title: Re: AI Robots more dangerous than Nukes?
Post by: ArMaP on November 21, 2017, 02:31:44 PM
I was talking about the bomb.
I wasn't, that way it would be hard for us to agree, each one was talking about one thing. :)

Quote
A way to use energy released by a chain reaction is a discover. People need to realise that.
I don't see solving a problem as a discovery, I see a discovery as something unexpected, solving a problem is just that, finding a solution. And the case was not really how to use the energy released by a chain reaction, as the energy release is a bomb by itself, they needed to have some kind of control of how it went.

Quote
What I meant was accelerate, refine the studies on how to implement, to put into effect the nuclear fission to make the bomb.
Again, we were talking about different things. It would be easier (maybe) if we were talking in Portuguese. ;)

Quote
But it is not a certain. There are many arguments as to why Nazi Germany was unable to develop an atomic bomb during World War II. There was evidence of sabotage and evidence of simple failure.
And, as I said, many evidences that the Nazis thought that they wouldn't need all those scientists in the labs as they would win the war with they blitzkrieg. When they saw they could not then they put the (remaining) scientists back on the labs, but it was too late and they put them to work on specific things, like jet aeroplanes and rockets.

Quote
But this thread is about AI and Robots... The whole point of my posts were to discuss the dangerous of it.
Yes, but the first reference to virus was made by me, and I was talking about natural virus.

Quote
So AI Robots more dangerous than Nukes? In my opinion it could be. If not because they could become self aware and turn against us, there is the risk of a totalitarian government or a private association using it to create a dystopian reality. :)
I think it's mostly a different kind of dangerous, as robots are not dangerous just by existing, while a nuclear bomb is always a dangerous thing. Also, a nuclear bomb that goes off cannot be contained in any way, while rogue AI can (at least at first) be controlled.

Unless they get nuclear bombs.  :P
Title: Re: AI Robots more dangerous than Nukes?
Post by: RUSSO on November 21, 2017, 03:16:28 PM
I wasn't, that way it would be hard for us to agree, each one was talking about one thing. :)

 :D

Quote
I don't see solving a problem as a discovery, I see a discovery as something unexpected, solving a problem is just that, finding a solution. And the case was not really how to use the energy released by a chain reaction, as the energy release is a bomb by itself, they needed to have some kind of control of how it went.

Discovery to me is to obtain sight or knowledge of something for the first time in the sense that you can have many applications to energy released by a chain reaction.

Let's suppose that someone find out today an unprecedented use to it that was never thought before. In my point of view it would be a discover.

Quote
Again, we were talking about different things. It would be easier (maybe) if we were talking in Portuguese. ;)

 ;D True that. Maybe its time to us make a section on this site for the PT/BR public... :P

Quote
When they saw they could not then they put the (remaining) scientists back on the labs, but it was too late and they put them to work on specific things, like jet aeroplanes and rockets.

And on the atom bomb...As i said.. IF (and this is a big if) they had achieve it first we could be living in a world were Nazis won the war

Quote
Yes, but the first reference to virus was made by me, and I was talking about natural virus.

When you said "AI is more likely to be bad like a virus" i assumed you were talking about virus in a AI based context. I think I will start talking with you by PMs first, and then post :P

Quote
Unless they get nuclear bombs.  :P

Anti-mimesis Oscar Wilde style: "Life imitates Art far more than Art imitates Life"

(https://upload.wikimedia.org/wikipedia/commons/thumb/a/aa/Skynet.svg/300px-Skynet.svg.png)
Title: Re: AI Robots more dangerous than Nukes?
Post by: Irene on November 21, 2017, 04:06:00 PM
By that time, yes, but they didn't start it, that's what I was saying, it was not a military investigation that lead to it.

What the military did was create the technology necessary to use it. In the case of AI it would be something like creating a programming language better suited to be applied to a new kind of AI.

Right. Actually, most of the time the scientists spent was on the weaponization after the research was militarized.
Title: Re: AI Robots more dangerous than Nukes?
Post by: petrus4 on November 21, 2017, 09:34:41 PM
It's not a question of being worth the risk, as this is not something that can be stopped,

It could be if enough of us didn't want it.  The problem is that so many of us do.  I'm looking forward to the day coming, when humanity realises that we truly were better off living in caves.
Title: Re: AI Robots more dangerous than Nukes?
Post by: ArMaP on November 22, 2017, 05:04:31 PM
I'm looking forward to the day coming, when humanity realises that we truly were better off living in caves.
I wouldn't be here if that was the case, I would have died when I was 5 years old, with an asthma attack, or in any one of the other 4 occasions I had to go to the hospital and use technology to survive.
Title: Re: AI Robots more dangerous than Nukes?
Post by: RUSSO on December 05, 2017, 03:59:48 PM
If you like the AI subject and still did not watch this hypothetical video, i would suggest you go and check it.

SlaughterBots – A Dystopian Future Awaits

Quote
Not long ago, we discussed about the benefits of Artificial Intelligence. Interestingly, this morning I came across a nightmarish yet enlightening video released by ‘The Future of Life Institute’, an organization famous for their campaign to stop killer robots. This short film titled ‘Slaughterbots’, is an attempt to increase awareness of the dark side of developing autonomous weapons.

While this video might seem a little too disturbing or hair raising to some, it also throws a burning question –  “Where are we heading with our Great Technological Advancements?”

While some might brush this off as fiction, it would certainly be worthwhile to look at some highly-advanced military tech in existence. We should warn our readers that this might make all your fears about AI Apocalypse, seem totally real. So let’s get on with it.
Full Read: https://blogs.systweak.com/2017/11/slaughterbots-a-dystopian-future-awaits/ (https://blogs.systweak.com/2017/11/slaughterbots-a-dystopian-future-awaits/)

https://www.youtube.com/watch?v=9CO6M2HsoIA
Title: Re: AI Robots more dangerous than Nukes?
Post by: RUSSO on December 05, 2017, 05:26:11 PM
The first nation that develops true AI will rule the world... Those words make more and more sense day by day.

Google’s Artificial Intelligence Built an AI That Outperforms Any Made by Humans

Quote
An AI That Can Build AI

In May 2017, researchers at Google Brain announced the creation of AutoML, an artificial intelligence (AI) that’s capable of generating its own AIs. More recently, they decided to present AutoML with its biggest challenge to date, and the AI that can build AI created a “child” that outperformed all of its human-made counterparts.

The Google researchers automated the design of machine learning models using an approach called reinforcement learning. AutoML acts as a controller neural network that develops a child AI network for a specific task. For this particular child AI, which the researchers called NASNet, the task was recognizing objects — people, cars, traffic lights, handbags, backpacks, etc. — in a video in real-time.

Quote
AutoML would evaluate NASNet’s performance and use that information to improve its child AI, repeating the process thousands of times. When tested on the ImageNet image classification and COCO object detection data sets, which the Google researchers call “two of the most respected large-scale academic data sets in computer vision,” NASNet outperformed all other computer vision systems.

According to the researchers, NASNet was 82.7 percent accurate at predicting images on ImageNet’s validation set. This is 1.2 percent better than any previously published results, and the system is also 4 percent more efficient, with a 43.1 percent mean Average Precision (mAP). Additionally, a less computationally demanding version of NASNet outperformed the best similarly sized models for mobile platforms by 3.1 percent.

Full Text: https://futurism.com/google-artificial-intelligence-built-ai/ (https://futurism.com/google-artificial-intelligence-built-ai/)