00:00:00
I think fundamentally if you're driving
00:00:02
productivity with AI, you're driving
00:00:06
leverage on human time and leverage on
00:00:09
capital. The question is how quickly can
00:00:11
you drive that up? And that's a function
00:00:13
of how much consumption there is, how
00:00:15
much capacity there is for consumption.
00:00:17
So if your earnings are the same, but
00:00:18
things are getting more expensive,
00:00:20
you're not happy. If your earnings go up
00:00:22
by 10%, things stay the same price, you
00:00:25
got 10% more than you had last year,
00:00:26
you're you're going to be happy. I just
00:00:27
think like all humans are driven by this
00:00:29
need to consume more each year than they
00:00:31
did last year. So I think for me that's
00:00:34
like the lower limit on consumptive
00:00:38
capacity in the world. The question that
00:00:40
we're now facing which we've never faced
00:00:42
in human history before is there a upper
00:00:46
limit
00:00:47
>> on consumptive capacity because AI
00:00:50
creates such a profound shift in
00:00:52
productivity and in leverage that
00:00:56
normally you would say hey when we get a
00:00:57
new tool or we get new leverage in a
00:00:59
system we build a new technology we can
00:01:01
make more with less. Therefore, everyone
00:01:03
gets access to more things for the same
00:01:05
price or the cost of things that they
00:01:07
consume come down by a certain price.
00:01:10
But there may be a situation now where
00:01:13
the ability to make stuff exceeds the
00:01:16
capacity to consume stuff. And that is
00:01:19
something that I don't think we've faced
00:01:21
before. And I think that's sort of where
00:01:23
a lot of the models start to break. Just
00:01:24
general economic models, just general
00:01:26
productivity models, and general social
00:01:29
models. And this goes to the point about
00:01:31
like what is everyone going to do? In
00:01:33
the same way that I think we've argued
00:01:35
that maybe SAS was a transitory business
00:01:38
phenomenon that existed between the
00:01:41
foundation of the internet and the era
00:01:43
of AI. It may be the case that knowledge
00:01:46
work in general is also a transitory
00:01:49
phenomenon that only existed between the
00:01:52
foundation of the computer or computing
00:01:54
tools and the existence of AI generally
00:01:57
speaking. And if all of that goes away
00:01:59
very quickly and all of those people can
00:02:02
be redistributed and recast into doing
00:02:04
other higher level more creative things,
00:02:06
their productivity goes up by 100x.
00:02:09
Is there really a consumer on the other
00:02:11
end of all of that productivity? Is
00:02:13
there really enough consumptive
00:02:14
capacity? And I think that's the
00:02:16
profound question that we all
