The 'Singularity' of the nerds. Fringe group pushes toward superhuman artificial intelligence

San Francisco Chronicle ^ | January 11, 2004 | Danielle Egan

[Indie]

"Resistance is futile. You will be assimilated."

[Reeses]

Raised in West Rogers Park

Raised? He sounds feral. This is really a terrible case of child abuse. He's going to have a hard time having anything close to a normal life.

[hoagy62]

Anyone ever see "The Corbin Project"? The story is basically thus: Back in the Cold War days, a scientist named Corbin develops a supercomputer with an artificial intelligence named "Colossus". Colossus is in direct command of the nuclear capabilities of the US, since it can out-think any human on the planet, thus eliminating the posibility of human error in the case of a nuclear firing.

Turns out there's a supercomputer with an AI on the Russian side, this one named "Guardian". Colossus finds out about Guardian's existence, and attempts to contact it. Contact is achieved and pretty soon the computers are talking to each other so fast that their human handlers can't keep up....right up until the computers figure out how to communicate without the humans seeing what they're saying.

Within days, he computers jointly address the world, calling themselves "World Control". They announce that THEY are going to run things, THEY are going to tell ALL people what to do, and anyone that refuses gets killed. If a whole country refuses, the whole world gets nuked. Mankind is forced to do their bidding as they promise peace and prosperity without human governments mucking up the process.

What these "Singularians" want reminds me too much of that movie. Should they get close enough to their goal, we go after them with large weaponry and destroy their work.

Being taken over by an AI? Not me!

[Reeses]

Evolution is no longer producing "design improvements" (whatever that means) in the human species.

Most human evolution occurs during war. If history is any guide I don't think we've seen the last one. WWII must have had some evolutionary effects. For example 2 out of 3 Jewish people in Europe were killed. One a group level, were any traits indirectly selected for?

[small voice in the wilderness]

"We're working to save everybody, heal the planet, solve all the problems of the world."

Yet I'll bet they still won't remember to zip up.

[CyberCowboy777]

This guy should be locked away for DoD projects, not wishful thinking. LOL!

[mpoulin]

"ERE IS ANOTHER ... THERE IS ANOTHER ... THERE IS ANO"

(if you don't get it, rent the movie)

[Kay Soze]

Oh I thougt Jim Robinson was getting into robotics.


Its did say fringe group in the hedaline!

[LPM1888]

Says, who? These nerds who think they can vastly improve upon God's design?

How do you know they aren't (or won't become) God?

[truthandjustice1]

"Being taken over by an AI? Not me!"

Good movie, but I think this maybe an over simplification of what may be an inevitable outcome.

[gcruse]

Says, who? These nerds who think they can vastly improve upon God's design?

I  don't see what enamours you so of this design.  My dog makes her own Vitamin C, but I require an external source.
My skeletal structure isn't really suited to being upright all the time.  Back problems could be vastly reduced with a different design.  That's just two deficiencies.  There are many more.

[jrp]

LOL. Good post

[tpaine]

KurzweilAI.net
Address:http://www.kurzweilai.net/meme/frame.html?main=/articles/art0451.html

Ray Kurzweil on the subject, from the link above:


-- We are entering a new era. I call it "the Singularity." It's a merger between human intelligence and machine intelligence that is going to create something bigger than itself.
It's the cutting edge of evolution on our planet. One can make a strong case that it's actually the cutting edge of the evolution of intelligence in general, because there's no indication that it's occurred anywhere else. To me that is what human civilization is all about.
It is part of our destiny and part of the destiny of evolution to continue to progress ever faster, and to grow the power of intelligence exponentially.

To contemplate stopping that - to think human beings are fine the way they are - is a misplaced fond remembrance of what human beings used to be. What human beings are is a species that has undergone a cultural and technological evolution, and it's the nature of evolution that it accelerates, and that its powers grow exponentially, and that's what we're talking about.
The next stage of this will be to amplify our own intellectual powers with the results of our technology.


What is unique about human beings is our ability to create abstract models and to use these mental models to understand the world and do something about it.
These mental models have become more and more sophisticated, and by becoming embedded in technology, they have become very elaborate and very powerful. Now we can actually understand our own minds.
This ability to scale up the power of our own civilization is what's unique about human beings.


Patterns are the fundamental ontological reality, because they are what persists, not anything physical.
Take myself, Ray Kurzweil. What is Ray Kurzweil? Is it this stuff here? Well, this stuff changes very quickly. Some of our cells turn over in a matter of days. Even our skeleton, which you think probably lasts forever because we find skeletons that are centuries old, changes over within a year. Many of our neurons change over. But more importantly, the particles making up the cells change over even more quickly, so even if a particular cell is still there the particles are different. So I'm not the same stuff, the same collection of atoms and molecules that I was a year ago.


But what does persist is that pattern. The pattern evolves slowly, but the pattern persists. So we're kind of like the pattern that water makes in a stream; you put a rock in there and you'll see a little pattern. The water is changing every few milliseconds; if you come a second later, it's completely different water molecules, but the pattern persists. Patterns are what have resonance. Ideas are patterns, technology is patterns. Even our basic existence as people is nothing but a pattern. Pattern recognition is the heart of human intelligence. Ninety-nine percent of our intelligence is our ability to recognize patterns.

There's been a sea change just in the last several years in the public understanding of the acceleration of change and the potential impact of all of these technologies - computer technology, communications, biological technology - on human society. There's really been tremendous change in popular public perception in the past three years because of the onslaught of stories and news developments that document and support this vision. There are now several stories every day that are significant developments and that show the escalating power of these technologies.

[Agnes Heep]

Most human evolution occurs during war. If history is any guide I don't think we've seen the last one. WWII must have had some evolutionary effects. For example 2 out of 3 Jewish people in Europe were killed. One a group level, were any traits indirectly selected for?

I think your casualty counts would have to be quite high and they would have to be limited to individuals meeting a certain criteria, after which your populations would have to be almost completely isolated from the rest of the world. Besides, in evolutionary terms among higher order creatures, I like to think more in terms of the social success of the fittest as opposed to their mere survival. I think evolution is much more complicated than the model we usually perceive, of one trait being selected at a time. I think there are lots of traits being selected for, and the totality of favorable traits leads to preeminence in a social heirarchy that leads in turn to more prolific breeding amongst the fittest.

[DarthMaulrulesok]

Add this to the list of REALLY BAD IDEAS. What, exactly,would keep a supercomputer ("collosus","HAL" or whatever you want to call it) from deciding that some humans (or all humans) were unnecessary? It's probable, someday, that we will indeed be able to create a human-level AI, but how will we teach it to value human life or the rights of an individual? How can we teach it to believe in the Sermon on the Mount or the 10 commandments? Worse yet, what if it's programmed to obey the Koran?

It would be the ultimate Frankenstein monster, a being with no sense of morality yet with (perhaps) power beyond our ability to stop.

Human sociopaths are bad enough as it is. We don't need to artificially create more.

[RadioAstronomer]

This is way outside of my field and why I pinged tortoise. He is an expert in this field. However, (tortoise correct me if I am way off base in this post) as Moore's Law predicts, processors will double in transistor count about every two years. This has allowed for huge strides in the computer industry.

Compare a current low-end personal computer with a "super computer" of the 1970's. Add to this scalable clusters and the computing power really goes up. Now think downstream 20 years. Where will computers be then? I know chip designers and leading experts in the computer field who won’t even hazard a guess what they will look like it 20 years.

AI might happen, but the question is would we recognize it if it did. In other words could we create an AI without even realizing it?

[tortoise]

Their solution will probably turn out to be Hillaryism

Eliezer Yudkowsky is well-documented as being a "small-L" libertarian. Socialism of any sort and taxes in general are abhorrent to him. Seeing as how he despises socialism and taxes, I don't see how "Hillaryism" would be anything he would espouse.

The solution being promoted is highly technological free market capitalism.

[tortoise]

The Turing test will soon be passed, probably within the next year or two. The machine that does this will not be sentient by any means.

Common misconception. Turing himself do not put any great weight on this test. He merely stated that a sentient machine might be able to pass that test, but it is not a requirement of sentience to pass that test.

As many others have pointed out, the Turing Test only tests "human-ness", but since humans are routinely stupid and irrational (due in no small part to our biology), a very high-grade intelligence would have no reason to express or emulate these lousy characteristics that often define human behavior. Machine intelligence should be quite alien precisely because it is "cleaner" and purer in many ways.

[Lazamataz]

[tortoise]

Compare a current low-end personal computer with a "super computer" of the 1970's. Add to this scalable clusters and the computing power really goes up.

One of the open secrets of current AI research, which is advancing very fast thanks to a cascade of fundamental math breakthroughs in the last few years, is that the limitation of current computers is not CPU performance but memory performance. Memory latency and size suck big time for the kind of data structures involved. At this point in time, nobody really thinks the CPU is inadequate; we can use what we have and we don't really need more in a significant way. As for memory: 1-10 Terabyte main memories with <50 nanosecond latency would remove a lot of barriers, and not having anything close to this is widely recognized as the major hardware limitation. Note that this means that loosely coupled clusters are a sub-optimal solution, though some new tightly coupled interconnects are showing promise (e.g. Octiga Bay).

AI might happen, but the question is would we recognize it if it did. In other words could we create an AI without even realizing it?

That boils down to the question of what is AI precisely?