Code Bytes: Essentially, this algorithm knows ABSOLUTELY NOTHING about the “software” of the human brain.It knows NOTHING of the complex neurological thought patterns, memories, creativity instincts, and logical reasoning of the human mind! So in essence… it’s basically an algorithm to simulate something much much less intelligent than a fetus brain! AGAIN: All it can do is simulate the data transfer between neurons (so they claim) if we ever build such a computer.
Even with massive supercomputers, it’s next to impossible to simulate 100% of the brain. With an earlier version of the algorithm, the researchers were able to reproduce only 1% on their petascale K supercomputer.
The reason is that the memory required per processor to simulate just 1% of the human brain is very high. If the entire brain comes in the picture, the memory requirements jump to almost 100 times per processor the current supercomputers have.
In the future, with exascale supercomputers (having more processors per node), it would be possible to scale the NEST algorithm to achieve faster whole-brain simulation. Then also, the memory per processor and the number of nodes would stay the same. But the advanced NEST algorithm would be able to optimize the memory required by the system.
I bet this is one of the very first things that the soon to be exa-scale supercomputers will do. But we better be careful. If we are able to tease out the nature of consciousness, which I suspect is an emergent phenomenon, we could inadvertently bring about an EI (emergent intelligence) We do not want an EI sharing the Earth with us.
Having said that, I think there is one other major consideration and that is that consciousness to include humanconsciousness may have a need for a quantum mechanics substrate. Roger Penrose, probably the smartest scientist on Earth since Hawking’s demise, has suggested this.
But of course we are working on quantum computers pretty diligently as well so in about 5 years time I bet we have the complete infrastructure to successfully develop an AI that could be more “versatile” than the narrow AI we have at this time. Could we call it AGI. Maybe not yet. But we would be on the path.
So will we learn to use the quantum computer to “enhance” the performance of classical computers and then mix in our AI R&D? Probably.
5, 10, even 20 years is a vanishingly small period of time in the history of human affairs and is why it is likely that Kurzweil is correct that the capability to bring about the “technological singularity” is well within view now.
Also don’t forget that our computer processing power is still doubling nearly every 2 years. That processing power plus the capacity for “big data” is rocket fuel for the development of all forms of AI to include the successful simulation of AGI, if not the real thing.