How close is AI to simulating all the incredible things that happen in the brain? Answer: not even to the first step. Not even in principle.
From Why the Brain Is Not At All Like a Computer:
Comment:
It appears that the human brain has approximately 10**12 neurons, where averaged out each neuron has approximately 10**5 synapses. Using the "brain is a computer" paradigm each synaptical connection is apparently now thought to actually be equivalent to a microprocessor in complexity. So, the number of equivalent microprocessors in the brain is approximately 10**17, approaching a billion billion.
As of 2017, the largest transistor count in a commercially available single-chip microprocessor was 19.2 billion, (1.9x10**10). As of 2018 the physically smallest microprocessor (smaller than a grain of sand) had 100,000 (10**5) transistors.
Therefore the number of equivalent logic circuit transistors in the brain is somewhere between 10**22 and 1.9x10**27. Between ten thousand billion billion and 1.9 billion billion billion.
Good luck to the researchers who are trying to build artificial simulations of the brain completely emulating the complexity of its logic. This would mean down to the basic digital logic unit level. Obviously just building it to the synapse level (assuming as they used to that each synapse is a single logic gate) would be totally inadequate. Simulating the entire complexity of the brain is totally and completely impractical for any foreseeable technology.
Of course, there are other even stronger reasons why such enterprises are futile, having to do with the fundamental nonequivalency of the brain to computer processors. For instance, as pointed out by the expert interviewee, as soon as you assume that each neuron is a microprocessor, you assume that somehow it was programmed. And programs are fundamentally algorithms. But there is no programmer. Accordingly there are no algorithms (as we understand them) in the brain. Whatever processing there is, is unknown. But the dilemma is then that even non-algorithmic programs also require a programmer. Very basically, we just don't have any understanding of how the brain really works.
Then there is also the problem of the origin of such masses of very complex specified information, with the only candidate of materialism being the blind purposeless semi-random Darwinian processes.
(This post was last modified: 2019-09-22, 07:01 PM by nbtruthman.)
From Why the Brain Is Not At All Like a Computer:
Quote:"Yuri Danilov: Again, it is a separate discussion, extremely painful for many but it is something that is happening right now. Remember, I talked today about our technological development morphing how our understanding of the brain works. And the attempt to make a parallel between the brain and a computer is a result of our evolution, if you wish. Because… in the Seventies … it was a transistor and everybody thought it was very simple. They thought that each neuron is a transistor.
Robert J. Marks: Yes.
Yuri Danilov: Then it was, “Each neuron is a microchip.”
Robert J. Marks: Yes.
Yuri Danilov: Then each neuron is a microprocessor.
Robert J. Marks: Yes.
Yuri Danilov: Right now people are saying, each synaptical connection is a microprocessor. So if it’s a microprocessor, you have 1012 neurons, each neuron has 105 synapses, so you have … you can compute how many parallel processing units you have in the brain if each synapse is a microprocessor.
But as soon as you assume that each neuron is a microprocessor, you assume that there is a programmer. There is no programmer in the brain; there are no algorithms in the brain."
Comment:
It appears that the human brain has approximately 10**12 neurons, where averaged out each neuron has approximately 10**5 synapses. Using the "brain is a computer" paradigm each synaptical connection is apparently now thought to actually be equivalent to a microprocessor in complexity. So, the number of equivalent microprocessors in the brain is approximately 10**17, approaching a billion billion.
As of 2017, the largest transistor count in a commercially available single-chip microprocessor was 19.2 billion, (1.9x10**10). As of 2018 the physically smallest microprocessor (smaller than a grain of sand) had 100,000 (10**5) transistors.
Therefore the number of equivalent logic circuit transistors in the brain is somewhere between 10**22 and 1.9x10**27. Between ten thousand billion billion and 1.9 billion billion billion.
Good luck to the researchers who are trying to build artificial simulations of the brain completely emulating the complexity of its logic. This would mean down to the basic digital logic unit level. Obviously just building it to the synapse level (assuming as they used to that each synapse is a single logic gate) would be totally inadequate. Simulating the entire complexity of the brain is totally and completely impractical for any foreseeable technology.
Of course, there are other even stronger reasons why such enterprises are futile, having to do with the fundamental nonequivalency of the brain to computer processors. For instance, as pointed out by the expert interviewee, as soon as you assume that each neuron is a microprocessor, you assume that somehow it was programmed. And programs are fundamentally algorithms. But there is no programmer. Accordingly there are no algorithms (as we understand them) in the brain. Whatever processing there is, is unknown. But the dilemma is then that even non-algorithmic programs also require a programmer. Very basically, we just don't have any understanding of how the brain really works.
Then there is also the problem of the origin of such masses of very complex specified information, with the only candidate of materialism being the blind purposeless semi-random Darwinian processes.
![[-]](https://psiencequest.net/forums/images/collapse.png)