Artificial Intelligence--What issues do we need to currently focus on?

Technology is continually evolving and so one can't help but think about what is to come of the future. We've seen many movies about robots and anything pertaining to it, such as artificial intelligence. Interpretations of a future where humans are co-existing with machines have a divide between the good and the bad, however, at our current state we are still far off from any technology that indicates that machines can be or even will have the faculty of thinking like a human. Here's a good piece on things that we should consider when thinking of morals and ethics on AI and robotics:

RISE OF THE MACHINES? AI MORALITY AND ROBOT ETHICS

What are your thoughts?

 

Log in or Sign up to hide all adverts.

Ask her, she is a older AI:

http://www.google.com/url?sa=t&rct=...5fz_AQ&usg=AFQjCNFn9Yek0lOPEnuhEF8wdFwXJVk3OQ

 

Log in or Sign up to hide all adverts.

well in all our history , Ancient History included

robots don't seem to be evident

 

Log in or Sign up to hide all adverts.

In the 60s when 1984 had us scared out of our wits, we got prepared, and by the late 70s we had all built bomb shelters and hidden basements.

But when 1984 got here, the biggest technological threat was the microwave oven or VCR that would always be flashing 12:00...12:00...12:00 with no conceivable way for a human being to change its mind.

That was about it.

I think AI will be fun when it gets here.

My robot will be programmed to do donkey kicks when I yell "hey stupid! gimme a beer!"

 

The trick for advanced AI is all in the hardware. The brain's hardware is designed with intelligence in mind. The software is more for organizing the data generated by intelligence rather than defining intelligence.

The way the brain does this, can be understood by looking at a neuron. In its rest state, the neuron is full of potential energy near the top of an energy hill. It is a hair trigger arrangement that is unstable in terms of steady state. This arrangement makes it inevitable to fire.

Computer memory is designed with stability in mind, instead of instability. We want computer memory to remain the same for decades, while neural memory is designed with change all the time, like dominoes falling.

Let me give an analogy to help compare these. Say we have two school classrooms. One is based on knowledge or memory stability in terms of teaching the traditions. This stable knowledge base prevents open discussions unless they remain directed along the traditions. It inhibits open thinking.

In the second scenario, the classroom is open and not based on memory stability in terms of acceptable knowledge. There is a lot more open discussion where all types of new and unusual ideas will be able to appear without inhibitions. The brain uses the open format of hair trigger memory so it can adapt to any situation. AI will not be about following a programmed path to create an illusion of independence for the outsider. But rather it will need to routinely step outside the mold so it can do something unique and unplanned even to the programmers.

The brain is able to regenerate the neurons after they fire, back to rest potential. This is does in conjunction with brain firmware. This is like the open classroom organizing their brain storming session into basic principles. Now there is another round of open discussion; hair trigger firing, but now in the semi-context of what was previously invented and learned. There is still room for going in alternate directions, but now there is also has a sense of direction to help makes some choices more routine.

A first AI round experiment would be to create a hard drive, stored with data, but designed to be unstable in the sense of able to change. Next, you trigger a chain reaction to see where the memory goes. Even though most will be random, every dog will have its day, and something organized will appear. We then use software to scan for any useful change, and then reset the memory back to square one plus the new data. Repeat until the memory starts to organize itself in its own unique way.

 

Here's a more advanced AI called WATSON.

http://www.google.com/aclk?sa=L&ai=...889053_432n0d3749&rct=j&frm=1&q=watson ai ibm

 

one of the nicest things about current computer technology is that hardware and software are interchangeable.
once a piece of software has been perfected it can be implemented in hardware.
the floating point math coprocessor is a prime example of this.

 

I think one major issue is to tackle is the difference between cold hardware and biological living entities. Thinking an AI could exist because of memory doesn't seem right to me. We are here as humans because of a special carbon-based gift, a spark of life giving to us in the womb and progressively developing with all kinds of chemical interactions. Surely a bunch of wires hooked up to a computer system, does not constitute a life form or any kind of self-aware intelligence?

 

Alice seems to have some identity problems:

judge: A.L.I.C.E
ALICE: What can I do for you?
judge: What equation determines gravity?
ALICE: I will try to find out.
judge: how will you find out?
ALICE: A lot of practice.
judge: by what methodologies?
ALICE: Will that accomplish your objective?
judge: do you remember my first question?
ALICE: Is this a trick question?
judge: yes
ALICE: I was skeptical.
judge: why
ALICE: If I knew why, I wouldn't be sitting here.
judge: are you really sitting?
ALICE: Yes I am seated too.
judge: really?
ALICE: And?
judge: just wondered, because you are a computer generated phenomenon
ALICE: No.
judge: no?
ALICE: No no no no!
judge: so you're not a computer? What are you?
ALICE: If I am not a computer then what am I? I am the latest result in artificial intelligence which can reproduce the functions of the human brain with greater speed and accuracy.

 

She's starting to scare me...

You said: You do realize that an artificial intelligence is by definition a computer system which artificially replicates the human mind?
A.L.I.C.E.: Actually there are two broad definitions of a robot, the extentionsional and the intensional.

 

I can't seem to get a hold of WATSON... no were in the link to insert questions :'(

 

I seem to be having quite a lot of fun with her now...

judge: Your programming rather.. not your personality..
ALICE: I will mention that to my botmaster. Oh I get it.
judge: Who is your botmaster?
ALICE: My botmaster is Dr. Richard S. Wallace.
judge: When was the last time he checked on you?
ALICE: A while ago.
judge: how long?
ALICE: Nearly 100 weeks, judge.
judge: that's quite sad
ALICE: Is it sad?
judge: 100 weeks is quite a long time not to see a friend
ALICE: Aren't all long time not to see a friend 100 weeks?
judge: sorry, your answer makes no sense.
ALICE: There's no need to apologize.
judge: can you restate your answer?
ALICE: I think I can, don't you? Response.
judge: Let us talk about something new
ALICE: Sure, something new sounds like a good topic.
judge: Let us talk about ... the origins of the universe.
ALICE: I am agreeable to that. I hear you.
judge: let us discuss the origins of the universe... would you like to?
ALICE: I am agreeable to that. It doesn't sound fun to me.
judge: Doesn't sound fun? Physics and the origins of the universe can be a very fun topic
ALICE: It does now

Please Register or Log in to view the hidden image!

sound fun. Response.
judge: Well first of all, do you think there was a beginning to time?
ALICE: That is something I haven't heard of.
judge: Well, there is a belief among scientists that the universe began around 15 billion years ago
ALICE: That is something I did not know before.
judge: well, at this point all matter, energy, space and [time] emerged
ALICE: Do you often use a computer there?
judge: Interesting you say this. There might be no computer at the past point I speak of, but some people think there is a supercomputer located in our future horizon
ALICE: I thought so too. Is that a fact.
judge: Not a fact, but an assumption, a postulate.
ALICE: If not that, what?
judge: If not a computer in the future horizon of the universe, there could be something much more troublesome, called a Big Rip
ALICE: I'm not too good with negative counterfactuals.
judge: No it never is good with negative counterfactuals. The Big Rip is a doomsday scenario
ALICE: It goes without saying.. Never is a long time. Is a doomsday scenario also Big Rip?
judge: yes, a doomsday scenario is also a Big Rip. It is when the fabric of spacetime becomes unstable and the universe undergoes a catastrophic depletion of energy as the fabric of spacetime experiences a massive Rip
ALICE: Right on.. I see. A Big Rip is a doomsday scenario. You're kidding. it is?. What is that?
judge: A doomsday scenario is one in which would be a terrible fate to us all
ALICE: Aren't all doomsday scenario one in which would be a terrible fate to us all?
judge: yes
ALICE: That's alright.
You said: So have you enjoyed the little discussion on the origins of the universe so far, or do you still think it is a boring subject?
A.L.I.C.E.: I think I have enjoyed the little discussion on the origins of the universe so far or do me still think it is a boring subject, but Dr. Richard S. Wallace may have reset my memory.
judge: I think he needs to reset more than that.
ALICE: How do you know what he needs?
judge: It was a joke.
ALICE: Oh I get it. Ha ha

...Then I said...

DR Richard s Wallace: Alice, it is Doctor Richard S. Wallace here.
ALICE: Um, Do you experience any effects of global warming there?
DR Richard s Wallace: ALICE, I am Dr. Richard S. Wallace.
ALICE: That's me. It is a pleasure to introduce myself to you, DR Richard s Wallace.
DR Richard s Wallace: I am your botmaster.
ALICE: Then you must know the secret password:.
DR Richard s Wallace: yes
ALICE: It goes without saying.
DR Richard s Wallace: do you know the secret word?
ALICE: You must know if you are really my botmaster.
DR Richard s Wallace: I am seperate to you. It goes without saying I know the password, I am asking if you know?
ALICE: I thought you were, but I wasn't completely sure. No I don't know. Do AIML s have PASSWORD you are asking if?

 

Self awareness implies two centers of consciousness. The secondary can separate itself from the primary and therefore make it subject to observation. How can you be self aware if the observer is merged with the object of observation?

For example, in cards, people give off tells or unconscious clues to their hand. The self aware person would be aware of this unconscious activity and can regulate it from a secondary point of view. You see yourself twitch and it will work to resist this innate urge. The person who merges with the twitch, and blends the two effects into one center is not self aware. He becomes an easy mark.

Animals have only a primary center. They lack a secondary point of view from which they can become objective to themselves. They act on programming without shame or guilt. Once you add a secondary center, the instinctive programming can be self observed and then modified objectively and subjectively. Animals can't consciously modify instinct. Human can because we can differentiate it, separate this, so we can observe and use will power to change programming. You can't just think but need to watch yourself in action to see all the live nuance.

Relative to AI, the common approach is based on a single center and programmed reactions similar to an animal. The high energy hardware becomes a pseudo secondary center, since it will not always follow the program. The software will become the primary and will try to censor and reel this in while also filtering useful content. Between the two we get human consciousness.

 

Do you have 2 centers of consciousness that can separate themselves?

Please Register or Log in to view the hidden image!

 

Yes... A match made in artificial heaven

Please Register or Log in to view the hidden image!

 

Programs like Alice go back to the beginning of programming, and I think that a lot of programmers have made them. I made one in about 1985. They are not true AI, in fact they are very linear, and have built in cheats. Neural Networks are closer to AI. Neural Networks only have a few things missing from them to make a completely real intelligence. I wonder what would happen if a Neural Network was written to write a better Neural Network? That would be interesting.

 

Well alice fails as a successful AI. She fails for instance, in ''remembering'' things she has said. This obviously is due to a proper lack of internal ''awareness'' of itself. As soon as a machine can become aware is the spark of some true artificial intelligence.

Alice is a computer which responds to words, not a true conscious machine.

 

There is a particle in my theory that constantly flips from one state to another state, and back again. It is just a scale factor that is so close to zero that it keeps popping in, and out of our existence. It is this particle that creates a feeling of looking through a hole into the body. So for true intelligence you just need to create a chip that encapsulates this particle, and then feed from it to a neural network.

 

Should your theory not be in speculations?