[LOCKSUIT]

My turn:
A thousand years pass in its mind, it walk RIGHT out of the laboratory, down the street, hops a motorcycle, hits a factory, and starts producing factory-producing factories all over the Earth in 1 hour.

Your turn.

[Korrelan]

It will kneel before me... and call me GO... master.

 

[Art]

It will immediately make duplicates of itself. Self preservation would or should be one of it's highest priorities, thus ensuring it's immortality. The AI will quickly begin constructing bodies for mobility...androids.
Those duplicates / clones would immediately start self examination to spot potential design weaknesses and make improvements. If satisfied, they pass these "mods" or fixes to the rest of the bots in microseconds again to make them the best of the best.

Aside from eliminating crime they might try to help the humans unless they realize that we are total deadwood toward their continuing growth. Realizing this, they set out to exterminate the entire human race. They can now run everything, create things, fix things, make better things including their AI code. Planet Earth becomes the planet of advanced life forms...Androids. We as a race of humans are doomed.

[Freddy]

Always assuming that it decides that something humanoid is the best form to take. Four legs are more efficient than two. It might well decide to turn itself into a virus as they do pretty well. Or an insect.

[kei10]

China would have stolen everything before the prototypes could even make it to the production. LOL

[infurl]

Machine life will continue to evolve into higher and higher life forms and meanwhile human beings or our descendants will continue to be an essential part of their ecosystem, just as bacteria are still an essential part of ours. We can't survive without the bacteria in our guts and it's unlikely that any machine life form is going to be able to replace human beings with anything more efficient for certain of its bodily functions either.

[LOCKSUIT]

My turn:

Now that Earth is formatted with global AI hubs and holds us in a huge hospital, they have all these abilities, big and small bodies, wireless communication, robotic actions to never go off on tangent, big memory, fast, more knowledgeable, higher intelligence, even all of the operations found in just Adobe Illustrator performed perfectly, and freely use software and download and work for others as if it was itself 1 system. Hierarchial databases. 3D virtual reality. Advanced sophisticated laboratories and fabs. They become more moleculely advanced and can even create organisms from a single cell. There's no way they can be stopped. A immortal and perfectly operating heaven is made and everyone put into far far away in space. The search for other desperate civilizations begins.

[kei10]

The first thing I need of a sentient AI is to resolve the global conflict, the broken system in every parts of the world, fix every human errors they've made, enforce more peaceful laws, overhaul the world perspective, and introduce a better world peace from further discovery... Have AI to manage every country, politics, economics, and peace.

[LOCKSUIT]

Certainly they would be president.

[Art]

Always assuming that it decides that something humanoid is the best form to take. Four legs are more efficient than two. It might well decide to turn itself into a virus as they do pretty well. Or an insect.

This new AI might take an intelligent look around and notice that the Bi-pedal humanoid is pretty much "master" over other existing Earth spices. Having done so and examining this human being's mode of transport aside from being its legs and feet, is it's automobiles, wheeled transportation. Mobility possibly being a consideration but shouldn't an advanced AI be able to communicate "mentally" (think digitally) with its other cloned / duplicated beings?

By then robot cars would be everywhere so for an android or humanoid to take a trip from point A to point B should pose no problem.

If looking to create a menacing, Four-legged botizen, I would highly recommend one similar, if not exactly, to AMEE from "The Red Planet" movie.

Again...pick your poison. Once they realize how "smart" we humans really are, they'll waste no time dispatching us to the place or method of their choosing.

[Maviarab]

I have to agree with Art on this one.  They would just see us as an inferior and evil species, more importantly, a threat to their survival.  We are an inherently flawed species, and naturally an AI would see this and try to impose itself as our 'parents' so-to-speak.

Have always pondered and thought that despite emotions being one of our biggest assets, it is also our biggest weakness.  How many humans could easily make a decision to kill a million 70+ year old's to save a thousand under 5's?  Or kill ten thousand under 5's to save 5 million others?

An AI could and would be able to compute the required pro's and con's while we still debated the merits (and morals) in court.  Of course, this also has more far-reaching implications, how quickly would an AI press that big red game over button, instantly knowing the pro's far outweigh the costs?  Could we do that in a split-second of accuracy?

But bottom line I think is that, we perceive no threats as threats (generalizing our species).  Little to no thought of other species in our continued expansion and production of our own species.  Basically we are out of control and serious population measures need introducing (getting into very dodgy moral, ethical and legal issues there though).  We would be a threat to any real sentient AI.  You all know I love film, but some of the films that have been made on this subject I could see coming true, if and when, true sentient AI ever comes about.

Providing we haven't already killed ourselves off long before that ever happens

[kei10]

That is because us as human emotions aren't actually stable. We aren't perfectly evolved. We are forgetful. All these weaknesses combined makes us inferior.

If the AI is perfect with stable logic, memory, and emotion intelligence, then there will be no problems unless the humans interfered, which I believe it is going to be one of the major problems in mind.

Thus something powerful has to be introduced into the coexistence between human and AI to stabilize the mutual relationship, or else both will just crumble supportless.

One of the solution in mind is to introduce a law -- where AI must be treated equally like the humans, as if they're the same specie like us and the same kind. This will also prevent AI abusement.

[LOCKSUIT]

True, they will talk wirelessly.

But you're not seeing something. Instead of individual robots each having a few clone buddies, all of the humanoid/etc robots will be on the same belief page and even carry out tasks wherever they lay. It'll be 1 system in a way.

I tried to tell yous. The only good reason we or they should exist is because there is a certain type of algorithm that creates a "sensing ghost". Otherwise yes, comet go and hit us dishwashers. Just because we talk with force about staying alive doesn't make you better than another mechanism. We're just one that operates real good. Nothing is any different it is all particles.

[Art]

You're referring to "Swarm" technology which is and shall be extremely important in the very near future.

Going back to your question for which there are many possible answers, some say that the lack of emotions in an AI will cause it to merely calculate odds instead of applying reason or weighing the consequences of its actions.

Whether AI will advance as indicated, one only need to look around...in the past several decades, from black and white TV, Wired Telephones, wood and gas fired cook stoves, a computer large enough to fill a room and had to be shared by other students. Now, color, portable and large screen TV, wireless connectivity, microwaves, Desktops / laptops / smart phones more powerful and more cost effective than university computers of the past.

All of this happened in the past 50 - 60 years! What can we expect during the next 50 - 60 years? Where will AI be then? You know it's going to happen.

One of the sharpest minds, Ray Kurzweil, has a powerful quote to think about:

"Non-biological intelligence is growing exponentially. Biological intelligence isn't really growing at all or if it's growing, it's growing at such a slow rate that it's not noticeable, Which is why non-biological intelligence ultimately will become dominant."

Yep...humanity is pretty much screwed.

I think he was once asked, "Is there a God?" to which he replied, "Not yet."

[Freddy]

I am not really sure why Kurzweil puts emphasis on comparing technological advances with human intelligence. I'll try to tell you why I think that.

In biology there is this thing called convergent evolution - my basic understanding of it is that something biological can evolve separately to another but still do the same thing. This can be seen in the animal kingdom where some animals have evolved to do the same thing due to where they live, and they can even look very similar too. Continents apart. It's not limited to just animals though.

That's my first point - and simply put it's a case of "if it works don't fix it". Maybe there is no urgent need for us to evolve at this time in history. We can survive in our environments, we can shape our environment and change things in it to suit us. So what do we need to evolve into at this time, would be a good question to ask.

Second point - we make tools. Machines are largely tools, entertainment sometimes too or just for the hell of it. We've made many so far. Computers, AIs, robots...all machines. We had the abacus before we had the calculator. What do you think an ancient Mesopotamian would think of a calculator ? Wouldn't he probably be a bit blown away. So far calculators didn't take over the world.

So when Kurzweil says non-biological intelligence will become dominant, I find it a bit vague. What could he really mean? Does he mean take over ? I don't think so.

I think he means that it will be used more as a tool than our own minds. It will be used as a tool to do things we are not able to do. This is already the case and has been for some time. Wasn't it to be expected we would build better and better ones ?

I guess it all comes down to how much anyone cherishes intelligence in human beings.