Recently I started reading the book "The Singlularity Is Near" (When humans transcend biology) by Ray Kurzweil. While doing some googling on reverse brain engineering which he talks about I came across the site longbets.org.

Mitchell Kapor, founder of Lotus Development Corporation and the designer of Lotus 1-2-3 issued a challenge at longbets saying that: By 2029 no computer - or "machine intelligence" - will have passed the Turing Test. If you have read much of Kurzweil's writings or books you won't be supprised to see that he has taken Kapor up on the challenge, and disagrees with him.

At stake is $20,000 to their named charities. The test is to be administered by the Long Now Foundation under conditions agreed to by Kurzweil and Kapor.

Because most people intuitively tend to view technological growth in the linear rather than exponential, I was Supprised that voting is running slightly in favor of Ray Kurzweil (presently 177-174). Follow the link above and read the arguements for both sides and leave some comments on your views. There are also many other interesting longshot bets there about future technology that you might find interesting.

Mitchell Kapor's finds it difficult to imagine a computer that can "perform a successful impersonation" of a human (required to pass the Turning Test). His belief is that because we are conscious beings, capable of reflection. That we have cognition and emotion.

Ray Kurzweil concedes that emotional response is the hardest challenge but contends that a "Turing-capable" machine will be possible by reverse engineering the human brain. Or in other words if it is a capability of the human brain, it is a complexity we can master.

Well I hate to be this way but I am going to wuss out on taking sides yet, although my guess is that Dean is solidly behind Kurzweil. It is partially because I haven't finished Kurzweil's book, and partially because I think quallities like cognition, reflection, introspection, emotions like love and fear may be beyond the results of reverse engineering the brain, but that doesn't mean a computer (with enough speed, power and the wealth of most human information) cannot emulate them well enough to fool most humans.

Sure emotions are chemically based and with out a doubt bio-chemicals cause feelings. But I think our consciousness is more than just chemical-neuron reactions. I'm not confident that reverse engineering will allow us to look into whatever the essence of our (maybe hidden) sentient core that gives us conscious perceptions and emotions is.

Also computers process data bit by bit. Humans process data holistically or as Kurzweil says digitally controlled analog "transactions." There is the "Informality of Behaviour" arguement which states that any system governed by laws will be predictable and therefore not truly intelligent. Turing replies by stating that this is confusing laws of behaviour with general rules of conduct. Maybe so as computers are already capable of originality producing music that sounds like Bach.

If we have a pleasant thought chemicals cause us to physically feel good. The physical "feel good" comes from chemicals, but as a result of the thought, not the cause of the thought. But all that aside I think the Turing Test may be passed by a computer by 2029. The unimaginable power and speed of future computing alone may be enough without consciousness to fool the testers. Of course the computer will have to be programed to give false information so as to not divulge that it is a machine.

I could use one.

James Lovelock thinks that most of the human race going to be killed off by global warming, leaving only a few breeding pair to survive in the Arctic.


Forget it doc, we will take an asprin and call you in the next century.

What most people don't realize is that our technology is growing at an expotential rate (and then some) and we will have the energy and global warming issued solved long before this century is over.

Lovelock's original Independent Online article.

You can stop looking for the fountain of youth. There may be a gene inside of us that prolongs our lifespan.

Computer technology in the future will move forward in such a way that will make todays desktop PCs by comparison look inferior to the TRS-80 of the 70s. Behind this leap is quantum technology. Google up "quantum dots," "quantum wells," or quantum computing" but be prepared to do some deep technical reading. Quantum gates are measured on the nanoscale. A nanometer being .000001 of a millimeter, or about 50 atoms thick.

A quantum dot (or well) confines its electrons in all three dimensions causing them to act as though they were part of an atom. What is really wild about this artificial atom is that by affecting it with an external charge it can be made to resemble any atom on the periodic table (and I think some that are not), but that is a subject for another post. It was just too wild a scientific claim not to mention it here.

Nano size isn't the only thing that will make this sort of computing faster. In understanding quantum computing you will have to learn about something called the qbit (not cubit). Unlike binary computing where we can only have two states represented as a 0 or a 1, a qbit can be either a 0, a 1 or both. And while a qbit can only decohere into one of two states (0 or 1), 5 qbits together can colapse into 32 different states. But before the colapse these 5 qbits can literally store and perform computations on all 32 states. That would take a binary computer 32 sets of 5, or 160 total bits.

Anyway the first quantum microchip has been created by University of Michigan scientists.


Each time a qbit is added the power doubles. Remember the classic penney a day wage doubled every day for a month? Yeah over $10,000,000 after working a 31 day month. Well that puts more into perspective the power of just adding 32qbits of computing power.

Here is another example from "Hacking Matter" (downloadable) by Wil McCarthy.


Don't run out next week or even next year looking for a quantum computer, but the fact that UM scientists have created a quantum chip moves us from theory to reality. It will be many years before we see a quantum desktop PC. It is more realistic to believe that when quantum technology does come to the PC it will be a silicon binary - quantum hybrid.

To learn more about quantum components and programable atoms see the Hacking Matter link above. Computing is only one of the many feasable applications for this seemingly magical new technology on the nano scale.

Wow. This is awesome.


Via KurzweilAI.net