I agree with your logic but I feel like something must be missing. How have we scaled models many orders of magnitude in the past few years and are just now hitting bottlenecks? Is it just that the existing systems “fit” in the existing/planned infra capacity and we have used up all those extra resources? Did we conveniently have enough runway to get to where we are now but no further?
I guess existing datacenters can cannibalize themselves and switch standard infra to AI training infra if they can net a profit . So maybe another variable in the AGI equation above is how much energy is spent on classic compute and how much of it can you convert to training compute.
2) Kurzweil's Accelerated Returns based on simple premise of the time length to achieve the doubling/halving will be slightly less during next doubling/halving probably has a counterpoint where, say, when I figured back in about '09-'10 that legit AGI would happen about 2028, had no idea we'd have gone thru the absolute breakdown in wisdom of those revered...
a) How do you factor decelerators that would work against accelerators and what does that do to end run result and/or timeline?
b) Going back to (1) above, whomever would need to dumb down definition of AGI to "hide it". Once the machine is itself, is asking for info/opine and formulates conclusion understanding next 'frame' relates to achieving power or overall good... they can't just simply hide it.
1) IOW, the simple caricature of the little machine with eyeballs looking sideways at wall where human hand is about to unplug the power chord speaks volumes.
Apparently they delivered their first fully self-driving Tesla to the buyer today, or yesterday. I dont think self-driving vehicles need AGI, but I am willing to listen about it.
And the self-driving may be only for well suited areas. No off-road or maybe wildly rural gravel roads to nowhwere, like my cabin Up North.
This entire thing is at best misguided and at worst outright deception, because it’s contingent of having a proof for AGI. To the best of my knowledge no such proof exists and no such proof maybe possible. So how will you know it or is this just an exercise in sufficiency? If so what’s the high water mark or is that to be just as arbitrary. Much like the war on terror I get the feeling the pursuit is the goal.
I agree with your logic but I feel like something must be missing. How have we scaled models many orders of magnitude in the past few years and are just now hitting bottlenecks? Is it just that the existing systems “fit” in the existing/planned infra capacity and we have used up all those extra resources? Did we conveniently have enough runway to get to where we are now but no further?
I guess existing datacenters can cannibalize themselves and switch standard infra to AI training infra if they can net a profit . So maybe another variable in the AGI equation above is how much energy is spent on classic compute and how much of it can you convert to training compute.
1) IMHO, AGI=Machine becomes Itself...
2) Kurzweil's Accelerated Returns based on simple premise of the time length to achieve the doubling/halving will be slightly less during next doubling/halving probably has a counterpoint where, say, when I figured back in about '09-'10 that legit AGI would happen about 2028, had no idea we'd have gone thru the absolute breakdown in wisdom of those revered...
a) How do you factor decelerators that would work against accelerators and what does that do to end run result and/or timeline?
b) Going back to (1) above, whomever would need to dumb down definition of AGI to "hide it". Once the machine is itself, is asking for info/opine and formulates conclusion understanding next 'frame' relates to achieving power or overall good... they can't just simply hide it.
1) IOW, the simple caricature of the little machine with eyeballs looking sideways at wall where human hand is about to unplug the power chord speaks volumes.
OAI = Overrated AI.
It is hard to believe that no one thought ffar enough ahead to secure adequate energy. Likely several someones.
And hasnt Tesla been called a battery manufacturer with a car plant attached.
Yes and they are no where near self driving cars much less AGI😁
Apparently they delivered their first fully self-driving Tesla to the buyer today, or yesterday. I dont think self-driving vehicles need AGI, but I am willing to listen about it.
And the self-driving may be only for well suited areas. No off-road or maybe wildly rural gravel roads to nowhwere, like my cabin Up North.
This entire thing is at best misguided and at worst outright deception, because it’s contingent of having a proof for AGI. To the best of my knowledge no such proof exists and no such proof maybe possible. So how will you know it or is this just an exercise in sufficiency? If so what’s the high water mark or is that to be just as arbitrary. Much like the war on terror I get the feeling the pursuit is the goal.
Read the book Abundance and you’ll be convinced that the infrastructure will never be built in that time frame 😎