amused

[mkh]

is it true that it is theoretically impossible to create artificially intelligent machines or simulate the human brain? if so what is the theory behind it?

see this link www.americanheritage.com/articles/magazine/ah/2001/1/2001_1_28.shtml

 

[Hero999]

Some say that if we make a computer clever enough to innovate then it will design a more complex computer and program it. Resulting in an even more powerful computer that will create a more powerful one. The whole cycle will increase exponentially until they take over the world.

This is total BS because even if computers become powerful enough to thoretically simulate a human brain it doesn't mean the software will be able to do the same.

The thing that's lacking (as the artical correctly states) is common sense reaoning and inventiveness, neither of the two have advanced much since computers were invented.

 

[Sceadwian]

The hardware isn't really the issue. Raw processing speed does not inteligence make =) It's the interconnectedness and 'fuzzy' weighting that the brain does and it's various interactions that make it difficult to create human like AI. We simply don't know enough about how our own brains work to atempt to mimic them in the digital world. We have difficulty discerning the differences between a psycho's mind and a normal human's mind, or people with autism, and those are extreme cases. The mind is a masterful concoction of different parts that do different things, and human conciousness itself doesn't exist in any one part, it's a side effect of so many complex parts interlinked in such complex manners. We can barely reconstruct insect inteligence which is only a few shades above packs of nerve clusters let alone understand the dynamics of the human mind.. Digital neurons are easy enough to research, but creating functional networks is a different story. As the years go by and medical tehcnology increases and the understanding of the human brain, or brain's in general increases the chances for being able to replicate a functional digital 'brain' will increase. It's exceptionally difficult if you realize one thing, that aside from the cellular connections the chemical nature of the human mind is little understood and even a SLIGHT upset in a single chemical in the brain leads to complete changes in the function of the brain.

 

[Dr.EM]

2 points really. If you did re-create it, how exactly would it have consciousness, self awareness? Does anyone know why we are self aware, have a soul as it were? It would be a depressing day indeed if someone found out why, I personally hope they don't and it leads onto my second point; why would you want to create an intelligent machine? Purely to satisfy a curiousity. If you succeeded in doing it, theres a good chance you'd be being terribly cruel by giving a machine self awareness whilst it would be trapped with no real communication to the outside world. A living nightmare for any human I think you'll agree. I don't know why there arn't laws preventing such research, it's totally uneeded and leads to terrible consequences via many routes. A lot of the "research" involves disgraceful procedures on animals anyway.

 

[mkh]

thinking with the present in mind,is it possible to produce that much processing power to for example process emotions?

 

[dknguyen]

The hardware isn't really the issue. Raw processing speed does not inteligence make =)

*Yoda voice*
"Force ability does not a Jedi make."

 

[The Mad Professor]

is it true that it is theoretically impossible to create artificially intelligent machines or simulate the human brain? if so what is the theory behind it?

see this link www.americanheritage.com/articles/magazine/ah/2001/1/2001_1_28.shtml

Wetware may provide a solution , the philosophical arguments have pretty much been thrashed out in Science-fiction circles for some time. Nowdays with the advances in "cloning" research and the demonstrations a couple of years back of brain material from a rat acting as the "pilot" connected to a flight simulator...

Well it appears to me it's not a hardware problem , more of a moral one.


If real is what you can feel, smell, taste and see, then real is simply electrical signals interpreted by your brain
By Morpheus on Reality

 

[Pommie]

If it was possible today to map my brain such that the state of every neuron, strength of every connection and any other relevant parameters were recorded. And, if this data was then used in a computer in 10,000 years time to simulate the actions of my brain at the chemical level, would I be aware? Would I be alive? Would I be capable of original thought? Would I be an intelligent machine?

If the same data was fed into the biggest computer that currently exists and my brain simulated at the chemical level and it ran at 1 millionth of what would be considered normal. Would I be aware? Time would have no meaning. 1 second for me would take 1 month for "real" people. After 100 years had passed, might I have learned to communicate?

It does my head in thinking about such things.

Mike.

 

[Papabravo]

*Yoda voice*
"Force ability does not a Jedi make."

Learn Forth quickly I will

 

[Marks256]

"If the human brain was so simple that we could understand it, we would be so simple that we couldn't." -Emerson Pugh

 

[Pommie]

"If the human brain was so simple that we could understand it, we would be so simple that we couldn't." -Emerson Pugh

There is no doubt in my mind that we will get to the point where we can simulate the human brain. This might be a million years from now but will eventually be possible. We don't have to understand it to simulate it. If we did simulate it, would the simulation be aware or would it be like Searle's Chinese room?

Mike.

 

[Marks256]

That is my question. Would an A.I. be self aware? What makes us ask questions? How do we generate questions? What make us self aware?

 

[Sceadwian]

The thing is you don't have to get AI to emulate the human brain, NOTHING can do that, only the human brain will EVER act like the human brain. We will eventually be able to create faxsimiles of some human brain functions though, at least enough to process information from it's raw form into a symoblized set. Say a vision system capable of determining the shape of an object, and from a set library determine if it was a chair, a car, a blue chair, a red car. an object shaped like a chair that's blue with some red, a car with a red body and orange stripes etc.. etc.. ad infinum. Properly network weighted logic systems (I hate to use the term fuzzy logic because it is so abused) can already do this and more, to greater or lesser degrees of reliability and sophistication. It's only a matter of time before there are language feedback systems that allow one computer to compile an image into a set of symbols and transfer it to another computer that is capable of reconstructing the basic scene from the descriptive language into a visual one. Basically forming a feedback system roughly equivilant to the human ability to imagine. Adding more computers to the system and various nodes of sensor and data storage and processing entire networks will begin to form that can produce 'inteligent' actions from it's raw sensor input and some intermediary stages. Weather or not this will ever actually occur is more a matter of our social structures instability rather than technical ability. It will at our present course only take time. And not likley within our lifetimes will we see inteligence in computers more than a young child or chipanse. Although A.I. will over time more significantly impact our interactions with our perceived reality, more in an augmentive way rather than as a seperate entitity.. that will change people as well. Various methods of networked communication have already started to change the way people interact with one another on varying levels, some dramatic.