On Wed, May 30, 2012 at 3:04 PM, Craig Weinberg <whatsons...@gmail.com>wrote:
> On May 29, 1:45 am, Jason Resch <jasonre...@gmail.com> wrote: > > > So which of the following four link(s) in the logical chain do you take > > issue with? > > > > A. human brain (and body) comprises matter and energy > > So does a cadaver's brain and body. The fact that a cadaver is not > intelligent should show us that the difference between life and death > can't be meaningfully reduced to matter and energy. > > That some organizations of matter/energy are intelligent and others are not is irrelevant, what matters is whether or not you agree that the brain is made of matter and energy. Do you agree the brain is made of matter and energy, and that the brain is responsible for your consciousness (or at least one of the many possible manifestations of it)? > > B. that matter and energy follow natural laws, > > No, laws follow from our observation of natural matter and energy. > You are mistaking our approximations and inferences concerning the natural laws for the natural laws themselves. Before there were any humans, or any life, there must have been laws that the universe obeyed to reach the point where Earth formed and life could develop. Do you agree that such natural laws exist (regardless of our human approximations of them)? > > > C. that these laws are describable in mathematical terms > > You have jumped from physics to abstraction. It's like saying 'I have > a rabbit > rabbits act like rabbits > Bugs Bunny is modeled after the > behavior of rabbits > Bugs Bunny is a rabbit'. > I haven't jumped there yet. All "C" says is that there exists some formal system that is capable of describing the natural laws as they are. You may accept or reject this. If you reject this, simply say so and provide some justification if you have one. Note that I have not made any statement to the effect that "an abstract rabbit is the same as a physical rabbit", only that natural laws that the matter and energy in (a rabbit or any other physical thing) follow can be described. > > > D. that mathematics can be simulated to any degree of precision by > > algorithms > > > > Precision only determines the probability that a particular detector > fails to detect the fraud of simulation over time. It says nothing > about the genuine equivalence of the simulation and the reality. > > It sounds like you accept that mathematics can be simulated to any degree of precision by algorithms, but your objection is that without absolutely perfect precision, the simulation will eventually diverge from the object being simulated in some noticeable way. I think this is a valid objection. However, I don't see this objection serving as the basis for Colin's argument against artificial general intelligence. Let's say we have a near perfect simulation of the physics of Einstein's brain running in a computer. It is near-perfect, rather than perfect, because due to rounding errors, it is predicted that there will be one neuron misfire every 50 years of operation. (Where a misfire is a neuron that fires when the actual brain would not have, or doesn't fire when the actual brain would not have). Maybe this misfire causes the simulated brain to develop a wrong idea when he would have otherwise had the right one, but who would argue that this simulated Einstein brain is not intelligent? Perhaps it has an IQ of 159 instead of the 160 of the genuine brain, but it would still be consider an example of AGI. If you don't like the 1 error every 50 years, then you can double the amount of memory used in the floating point numbers (going from 64 bits to 128 bits per number), and then you make the system have a precision that is 2^64 times finer, so there would not be a deviation in the simulation during the whole life of the universe. So while I accept your argument that a digital machine cannot perfectly simulate a continuous one perfectly, I do not see how that could serve as a practical barrier in the creation of AGI. Jason > -- You received this message because you are subscribed to the Google Groups "Everything List" group. To post to this group, send email to email@example.com. To unsubscribe from this group, send email to everything-list+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/everything-list?hl=en.