[ExI] Oxford scientists edge toward quantum PC with 10b qubits.

[Richard Loosemore]

Adrian Tymes wrote:
> 2011/1/28 John Clark <jonkc at bellsouth.net>:
>> On Jan 28, 2011, at 11:01 AM, Richard Loosemore wrote:
>>> do you have any idea how trivial this is?
>> No.
> 
> Richard's already stated it well.  I have some expertise in
> the field, and I agree with what he has said here.
> 
> That said, getting back to the original point: better quantum
> computers make certain theories of how human-level AI
> might work, easier to test.  If one of those theories turns out
> to be The One, then yes, we could be a bit closer because
> of this.  If not, we're not.  It's impossible to say for sure yet.
> 
> Even if not, a single computer with 10 billion qubits is
> potentially commercially usable, for the things we know
> quantum computers are useful for...if someone
> commercializes this discovery.  ("Exists in lab" is one
> thing, but "is available to, and practical for use by, people
> who would benefit from using it" is quite another.)

I'll concur with this, and I should add that I don't want to overstate 
my case.

Faster hardware is always useful for doing the research itself.  In act, 
because my own approach to AGI requires the execution of very large 
numbers of simulation experiments, the more horsepower, the better.  So 
it could speed up my attempts to understand AGI theory.

I guess what I am fighting is the idea (which is very common, it seems 
to me) that lack of horsepower is what is holding back AGI.



Richard Loosemore