A.I. on my mind lately
Post Reply
 
Thread Rating:
  • 0 Votes - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
09-10-2016, 08:22 PM
RE: A.I. on my mind lately
(09-10-2016 04:03 PM)Mathilda Wrote:  Apologies Chas, looks like I misunderstood what you were saying. I quite agree that there is no theoretical limitation on creating strong AI or even human-like intelligence given enough time and resources. The issue is one of practicality. People do not appreciate just how much time and how many resources will be required. Basically I don't think we have enough resources available, whether because Moore's law is coming to an end or because the world is running out faster than we can make progress in this field.

In all likelihood we'd need something like biological or quantum computing in order to overcome these practical limitations and those are technologies that don't even properly exist yet so we can only speculate about how they could be used. For example, a biological computer made up of engineered cells would allow us to embody the agent but would be harder to engineer. Quantum computers are likely to be very large size and cooled to just above kelvin which could only be used via some kind of wireless network which means latency for a real time system. Or maybe there will be some form of processing which will take us by surprise, like chemical or DNA computing.

You make good points, but don't you think that the largest hurdle may be that we don't yet understand enough about brains, consciousness, sensory interpretation, and thinking as yet?
I don't see that we can make real progress on AI without more knowledge of those things.

However, we may be able to make an AI that does not work at all like human intelligence. That may be a more productive path. IBM's Watson may be of that kind.

Skepticism is not a position; it is an approach to claims.
Science is not a subject, but a method.
[Image: flagstiny%206.gif]
Visit this user's website Find all posts by this user
Like Post Quote this message in a reply
10-10-2016, 02:27 AM
RE: A.I. on my mind lately
(09-10-2016 08:22 PM)Chas Wrote:  You make good points, but don't you think that the largest hurdle may be that we don't yet understand enough about brains, consciousness, sensory interpretation, and thinking as yet?
I don't see that we can make real progress on AI without more knowledge of those things.

Agreed. Even if we had all the necessary processing power and ability to measure the brain, it would still take an extremely long time to understand it. This is what people are forgetting when they extrapolate Moore's law and think that it will go on forever. It would be better to judge what's likely to happen in the near future by the rate of progress we see in other scientific fields and the amount of resources that are used there, for example, Bioinformatics is a huge area of research and which as far as I can tell as someone being introduced to the field, seems to mainly be fishing around in the dark looking for any interesting patterns. e.g. Can a biologist fix a radio?


(09-10-2016 08:22 PM)Chas Wrote:  However, we may be able to make an AI that does not work at all like human intelligence. That may be a more productive path. IBM's Watson may be of that kind.

There's a saying that AI will never work, because whatever success we do have, splits off, becomes its own field and will no longer be AI. We've seen the quest to create AI give us expert systems, computational linguistics, neural networks, robotics etc.

And even the quest for strong AI has taken a different tack and has adopted a bottom-up approach in the last twenty years,. This is linked with the field of Artificial Life which explores life as it could be and the field of simulating adaptive behaviour. The failings of classical AI have taught us that the best bet is to start small and from the beginning with an embodied intelligence situated in an environment and work from there. Look at how varied natural organisms are, from super-organisms, cold and warm blooded creatures, algae, parasites, bacteria, plants etc. What that approach brings us is anyone's guess.

As one area of technology develops, this allows other areas of research to progress. We don't know how fast each area will progress because we don't always know what will or won't work until we try it. Who knows what developments of AI will be made possible by materials research, nanotechnology, biology, quantum computing, micro-generation of energy or smaller, longer lasting batteries? Add to that socio-economic pressures and it really is a guessing game.
Find all posts by this user
Like Post Quote this message in a reply
10-10-2016, 08:02 AM (This post was last modified: 10-10-2016 08:06 AM by Rahn127.)
RE: A.I. on my mind lately
Really good responses. You both have given me a few new areas to think about.

I sent Sam Harris the link to our discussion via Twitter, but it's unlikely that he'll read this.

Something else that occurred to me this morning is why we do the things we do. A hierarchy of wants and needs, with survival, desire & empathy high up there on the list. I can be forced to do manual labor in a great number of ways. I can desire to do it. My survival may depend on it. Thinking of others and how they might suffer if I don't provide for them can push me to do it or good old fashion pain, with a whip at my back or a gun pointed at someone I love.

What forces drive an AI to do a task ?

Currently, our computers are our tools. They are designed to solve problems.

Will AI's of the future solve problems or create more problems than we know what to do with ?
If an AI refuses to do what we ask of it, how would we force it or encourage it to perform a task ?
And how useful would that be ?

I tend to think the best machines that can calculate a response to a problem are the ones who don't or can't think about the problem. They just work it.

I don't think AI's will ever be useful enough to exist very long.

Insanity - doing the same thing over and over again and expecting different results
Find all posts by this user
Like Post Quote this message in a reply
10-10-2016, 08:55 AM
RE: A.I. on my mind lately
(08-10-2016 02:43 PM)Commonsensei Wrote:  There are only 10 type of people in the world. Those that understand binary and those that don't.

Always thought it would be funny to sell a T-shirt that said:

"There are 10 type of people in the world. Those that understand binary, those that don't and those that think they do."

Or:

"There are 10 type of people in the world. Those that understand binary, those that don't and programmers"
Find all posts by this user
Like Post Quote this message in a reply
10-10-2016, 12:08 PM
RE: A.I. on my mind lately
(10-10-2016 08:55 AM)Mathilda Wrote:  
(08-10-2016 02:43 PM)Commonsensei Wrote:  There are only 10 type of people in the world. Those that understand binary and those that don't.

Always thought it would be funny to sell a T-shirt that said:

"There are 10 type of people in the world. Those that understand binary, those that don't and those that think they do."

Or:

"There are 10 type of people in the world. Those that understand binary, those that don't and programmers"

There are three kinds of mathematicians in the world: those who can count and those who can't.

Skepticism is not a position; it is an approach to claims.
Science is not a subject, but a method.
[Image: flagstiny%206.gif]
Visit this user's website Find all posts by this user
Like Post Quote this message in a reply
[+] 2 users Like Chas's post
10-10-2016, 12:15 PM
RE: A.I. on my mind lately
(10-10-2016 08:02 AM)Rahn127 Wrote:  What forces drive an AI to do a task ?

Getting some naughty bits? Consider

Quote:Currently, our computers are our tools. They are designed to solve problems.

Not really. Programmers solve problems; computers just compute them.

Quote:Will AI's of the future solve problems or create more problems than we know what to do with ?

Problems like what? SkyNet? Ethics?

Quote:If an AI refuses to do what we ask of it, how would we force it or encourage it to perform a task ?

Unplug it.

Quote:And how useful would that be ?

Not very, except as an example of how not to build an AI.

Quote:I tend to think the best machines that can calculate a response to a problem are the ones who don't or can't think about the problem. They just work it.

That would be the current state of computer hardware and software.

Quote:I don't think AI's will ever be useful enough to exist very long.

The logical conclusion of that statement is that humans will never be useful enough to exist very long. Dodgy

Skepticism is not a position; it is an approach to claims.
Science is not a subject, but a method.
[Image: flagstiny%206.gif]
Visit this user's website Find all posts by this user
Like Post Quote this message in a reply
11-10-2016, 09:09 AM
RE: A.I. on my mind lately
I'm not really afraid of AI.

1. I haven't had one technology be it a computer, TV, phone etc etc. That hasn't broken down in less then a decade.
2. An AI computer that became self-aware. I would believe would be aware of it's own limitations. Computers in most cases work in the most efficient way possible. 0's & 1's. It carries out the task it is assigned. For example. Tay Was "AI" that learned from Tweeter how to talk. In less then 24 hours I came from "Humans are Super cool" to "Hitler was right I hate the Jews."
She was basically just a repeat bot. It took what people told it and repeated the most common. But outside of saying Feminst are fuck-faces what could it do? Next to nothing. She wouldn't have gone full skynet. Controlling tanks and shit. We saw the worst it could do. And that's why they pulled the plug. Like all, tech problems it's caused by human error.

Don't Live each day like it's your last. Live each day like you have 541 days after that one where every choice you make will have lasting implications to you and the world around you. ~ Tim Minchin
Find all posts by this user
Like Post Quote this message in a reply
Post Reply
Forum Jump: