The problem of designing an AI to understand what it sees

@deepthought said
It just follows the rules one's given it. The problem is that we're talking about these things as if they're algorithms in the traditional sense. The type of AI humy is talking about is something more like a brain simulation. A brain simulation would make these judgements in the same way a brain does.

Here is the difference, and it is a big one! We think through problems, and we can calculate mathematical questions; we are self-generating these things owning all of it. When we come up with an answer right or wrong, we own our answers; they are what we came up with in the end.

Not so a computer, it doesn't know, grasp, care, understand anything about what it does. It does one thing or another it is meaningless, it is a device, it is just doing what it must without understanding! We can program a computer to play chess, win or lose it is meaningless to the computer! We can program it to calculate 1+1=2; it doesn't know what 1+1=2 is, or why, there may be an "AND GATE" dealing with a truth table within it. It only does what it was designed to do, and there is no intelligence within the system that has meaning to the computer, only us! Projecting intelligence into a lifeless thing is like giving human traits to an animal; we are merely attempting to relate with it.

There is no understanding in the computer it is just doing is what we program into it. We could claim building a water system is intelligent because we use gravity to cause water to go where we want it. I don't see why we even attempt to make such claims, humy's logic and programming abilities are inspiring! The software and hardware are to, but on their own without humy's skill sets, they would sit there.

@KellyJay
The same could be said of the brain of a bee. You would have to admit it doesn't know how it knows to find a flower with nectar. So it takes more organization of larger numbers of brain cells to get to the point of self awareness like the great apes, not able to design tools only use a branch as one and so forth but it knows what it is doing and why.
I think the same will happen to computers.
Remember even the fastest strongest of the best of the world right now would have the intelligence of a bee and computers coming up in the next few decades will be orders of magnitude stronger and I think quantum computers will be matched to classical computers of the future so they WILL become self aware.

It is clear your objection is coming from religious values, such as humans not EVER being able to do what you think a god did in terms of human intelligence.

@sonhouse said
@KellyJay
The same could be said of the brain of a bee. You would have to admit it doesn't know how it knows to find a flower with nectar. So it takes more organization of larger numbers of brain cells to get to the point of self awareness like the great apes, not able to design tools only use a branch as one and so forth but it knows what it is doing and why.
I think the ...[text shortened]... s, such as humans not EVER being able to do what you think a god did in terms of human intelligence.

The bee knows what it is doing it is aware and alive, a computer is not, as I pointed out earlier there is more intelligence in a piss ant than a computer it has none. You project a great deal about what you think is going to occur in the future, to the point of ignoring what is in the present.

When it comes to God and man, your correct. God knows it all from the beginning and the end, when do you think man will arrive at that state? God sees it all from the beginning to the end, when do you think man will arrive at that state? It isn't close, the universe cannot contain all that God is, and you want to compare Him to man?

That, however, doesn't address my complaints on a lifeless computer acquiring real intelligence by our ability to design hardware and software. You can input the total human totality of knowledge into a computer, it will still no nothing, it isn't able to understand and know, it is a lifeless thing.

@kellyjay said
The bee knows what it is doing it is aware

Define what you mean by "aware". And "aware" of what? And how is whatever your meaning by "aware" such that an AI cannot be "aware"? You keep making assertions that are not only without evidence or premise but are also pretty vague and badly worded.

@humy said
Define what you mean by "aware". And "aware" of what? And how is whatever your meaning by "aware" such that an AI cannot be "aware"? You keep making assertions that are not only without evidence or premise but are also pretty vague and badly worded.

Aware is a matter of understanding, there is no understanding in a computer before a problem is presented to a computer, during a calculation within a computer as it is solving what we put to it, or after the calculation is completed with a correct or incorrect answer, the computer remains the same there is no understanding where it becomes aware of something new.

@kellyjay said
Aware is a matter of understanding,

Don't know what that's supposed to mean.
Define what YOU mean by "aware" and "understanding" and explain in exactly what way it differs from what I mean by the words so that that above assertion makes some sort of sense...

Me speaking here as an AI EXPERT; Some AIs ALREADY have at least some rudimental level of understanding and AI designs are incrementally improving all the time and there is no 'barrier' to AIs eventually general gaining understanding of most things at least as well as that of the average human. Thus its just a matter of when, not if, an AI will come to understand most things at least as well as that of the average human.

Understanding is just being able to define both the intrinsic properties of things and relationships between things. Some AIs have already been demonstrated to be able to do both. How does what YOU mean by the word "understanding" differ from that?

@humy said
Don't know what that's supposed to mean.
Define what YOU mean by "aware" and "understanding"...

I think you are purposely ignorant now, a state of awareness on my part about you and your argument, or I see what you mean.

@kellyjay said
I think you are purposely ignorant now,

I am an AI EXPERT and with AI university qualifications + years of experience doing actual AI research and still doing AI research. You are not. Its YOU who are ignorant about AI and what they can do, not me; And by choice and I can only assume because you think the truth offends your religion, which is why you don't listen to the experts when they say something you don't want to be true.

@humy said
I am an AI EXPERT and with AI university qualifications + years of experience doing actual AI research and still doing AI research. You are not. Its YOU who are ignorant about AI and what they can do, not me; And by choice and I can only assume because you think the truth offends your religion, which is why you don't listen to the experts when they say something you don't want to be true.

If there is something about my religion you think I've been injecting bring it up. I've been speaking about a object we build that you have assigned traits to that do not belong. I don't care what your occupation or degrees are, they don't alter the fact that you can be right or wrong, and in this case your wrong.

@kellyjay said
I don't care what your occupation or degrees are,

So you assume you know better about a subject than the experts on that subject?
That's pretty arrogant and delusional. I never assume I know better about a subject than the experts on that subject UNLESS, and even then not usually, I AM an experts on that subject.

@sonhouse said
@KellyJay
The same could be said of the brain of a bee. You would have to admit it doesn't know how it knows to find a flower with nectar. So it takes more organization of larger numbers of brain cells to get to the point of self awareness like the great apes, not able to design tools only use a branch as one and so forth but it knows what it is doing and why.
I think the ...[text shortened]... s, such as humans not EVER being able to do what you think a god did in terms of human intelligence.

...I think quantum computers will be matched to classical computers of the future so they WILL become self aware.

I think this is where critics like Musk are raising alarms. Computational self awareness is a problem if you're a human, especially if we want them thinking like us. Humans have made some terrrrrrible decisions throughout history, justified in twisted logic and religious dogma and power. Our empathy in general is limited to our own kind above others. If we create self-aware computers and set them loose on the internet, bad things are likely to happen and we don't have adequate safeguards in place.

I think generally people underestimate the capability of AI. They sort of think like it’s a smart human.

“But it’s really going to be much more than that. It’ll be much smarter than the average human.

“It’ll be like, can a chimpanzee really understand humans? Not really, you know, we just seem like strange aliens. They mostly just care about other chimpanzees.

“And how it is more or less and if the difference is only that small, that would be amazing. Probably, it’s much greater." - Elon Musk

@wildgrass said

...I think quantum computers will be matched to classical computers of the future so they WILL become self aware.

I think this is where critics like Musk are raising alarms. Computational self awareness is a problem if you're a human, especially if we want them thinking like us. Humans have made some terrrrrrible decisions throughout history, justified in ...[text shortened]... ference is only that small, that would be amazing. Probably, it’s much greater." - Elon Musk[/quote]

What is a self-aware computer?
One that thinks for itself, and only does what it wants?
Could one then choose to give you a bad answer to a math problem, because it doesn't like you?

@KellyJay
You are catching on. For instance, suppose at some future time, a computer is developed after centuries of development like we have seen to this date, so this computer is a thousand times smarter than any human. Suppose it is asked by humans to develop a bomb so powerful it can blow up an entire planet in a second.
Do you think said computer will go, sure, I already knew how to do that, here are the specifications for such a weapon and oh BTW, here is one even more powerful, it can blow up the SUN.
Do you seriously think it would comply with such?