00:00:00
I have a little surprise for my besties
00:00:02
everybody's been getting incredible
00:00:04
adulation people have been getting
00:00:06
incredible feedback on the podcast it is
00:00:09
a phenomenon As We Know
00:00:11
and I thought it was time for us to do
00:00:14
performance reviews of each bestie now I
00:00:17
as executive producer I'm not in a
00:00:20
position to do uh your performance
00:00:22
reviews I too need to have a performance
00:00:23
review so I thought I would let the
00:00:25
audience do their performance reviews so
00:00:27
we went and we're debuting a new feature
00:00:29
today at this very moment it's called
00:00:31
Reddit performance review oh God cue
00:00:34
some music here some graphics Reddit
00:00:37
performance reviews and so we'll start
00:00:39
off with you David Friedberg this proves
00:00:42
why you haven't been successful in life
00:00:43
so far
00:00:48
Enterprise in
00:00:50
set to yourself which is what all Elite
00:00:52
performers do yes you turned it over to
00:00:55
a bunch of mids on Reddit yes I'm
00:00:58
already doing it they were already doing
00:00:59
it we just collected it what elucidation
00:01:01
were you going to get from Reddit yeah
00:01:03
really let's not ruin the bit do you
00:01:04
think Elon does 360 performance reviews
00:01:07
on Reddit on Reddit of all things you
00:01:10
know how many 360 performance trees I've
00:01:12
done in my life zero of course and
00:01:14
that's why this will be so interesting
00:01:16
okay go go start off with me
00:01:19
you are going to be presented with the
00:01:22
candid feedback that you've gotten in
00:01:23
the last 30 days on Reddit
00:01:25
but for the first time and you have to
00:01:28
read it out loud to the class Freeburg
00:01:30
you'll go first here's your first piece
00:01:33
of candid feedback in your 360 from the
00:01:35
Reddit in cells go ahead Freeburg read
00:01:37
it out loud David Friedberg deserves
00:01:40
more hate he and the others have made it
00:01:42
their mission to convince us that
00:01:44
reforming Social Security is the only
00:01:46
way forward to survive he hides behind
00:01:48
this nerdy apolitical Persona and then
00:01:51
goes hard right out of nowhere it's that
00:01:55
fear-mongered about the deficit as an
00:01:57
excuse to restructure entitlement
00:01:59
programs we would see that as the
00:02:01
partisan right-wing take it is when
00:02:03
Friedberg does it we're supposed to act
00:02:05
like he has no skin in this game he's
00:02:07
just the Science Guy no he's a rich guy
00:02:11
who would rather work your grandparents
00:02:13
to death than pay an extra five percent
00:02:14
tax all right there's your review very
00:02:16
good I think you took it well and you
00:02:19
don't have to respond now don't be
00:02:20
defensive just take it in just a counter
00:02:22
briefly I have highlighted multiple
00:02:24
times I think we're going to 70 tax
00:02:26
rates but hey you know all right they're
00:02:29
wrong everyone's got an opinion that
00:02:30
won't generate more Revenue yeah the
00:02:32
history is any guide yeah well the
00:02:34
audience has been waiting for this David
00:02:35
sacks has never taken a piece of
00:02:37
feedback and the feedback he has gotten
00:02:39
he hasn't taken well so here we go David
00:02:41
here's your performance review go ahead
00:02:43
I'm supposed to read this go with the
00:02:45
bit come on it's a little feedback for
00:02:46
you come on all right and I'm gonna do
00:02:48
this too and I haven't seen mine
00:02:50
oh Nick pulled these Nick pulled these
00:02:52
this is nice
00:02:54
these are actual pieces of real feedback
00:02:57
Nick you're fired that's it get out of
00:03:00
here
00:03:01
good sex
00:03:02
the thing about David sacks if he wasn't
00:03:04
Rich everyone would dismiss him as being
00:03:06
both stupid and boring and the secret to
00:03:09
his wealth is just follow theater since
00:03:10
College it's like if turtle from ontar
00:03:12
pretend to be a public intellectual
00:03:20
has never had a 360 review he informs us
00:03:23
and I think his staff you know for all
00:03:25
the people at Social Capital you can get
00:03:27
in on this by just posting to Reddit
00:03:28
since he went to 360 reviews at Social
00:03:30
Capital go ahead let's pull up your
00:03:31
months feedback for the quarter
00:03:34
moth is the biggest self-serving leech
00:03:36
as long as he can make a dollar trade on
00:03:40
it he will burn anything down to the
00:03:42
ground [ __ ] the consequences to society
00:03:44
or anyone else this was actually the
00:03:46
good part
00:03:48
yeah why'd you give him such a good
00:03:50
review
00:03:52
truth it is
00:03:55
okay I gotta read mine I haven't seen
00:03:57
this I'm embracing for impact here
00:03:58
what's the critique exactly what is the
00:04:02
critique
00:04:03
from that I mean that's like it's pretty
00:04:05
active yeah okay all right here we go
00:04:07
I gotta read mine okay I can't wait for
00:04:09
AI to replace Jay cow Jay cow is the
00:04:12
least skilled knowledge worker on the
00:04:14
show I think he has about three shows
00:04:16
left before AI replaces his hosting
00:04:18
skills and ability to trick dentists
00:04:20
into investing in hype
00:04:26
that's pretty good that last part
00:04:30
I do have a lot of dentist friends in
00:04:32
the funds um okay I'm gonna pay for a
00:04:34
lot of kids then to school great idea
00:04:35
great idea Jay Cal to give us what a
00:04:38
great big there's one group one there's
00:04:39
one for the whole group this is a group
00:04:41
survey this is our group 360. vote to
00:04:44
rename the podcast the vote is binding
00:04:47
and the podcast will be renamed three
00:04:48
billionaires in jaycal three kind of
00:04:51
smart guys and jcal three [ __ ] and
00:04:53
Free Bird that's a pretty good survey
00:04:56
that was pretty good yeah that's a good
00:04:57
one
00:05:01
Rain Man
00:05:02
[Music]
00:05:06
we open source it to the fans and
00:05:09
they've just gone crazy
00:05:10
[Music]
00:05:13
all right everybody Welcome to the all
00:05:16
in podcast we're still here episode 129
00:05:20
all in Summit 2023 general admission
00:05:22
sold out
00:05:23
too many people apply for scholarships
00:05:25
that's on pause and there's a couple of
00:05:27
VIP tickets left get them while they're
00:05:29
hot just search for all in Summit
00:05:30
Freeburg anything to add we'll just get
00:05:32
through the grift real quick here at the
00:05:33
beginning no it's gonna get it all right
00:05:35
that was it yeah I mean just more demand
00:05:37
than we predicted we look for a bigger
00:05:39
venue couldn't find one we're I think
00:05:41
we're excited about Royce Hall
00:05:43
it's still as you pointed out two and a
00:05:45
half times the size of last year so we
00:05:46
want to make sure it's a great quality
00:05:48
event but unfortunately way too many
00:05:50
folks want to go so we have to kind of
00:05:52
pause ticket sales what's my wine budget
00:05:55
300 per person per night a thousand
00:05:58
dollars per VIP per event thank you okay
00:06:01
I will handle it from here so there's
00:06:03
750 of them so I think you have 750 000
00:06:06
in one budget thank you
00:06:09
I mean I just can't believe I just gave
00:06:11
extra mod 750 dollars to buy one all
00:06:15
right guys I'm gonna curate an
00:06:17
incredible guideline here today by the
00:06:19
way
00:06:19
oh he's doing
00:06:21
yes Melanie Josh somebody hey Josh all
00:06:24
right let's get to work okay the senate
00:06:27
had a hearing this week for AI Sam
00:06:31
Altman was there as well as Gary Marcus
00:06:33
a professor from NYU
00:06:36
that's an overpriced College in New York
00:06:38
City and Christina Montgomery the Chief
00:06:41
privacy and Trust officer from IBM which
00:06:44
had Watson before anybody else was in
00:06:46
the AI business and I think they
00:06:47
deprecated it or they stopped working on
00:06:49
it which is quite paradoxical
00:06:52
there were a couple of very interesting
00:06:53
moments
00:06:55
Sam claimed the U.S should create a
00:06:56
separate agency to oversee AI I guess is
00:06:59
in the chamoth camp he wants the agency
00:07:02
to issue licenses to train and use AI
00:07:05
models a little regulatory capture there
00:07:07
as we say in the biz he also claims and
00:07:10
this was interesting dovetailing with
00:07:12
elon's CNBC interview with I think Dave
00:07:15
Farber which is very good that he owns
00:07:18
no equity
00:07:20
in open AI whatsoever and was quote
00:07:23
doing it because he loves it
00:07:26
any thoughts shamath you did say that
00:07:29
this would happen
00:07:31
two months ago and here we are two
00:07:32
months later and exactly what you said
00:07:34
would happen is in the process of
00:07:36
happening regulation Licensing and
00:07:39
Regulatory capture Sam went a little
00:07:41
further
00:07:42
then I
00:07:45
sketched out a few months ago which is
00:07:48
that he also said that it may make sense
00:07:50
for us to issue licenses for these
00:07:53
models to even be compiled
00:07:55
um
00:07:57
and for these models to actually do the
00:07:59
learning
00:08:01
and I thought that that was really
00:08:02
interesting because what it speaks to is
00:08:04
a form of kyc right know your customer
00:08:06
and again when you look at markets that
00:08:10
can be subject to things like fraud and
00:08:12
manipulation right where you can have a
00:08:14
lot of Bad actors banking is the most
00:08:16
obvious one
00:08:17
we use things like kyc to make sure that
00:08:19
money flows are happening appropriately
00:08:21
and between parties that where the
00:08:23
intention is legal
00:08:26
and so I think that that's actually
00:08:29
probably the most important new bit of
00:08:32
perspective that he is adding as
00:08:34
somebody right in the middle of it which
00:08:36
is that
00:08:37
you should apply to this agency to get a
00:08:40
license to then allow you to compile a
00:08:42
model
00:08:43
and I think that that was a really
00:08:44
interesting thing the other thing that I
00:08:47
said and I said this in my in a tweet
00:08:49
just a couple days ago is I'm really
00:08:51
surprised actually where this is the
00:08:53
first time in modern history that I can
00:08:55
remember
00:08:56
where we've invented something we being
00:08:58
Silicon Valley
00:09:00
and the people in Silicon Valley are the
00:09:02
ones that are more circumspect than the
00:09:04
folks on Wall Street or other areas and
00:09:07
if you see if you gauge the sentiment
00:09:10
the hedge funds and family offices right
00:09:12
now are just giddy about Ai and it turns
00:09:16
out if you look at their 13fs they're
00:09:17
all long Nvidia and AMD
00:09:20
but if you actually look at the other
00:09:22
side of the coin which is the folks in
00:09:23
Silicon Valley that's actually making it
00:09:25
the rest of us are like hey let's crawl
00:09:27
before we walk before we run yeah let's
00:09:29
think about guard rails let's be
00:09:31
thoughtful here and so the the big money
00:09:33
people are saying let's place bets and
00:09:35
the people building in are saying hey
00:09:37
let's be thoughtful which is opposite to
00:09:40
what it's always been I think right
00:09:41
we're like hey let's let's run with this
00:09:43
and wall Street's like prove it to me
00:09:46
sax you are a less regulation guy you
00:09:49
are a free market monster I've heard
00:09:51
you've been called uh you don't believe
00:09:53
that we should license this what do you
00:09:55
think about what you're seeing here and
00:09:57
there is some cynical cynical thoughts
00:10:01
about what we just saw happen in terms
00:10:03
of people in the lead wanting to
00:10:05
maintain their lead by creating red tape
00:10:09
what are your thoughts
00:10:10
yeah of course I think you know Sam just
00:10:12
went straight for the end game here
00:10:14
which is regulatory capture
00:10:16
normally when a tech executive goes and
00:10:17
testifies at these hearings they're in
00:10:19
the hot seat and they get grilled and
00:10:21
that didn't happen here because you know
00:10:23
Sam Allman basically bought Into The
00:10:25
Narrative of these senators and he
00:10:28
basically conceded all these risks
00:10:30
associated with with AI he talked about
00:10:33
how
00:10:34
GPT style models if unregulated could
00:10:38
increase online misinformation bolster
00:10:39
cyber criminals even threatened
00:10:41
confidence in election systems
00:10:45
so he basically bought into the center's
00:10:48
narrative and like you said agreed to
00:10:50
create a new agency that would license
00:10:52
models and can take licenses away
00:10:55
he said that he would create safety
00:10:57
standards specific tests that a mall has
00:10:59
to pass before it can be deployed he
00:11:02
says he would require independent audits
00:11:04
who can say the model is or isn't in
00:11:06
compliance
00:11:08
and by basically buying into their
00:11:10
narrative and agreeing to everything
00:11:12
they want which is to create all these
00:11:14
new regulations and a new agency I think
00:11:17
that Sam is pretty much guaranteeing
00:11:19
that he'll be one of the people who gets
00:11:20
to help shape the new agency and and the
00:11:24
rules they're going to operate under and
00:11:25
what these independent audits are gonna
00:11:27
how they're gonna determine what's in
00:11:28
compliance so he is basically putting a
00:11:31
big moat around his own incumbency here
00:11:35
and so yes it is a smart strategy for
00:11:39
him but the question is do we really
00:11:41
need any of this stuff and you know what
00:11:43
you heard at the hearing is that just
00:11:45
like with just about every other Tech
00:11:47
issue the centers on the Judiciary
00:11:49
Committee didn't exhibit any real
00:11:51
understanding of the technology and so
00:11:54
they all generally talked about their
00:11:55
own hobby horses so you know you heard
00:11:58
from Senator Blackburn she wants to
00:11:59
protect songwriters Holly wants to stop
00:12:02
anti-conservative bias Klobuchar was
00:12:04
touting a couple of bills that have her
00:12:07
name on them one's called the jcpa
00:12:09
journalism competition presentation
00:12:10
Bernie Sanders want to do he wants to
00:12:12
protect the one percent of the one
00:12:13
Durbin hates section 230 that was the
00:12:16
hobby horse he was riding and then
00:12:17
Senator Blumenthal was obsessed that
00:12:19
someone had published deep fakes of
00:12:20
himself so you know all of these
00:12:23
different Senators had different
00:12:25
theories of harm that they were
00:12:26
promoting and they were all basically
00:12:29
hammers looking for a nail you know they
00:12:31
all wanted to regulate this thing and
00:12:33
they didn't really pay much of any
00:12:36
attention to the ways that existing laws
00:12:37
could already be used absolutely to stop
00:12:40
any of these things if you commit a
00:12:42
crime with AI there are plenty of
00:12:44
criminal laws every single thing they
00:12:45
talked about could be handled through
00:12:47
existing law if they came out to be
00:12:49
harmed at all yeah but they want to jump
00:12:51
right to creating a new agency and new
00:12:53
regulations and Sam I think did the
00:12:57
you know expedient thing here which is
00:12:59
basically buy into it in order an
00:13:02
Insider if this was a chess game Sam got
00:13:04
to the mid game he traded all the pieces
00:13:06
and went right to the end game let's
00:13:08
just try to Checkmate here I've got the
00:13:09
lead I got the 10 billion from Microsoft
00:13:12
everybody else get a license and try to
00:13:13
catch up Friedberg we have trimath Pro
00:13:16
regulation licensing I think or just
00:13:19
being pretty thoughtful about it there
00:13:20
you got
00:13:21
sax being typically a free market
00:13:24
monster let the laws be what they are
00:13:26
but these senators are going to do
00:13:27
regulatory capture where do you
00:13:28
as the Sultan of science stand on this
00:13:31
very important issue I think there is a
00:13:35
more important
00:13:37
kind of broader set of trends that are
00:13:40
worth noting and that the folks doing
00:13:42
these hearings and having these
00:13:43
conversations are aware of
00:13:45
which implies why they might be saying
00:13:47
the things that they're saying that's
00:13:49
not necessarily about regulatory capture
00:13:51
and that is that a lot of these models
00:13:53
can be developed and generated to be
00:13:56
much smaller we're seeing you know
00:13:57
models that can effectively run on an
00:13:59
iPhone we're seeing a number of Open
00:14:01
Source models that are being published
00:14:02
now there's a group called Mosaic ml
00:14:04
last week they published what looks like
00:14:07
a pretty good quality model
00:14:09
that has you know a very large kind of
00:14:12
token input which means you can do a lot
00:14:14
with it and that model can be downloaded
00:14:17
and used by anyone for free you know
00:14:19
really good open source license that
00:14:21
they've provided on that model and
00:14:22
that's really just the tip of the
00:14:23
iceberg on what's going on which is that
00:14:25
these models are very quickly becoming
00:14:27
ubiquitous to monetized small and
00:14:30
effectively are able to move to and be
00:14:32
run on the edge of the network as a
00:14:34
result that means that it's very hard to
00:14:36
see who's using what models how behind
00:14:39
the products and tools that they're
00:14:40
building
00:14:41
and so if that's the trend then it
00:14:45
becomes very hard for a regulatory
00:14:47
agency to go in and audit every server
00:14:50
or every computer or every computer on a
00:14:52
network and say what model are you
00:14:54
running is that an approved model is
00:14:55
that not an approved model it's almost
00:14:57
like having a regulatory agency that has
00:14:59
to go in and
00:15:01
audit and assess whether a Linux upgrade
00:15:05
or some sort of you know open source
00:15:07
platform that's that's being run on some
00:15:09
server is appropriately vetted and
00:15:11
checked and so it it's almost like a
00:15:14
Fool's errand and so if I'm running one
00:15:17
of these companies and I'm trying to you
00:15:19
know get Congress off my butt and get
00:15:21
all these Regulators off my butt I'm
00:15:22
going to say go ahead and regulate us
00:15:23
because the truth is there really isn't
00:15:26
a great or easy path or ability to do
00:15:28
that and there certainly won't be in
00:15:29
five or ten years once these models all
00:15:31
move on to the edge of the network and
00:15:34
they're all being turned around all the
00:15:35
time every day
00:15:37
and there's a great Evolution underway
00:15:39
so I actually take a point of view that
00:15:42
it's not just that this is necessarily
00:15:44
bad and there's cronyism going on I
00:15:47
think that the point of view is just
00:15:48
that this is going to be a near
00:15:50
impossible task to try and track and
00:15:53
approve llms and audit servers that are
00:15:57
running llms and audit apps and audit
00:15:59
what's behind the tools that everyday
00:16:01
people are using
00:16:03
and I wish everyone the best of luck in
00:16:05
trying to do so but that's kind of the
00:16:07
joke of the whole thing it's like let's
00:16:08
go ahead and paddle these Congress
00:16:10
people on the shoulder and say you got
00:16:11
it you're right there you have it folks
00:16:13
wrong answer chamatha sacks right answer
00:16:16
Friedberg if you were to look at hugging
00:16:18
face if you don't know what that is it's
00:16:19
a basically an open source repository of
00:16:22
all of the llms the cat is out of the
00:16:25
bag the horses have left the barn if you
00:16:28
look at what I'm showing on the screen
00:16:30
here this is the open llm leaderboard
00:16:32
it's kind of buried on hugging face if
00:16:34
you haven't been to hugging face this is
00:16:35
where developers show their work they
00:16:37
share their work and they kind of
00:16:39
compete with each other in a social
00:16:40
network showing all their contributions
00:16:42
and what they do here is and this is
00:16:44
super fascinating they have a series of
00:16:47
tests that will take an llm a language
00:16:50
model and they will
00:16:52
have it do science questions that would
00:16:54
be for grade school they'll do a test of
00:16:57
mathematics U.S history computer science
00:16:59
Etc there's a Jeopardy test too I don't
00:17:01
know if it's on here but the Jeopardy
00:17:03
test is really good it's like straight
00:17:04
up Jeopardy trivia and see if it can
00:17:05
answer the questions yeah which actually
00:17:07
uh Friedberg was actually his high
00:17:09
school uh Jeopardy Championship three
00:17:11
years in a row
00:17:12
but anyway on this line report
00:17:15
you can see the language models are
00:17:18
outpacing what openai did I'm sorry
00:17:22
closed AI is what I call it now because
00:17:24
they're closed Source closed Ai and Bard
00:17:26
have admitted that internal person at
00:17:29
Bart said the language models here are
00:17:31
now outpacing what they're able to do
00:17:33
with much more resources many hands
00:17:35
makes for light work the Open Source
00:17:37
models are going to fit on your phone or
00:17:38
the latest you know Apple silicone so I
00:17:41
think the cat's out of the bag I don't
00:17:43
know how they post something
00:17:44
incompatible about that what Freebird
00:17:46
just said with what I said in fact three
00:17:48
verse point bolsters my point it's
00:17:49
highly impractical to regulate open
00:17:53
source software in this way also when
00:17:55
you look at that list of things that
00:17:56
people are doing on hogging face there's
00:17:58
nothing nefarious about it yeah and all
00:18:00
the harms that were described are
00:18:02
already illegal and can be prosecuted
00:18:03
exactly I need some special agency you
00:18:06
know giving it seal of approval again
00:18:08
this is going to replace permissionless
00:18:10
Innovation which is what has defined the
00:18:12
software industry and especially open
00:18:14
source with the knee need to develop
00:18:17
some connection or relationship and
00:18:18
lobbying in Washington to go get your
00:18:20
project approved and there's no really
00:18:23
good reason for this except for the fact
00:18:24
that the Senators on the Judiciary
00:18:27
Committee and all of Washington really
00:18:28
wants more control so they can get more
00:18:31
donations sax I have a question do you
00:18:33
think that creating the DMV
00:18:35
and requiring a driver's license limits
00:18:38
the ability for people to learn how to
00:18:40
drive the DMV is like the classic
00:18:42
example of how government doesn't work I
00:18:44
don't know why you'd want to make that
00:18:45
your example I mean people have to spend
00:18:48
the whole time he sends people to it
00:18:50
he's got a VIP person all day waiting in
00:18:53
line to get your photo taken it's insane
00:18:54
I mean everyone has a miserable
00:18:56
experience with it no but it's highly
00:18:58
relevant because you're you're right if
00:19:00
you create an agency where people have
00:19:02
to go get their permission it's a
00:19:03
licensing scheme you're gonna be waiting
00:19:06
in some line of Untold length it won't
00:19:08
be like a physical line at the DMV
00:19:10
building it's going to be a virtual line
00:19:12
where you're in some queue where there's
00:19:13
probably going to be some overwork
00:19:14
regulator who doesn't even know how this
00:19:17
was to approve your project they're just
00:19:19
going to be trying to cover their ass
00:19:20
because if the project ends up being
00:19:22
something nefarious
00:19:24
then they get blamed for it so that's
00:19:26
what's going to end up happening let me
00:19:28
also highlight something that I think is
00:19:29
maybe I think a little bit misunderstood
00:19:33
but you know an AI model is an algorithm
00:19:37
so it's a it's a piece of software that
00:19:40
takes data in and spits data out
00:19:42
and you know we have algorithms that are
00:19:44
written by humans we have algorithms
00:19:46
that have been you know written by
00:19:48
machines either machine learn models
00:19:50
which is what a lot of what people are
00:19:52
calling AI today is effectively an
00:19:54
extension of and out of
00:19:56
and so the idea that a particular
00:19:58
algorithm is differentiated from another
00:20:02
algorithm is also what makes this very
00:20:04
difficult because these are algorithms
00:20:06
that are embedded and sit Within
00:20:07
products and applications that an end
00:20:10
user and End customer ultimately uses
00:20:12
and I just sent you guys a
00:20:15
a link to the you know the EU has been
00:20:17
working towards passing this AI good
00:20:19
here we go there are a couple of weeks
00:20:21
ahead of these conversations in the U.S
00:20:23
but I mean as you read through this AI
00:20:26
Act and the proposal that it's uh that
00:20:28
it's put forth it almost becomes the
00:20:31
kind of thing that you say
00:20:33
I just don't know if these folks really
00:20:35
understand how the technology works
00:20:37
because it's almost as if they're going
00:20:39
to audit and have
00:20:41
you know an assessment of the risk level
00:20:43
of every software application out there
00:20:46
and that the tooling and the necessary
00:20:49
infrastructure to be able to do that
00:20:50
just makes no sense in the context of
00:20:52
Open Source software in the context of
00:20:54
an open internet uh in the context of
00:20:56
how quickly software and applications
00:20:59
and tools evolve and you make tweets to
00:21:01
an algorithm and you got to resubmit it
00:21:02
for authorization sure their number one
00:21:06
job freedberg is going to be to protect
00:21:09
jobs so anything there that in any way
00:21:12
infringes on somebody's ability to be
00:21:14
employed in a position whether it's a
00:21:16
artist or a writer or a developer
00:21:18
they're going to say you can't use these
00:21:20
tools or they're going to try to follow
00:21:21
them to try to protect jobs because
00:21:23
that's their second question of all
00:21:25
three of you do you guys think that this
00:21:27
was Sam's way of pulling up the ladder
00:21:28
behind him of course no 100 just like
00:21:31
okay absolutely it is and it's because
00:21:33
no you can you can prove it he made open
00:21:36
AI closed AI by making it not open
00:21:38
source if you're Sam you're smart enough
00:21:40
to know how quickly the models are
00:21:44
commoditizing and how many different
00:21:46
models there are that you know can
00:21:49
provide similar degrees of functionality
00:21:50
as you just pointed out jaycal so I
00:21:52
don't think it's about trying to lock in
00:21:53
your model I think it's about
00:21:55
recognizing the impracticality of
00:21:57
creating some regulatory regime around
00:21:59
model auditing and so you're very so
00:22:02
that in that in that world in that
00:22:04
scenario where you have that Vision you
00:22:06
have that foresight do you go to
00:22:07
Congress and tell them that they're dumb
00:22:08
to regulate AI or do you go to Congress
00:22:10
and you say great you should regulate AI
00:22:12
knowing that it's like hey yeah you
00:22:14
should go ahead and stop the Sun from
00:22:15
shining you know like it's just yeah so
00:22:18
basically he's telling him to do that
00:22:20
because he knows they can't no therefore
00:22:22
he gets all the points all the joy
00:22:24
points all the social credit I don't
00:22:27
want to say virtual signaling but he
00:22:28
gets all the credit relationship credit
00:22:30
with Washington for saying what they
00:22:32
want to hear reflecting back to them
00:22:34
even though he knows they can't compete
00:22:36
with Facebook's open model which is
00:22:38
number one yeah there is historical
00:22:39
precedent interesting for companies that
00:22:42
are facing Congressional scrutiny to go
00:22:45
to Congress and say go ahead and
00:22:46
regulate us as a way of Premium relief
00:22:48
yeah and I think that it doesn't
00:22:50
necessarily mean you're going to get
00:22:51
regulated but it's a way of kind of
00:22:53
creating some relief and getting
00:22:54
everyone to take a breather and a sigh
00:22:56
of relief and be like okay the industry
00:22:57
is with us you know what do you think
00:23:00
the gardas strategy he's pulling here
00:23:03
list guard does I think that's in chess
00:23:05
when you are going to take the queen
00:23:07
anyway what do you think of his chess
00:23:10
movies that's uh not a strategy in chess
00:23:14
so I think it is a chess move
00:23:16
nonetheless is he pulling up the ladder
00:23:19
sacks or no I don't think that's his
00:23:21
number one goal but I think it is the
00:23:23
result
00:23:24
and so I think the the goal here is I
00:23:28
think he's got two passing from and when
00:23:30
you go to testify like this you can
00:23:32
either resist and they will put you in
00:23:35
the hot seat and just Grill you for a
00:23:36
few hours or you can sort of concede and
00:23:39
you buy into their narrative and then
00:23:42
you kind of get through the hearing
00:23:43
without being grilled and so I think on
00:23:46
that level it's preferable just to kind
00:23:48
of play ball and then the other thing is
00:23:51
that by playing ball you get to be part
00:23:53
of the Insiders Club that's going to
00:23:54
shape these regulations and that will I
00:23:56
wouldn't say it's a ladder coming up I
00:23:58
think it's more of a moat where because
00:24:00
it's not that the latter comes up and
00:24:02
nobody else can get in but the
00:24:04
regulations are going to be a pretty big
00:24:06
moat around major incumbents who know
00:24:09
they qualify for this because they're
00:24:10
going to write these standards so at the
00:24:13
end of the day if you're someone in
00:24:15
Sam's shoes you're like why resist and
00:24:17
make myself a Target or I'll just buy
00:24:21
into the narrative and help shape the
00:24:22
regulations and it's good for my
00:24:24
business I like the analysis this
00:24:25
gentlemen this is a perfect analysis let
00:24:27
me ask you about the question tomorrow
00:24:28
what is the commercial incentive from
00:24:31
your point of view
00:24:33
to ask for regulation and to be
00:24:35
pro-regulation your pro regulation can
00:24:37
you just highlight for me at least what
00:24:38
you think
00:24:39
you know the the commercial reason is to
00:24:42
do that you know how do you benefit from
00:24:44
that like not you personally but
00:24:45
generally like where does benefit arise
00:24:48
I think that certain people in a sphere
00:24:51
of influence and I would put us in that
00:24:53
category
00:24:54
have to have the intellectual capacity
00:24:56
to see beyond ourselves and ask what's
00:24:59
for
00:25:00
the greater good
00:25:02
I think Buffett is right two weeks ago
00:25:05
he equated AI to nuclear weapons which
00:25:10
is an incredibly powerful technology
00:25:11
whose Genie you can't put back in the
00:25:14
bottle who's 99.9 of use cases are
00:25:19
generally quite societally positive but
00:25:22
the point one percent of use cases
00:25:24
destroys Humanity
00:25:26
and so I think you guys are unbelievably
00:25:28
naive on this topic and you're letting
00:25:31
your ideology
00:25:34
fight your common sense the reality is
00:25:36
that there are probably 95 billion
00:25:40
trillion use cases that are incredibly
00:25:42
positive
00:25:43
but the 1000 negative use cases are so
00:25:46
destructive and they're equally possible
00:25:49
and the reason they're equally possible
00:25:50
and this is where I think there's a lot
00:25:51
of intellectual dishonesty here is we
00:25:54
don't even know how Transformers work
00:25:56
the best thing that happened when
00:25:58
Facebook open sourced llama was also
00:26:00
that somebody stealthily released all
00:26:02
the model weights yeah
00:26:04
okay so I don't think a little bit for a
00:26:07
neophyte what we're talking about so so
00:26:09
there's the model and there's the
00:26:10
weights think about it as it's a
00:26:12
solution to a problem the solution looks
00:26:14
like a polynomial equation okay let's
00:26:17
take a very simple one let's take
00:26:18
Pythagorean theorem you know x squared
00:26:20
plus y squared equals z squared okay so
00:26:23
if you want to solve an answer to a
00:26:25
problem you have these weights you have
00:26:27
these variables and you have these
00:26:28
weights associated with it the slope of
00:26:31
a line Y equals MX plus b okay what a
00:26:34
computer does with AI is it figures out
00:26:36
what the variables are
00:26:37
and it figures out what the weights are
00:26:39
the answer to identifying images
00:26:42
flawlessly turns out to be 2x plus 7
00:26:46
where x equals this thing
00:26:49
now take that example and multiply it by
00:26:52
500 billion parameters
00:26:54
and 500 billion weights and that is what
00:26:58
an AI model essentially gives us to as
00:27:00
an answer to a question
00:27:02
so even when Facebook released llama
00:27:05
what they essentially gave us was the
00:27:07
equation but not the weights
00:27:09
and then what this guy did I think it
00:27:11
was an intern apparently or somebody he
00:27:13
just looked the weights so that we
00:27:15
immediately knew what the structure of
00:27:17
the equation looked like
00:27:19
so that's what we're basically solving
00:27:21
against but we don't know how these
00:27:23
things work we don't really know how
00:27:25
Transformers work and so this is my
00:27:26
point when I think you guys are right
00:27:29
about yes the overwhelming majority of
00:27:31
the use cases but there will be people
00:27:34
who can nefariously create havoc and
00:27:36
chaos and I think you got to slow the
00:27:39
whole ship down to prevent those few
00:27:42
folks from ruining it for everybody
00:27:46
I haven't had a chance to chime in on my
00:27:48
position so I'd like to just move mine
00:27:50
nobody cares okay well I do I think
00:27:54
actually I split the difference here a
00:27:55
little bit I don't think it needs to be
00:27:56
an agency in licensing I do think we
00:27:58
have to have a commission and we do need
00:28:00
to have people being thoughtful about
00:28:01
those thousand use cases chamoth because
00:28:03
they are going to cause societal harm or
00:28:07
things that we cannot anticipate and
00:28:09
then number two for the neophyte with
00:28:11
the 1600 rating on chess.com sax gardes
00:28:15
an announcement to the opponent that
00:28:17
their Queen is under direct attack
00:28:18
similar to the announcement of check the
00:28:20
warning was customary until the early
00:28:21
20th century so since you do not know
00:28:23
the history of check now you've learned
00:28:25
something here early 20th century okay
00:28:26
well since I've only played chess in the
00:28:28
20th and 21st centuries I'm unaware of
00:28:30
that Jake Owen French is pronounced
00:28:37
in the context of what we're talking
00:28:39
about that models are becoming smaller
00:28:42
and can be run on the edge and there's
00:28:44
obviously hundreds and thousands of
00:28:45
variants of these open source models
00:28:47
that have you know good effect and
00:28:50
perhaps compete with some of these these
00:28:53
models that you're mentioning that are
00:28:54
closed source
00:28:56
how do you regulate that how do you and
00:28:58
then they sit behind an application and
00:29:00
they sit behind tooling yeah I think in
00:29:03
order for you to be able to compile that
00:29:04
model to generate that initial
00:29:06
instantiation you're still running it in
00:29:09
a cluster of thousands of gpus but let's
00:29:11
say we're past that you can't be past
00:29:14
that we're not past that yet okay we
00:29:16
don't have five million models we don't
00:29:18
have all kinds of things that solve all
00:29:20
kinds of problems we don't have an open
00:29:22
source available simulation of every
00:29:24
single molecule in the world including
00:29:26
all the toxic materials that could
00:29:28
destroy humans we don't have that yet so
00:29:30
before that is created and shrunk down
00:29:32
to an iPhone I think we need to put some
00:29:35
stage gates up to slow people down what
00:29:37
do you mean by those stage kids yeah I
00:29:38
think you need some form of kyc I think
00:29:40
before you're allowed to run a massive
00:29:43
cluster to generate the model that then
00:29:45
you try to shrink
00:29:46
you need to be able to show people that
00:29:48
you're not trying to do something
00:29:49
absolutely chaotic or have it creating
00:29:52
that I don't think that that could be as
00:29:54
simple as putting your driver's license
00:29:55
in your social security number that
00:29:58
you're working on an instance in a cloud
00:29:59
right it could be your putting your name
00:30:01
on your work it becomes slightly more
00:30:03
nuanced than that it's like I think that
00:30:05
Jake out that's probably the simplest
00:30:07
thing for AWS gcp and Azure to do which
00:30:10
is that if you want to run over a
00:30:13
certain number of GPU clusters you need
00:30:14
to put in that information I think you
00:30:16
also need to put in your tax ID number
00:30:18
so I think if you want to run a real
00:30:19
high scale model that's still going to
00:30:21
run you tens or hundreds of millions of
00:30:23
dollars I do think there aren't that
00:30:25
many people running those things and I
00:30:27
do think it's easy to police those and
00:30:29
say what are you trying to do here so
00:30:31
let me just push back on that because
00:30:32
Mosaic ml published this model that is
00:30:35
let me I can pull up the the performance
00:30:37
chart or Nick maybe you can just find it
00:30:39
on their website real quick or the new
00:30:40
model they publish to mock they trained
00:30:42
this model
00:30:43
on open source data that's publicly
00:30:47
available
00:30:48
and they spent two hundred thousand
00:30:50
dollars on a cluster run to build this
00:30:52
model and look at how it performs
00:30:54
compared to some of the top models that
00:30:56
are closed Source just say it for the
00:30:57
people who are listening yeah so for
00:30:59
people that are listening basically this
00:31:00
model is called MPT 7B that's the name
00:31:04
of the AI model the llm model that was
00:31:07
generated by this group called Mosaic ml
00:31:10
and they spent two hundred thousand
00:31:11
dollars creating this model from scratch
00:31:14
and the data that they trained it on is
00:31:16
all listed here it's all publicly
00:31:18
available data that you can just
00:31:19
download off the internet then they
00:31:20
score how well it performs on its
00:31:22
results against other big models out
00:31:25
there like llama 7B
00:31:27
I know but I don't exactly know what the
00:31:30
actual problems they're trying to ask it
00:31:32
to compare right so the but the point is
00:31:35
that this model theoretically could then
00:31:37
be applied to a different data set once
00:31:39
it's been you know built I don't know
00:31:42
this is you know I just want to use your
00:31:43
point earlier about toxic chemistry
00:31:45
because models were generated and then
00:31:47
other data was then used to fine-tune
00:31:50
those models and deliver an output hold
00:31:52
on a second like those answers were to
00:31:54
specific kinds of questions
00:31:56
if you wanted to all of a sudden ask
00:31:58
totally orthogonal thing of that model
00:32:00
that model would fail you would have to
00:32:02
go back and you'd have to retrain it
00:32:04
that training does cost some amount of
00:32:06
money so if you said to me hmoth I could
00:32:09
build you a model trained on the
00:32:11
universe of every single molecule in the
00:32:13
world and I could actually give you
00:32:15
something that could generate the toxic
00:32:17
list of all the molecules and how to
00:32:18
make it for two hundred thousand dollars
00:32:20
I would be really scared I don't think
00:32:22
that that's possible today
00:32:24
so I don't understand these actual tests
00:32:26
but I don't think it's true that you
00:32:28
could take this model and these model
00:32:29
weights apply to a different set of data
00:32:32
and get useful answers but let's let's
00:32:34
assume for a minute that you can in fact
00:32:36
take two hundred thousand dollars worth
00:32:38
it
00:32:39
here's my point I want to tell you
00:32:41
what's happening right now which is
00:32:43
that's not possible so we should stop so
00:32:46
that then I don't have to have this
00:32:47
argument with you in a year from now
00:32:49
which is like hey some jack jerk off
00:32:51
just created this model now the cat's
00:32:53
out of the bag so let's not do it yeah
00:32:55
and and then what's going to happen is
00:32:57
like some chaotic seeking organization
00:32:59
is going to print one of these materials
00:33:01
and release it into the wild to prove it
00:33:03
but here's the point for the audience we
00:33:04
are at a moment in time where this is
00:33:06
moving very quickly and you have very
00:33:08
intelligent people here who are very
00:33:09
knowledgeable
00:33:10
talking about to the degree to which
00:33:13
this is going to manifest itself not if
00:33:15
it will manifest you are absolutely 100
00:33:17
certain Freeburg that somebody will do
00:33:19
something very bad in terms of the
00:33:22
chemical example as but one we're only
00:33:25
determining here what level of hardware
00:33:27
and what year that will happen chimatha
00:33:29
saying we know it's going to happen
00:33:31
whether it's 2 or 10 or you know five
00:33:33
years let's be thoughtful about it and I
00:33:36
think you know this discussion we're
00:33:38
having here I think
00:33:39
on a spectrum
00:33:42
it's this is a unique moment where the
00:33:45
most knowledgeable people
00:33:47
across every single political Spectrum
00:33:49
persuasion for-profit non-profit
00:33:51
Democrat Republican right Elon and Sam
00:33:54
will just use those as the two canonical
00:33:56
examples to demonstrate
00:33:58
are pro-regulation and then the further
00:34:00
and further you get away the less
00:34:02
technically astute you are the more
00:34:04
anti-regulation and like pro-market you
00:34:07
are and all I'm saying is I think that
00:34:08
should also be noted that that's a
00:34:10
unique moment that the only other time
00:34:12
that that's happened was around nuclear
00:34:14
weapons and you know that's when
00:34:16
Einstein I actually think I think it's
00:34:20
politically
00:34:21
incorrected right now I think because of
00:34:24
what you're saying well just give me a
00:34:25
second I think because of what you're
00:34:26
saying everyone on the left and the
00:34:27
right it's it's become popular to be
00:34:30
pro-regulation on AI and to say that AI
00:34:32
is going to Doom the world and it's
00:34:33
unpopular and absolutely
00:34:37
explained my point of view on Sam elon's
00:34:39
different but I think like it I think
00:34:42
it's become Politically Incorrect to
00:34:43
stand up and say you know what this is a
00:34:45
transformative technology for Humanity I
00:34:47
don't think that there's a real path to
00:34:48
regulation I think that I totally
00:34:50
there's laws that are in place that can
00:34:52
protect us in other ways with respect to
00:34:54
privacy with respect to fraud with
00:34:56
respect to biological warfare and all
00:34:58
the other things that we should worry
00:34:59
about Elon has said pretty clearly he
00:35:00
doesn't give a [ __ ] about what it does
00:35:02
to make money or not he cares about what
00:35:04
he thinks so all I'm saying is that's a
00:35:06
guy that's not trying to be politically
00:35:07
correct Elon has a very specific concern
00:35:09
which is Agi he's concerned that we're
00:35:12
on a path to a digital superintelligence
00:35:14
it's a singularity and if we create the
00:35:16
wrong kind of
00:35:18
artificial general intelligence that
00:35:20
decides that it doesn't like humans that
00:35:23
is a real risk to the human species
00:35:25
that's the concern he's expressed but
00:35:28
that's not what the hearing was really
00:35:31
about and it's not what any of these
00:35:33
regulatory proposals are about and the
00:35:35
reality is none of these Senators know
00:35:37
what to do about that even the industry
00:35:39
doesn't know what to do about the
00:35:41
long-term risk of creating an AGI nobody
00:35:44
knows nobody really and so and so I
00:35:46
actually I disagree with this idea that
00:35:48
tomorrow early earlier you said that
00:35:50
there's a thousand use cases here that
00:35:52
could destroy the human species I think
00:35:54
there's only one there's only one
00:35:57
species level risk which is Agi but
00:36:00
that's a long-term risk we don't know
00:36:01
what to do about it yet I agree we
00:36:03
should have conversations what we're
00:36:04
talking about today is whether we create
00:36:07
some new licensure regime in Washington
00:36:09
so that politically connected insiders
00:36:12
get to control and shape the software
00:36:13
industry and that's a disaster let me
00:36:16
give you another detail on this so
00:36:18
and one of the
00:36:20
chat groups I'm in there was somebody
00:36:22
who just got back from Washington I
00:36:25
won't say who they are it's not someone
00:36:26
who's famous outside the industry but
00:36:28
they're kind of like a tech leader and
00:36:30
what they said is they just got back
00:36:32
from Capitol Hill and the White House
00:36:34
and I guess there's like a White House
00:36:36
Summit
00:36:37
on AI you guys know about that yeah so
00:36:41
what this person said is that the White
00:36:45
House meeting was super depressing some
00:36:47
smart people were there to be sure but
00:36:51
the White House and vp's teams were
00:36:53
rapidly negative no real concern for the
00:36:55
opportunity or economic impact just
00:36:58
super negative of course basically the
00:36:59
mentality was that Tech is bad we hate
00:37:02
social media this is the hot new thing
00:37:04
we have to stop it of course that
00:37:06
basically is their attitude they don't
00:37:07
understand the technology White House
00:37:09
yeah and the VP specifically because
00:37:11
she's the now the AIS are to put Kamala
00:37:14
Harris in charge of this makes no sense
00:37:15
I mean does she have any background in
00:37:18
this like it just shows like a complete
00:37:19
utter lack of awareness where is the
00:37:22
Megan Smith of course somebody like a
00:37:24
CTO to be put in charge of this remember
00:37:26
Megan Smith was CTO under I guess Obama
00:37:28
like you need somebody with a little
00:37:30
more depth of experience here like
00:37:32
hopefully you guys than saying your your
00:37:34
pro-regulation depending on who's in
00:37:36
charge well I'm proud thoughtfulness I'm
00:37:38
pro-lawfulness I'm illustrating that
00:37:41
really this whole new agency that's
00:37:43
being discussed is just based on Vibes
00:37:45
you're not down with the Vibes of Vibes
00:37:48
the vibe is that a bunch of people in
00:37:51
Washington don't understand technology
00:37:52
and they're afraid of it yeah
00:37:55
these are socialist David they're
00:37:57
socialists they hate progress they are
00:38:00
scared to death that jobs are going to
00:38:02
collapse they're socialists they they're
00:38:04
union leaders this is their worst
00:38:06
nightmare Because the actual truth of
00:38:08
this technology is 30 more efficiency
00:38:11
and it's very mundane this is the truth
00:38:14
here I think that Representatives have
00:38:16
30 more efficiency means Google Facebook
00:38:19
and many other companies Finance
00:38:21
education they do not add staff every
00:38:24
year they just get 30 more efficient
00:38:26
every year and then we see unemployment
00:38:28
go way up and Americans are going to
00:38:30
have to take service jobs and why collar
00:38:32
jobs are going to be refined to like a
00:38:35
very elite few people who actually do
00:38:37
work in the world there is absolutely a
00:38:40
lot of new companies if humans can
00:38:42
become if knowledge workers can become
00:38:43
30 more productive there'll be a lot of
00:38:45
new companies absolutely and the biggest
00:38:47
shortage on our economy is coders right
00:38:49
and we're going to have an unlimited
00:38:50
number of them now they're all going to
00:38:51
go I don't know if it's unlimited but
00:38:53
yes it's a good thing if you give them
00:38:54
superpowers we've talked about this
00:38:55
before yeah yeah so I think it's too
00:38:57
soon to be concluding that we need to
00:39:00
stop job displacement that hasn't even
00:39:02
occurred yet I'm not saying it's
00:39:03
actually going to happen I do agree
00:39:04
there'll be more startups I'm seeing it
00:39:06
already I just think that's what they
00:39:08
fear that's their fear is and that's the
00:39:10
fear of the EU the EU is going to be
00:39:12
protection unionist protect Pro workers
00:39:15
aren't going to be affected because
00:39:17
these are not blue-collar jobs we're
00:39:18
talking about these are knowledge unions
00:39:20
at all the media companies created
00:39:22
unions and look at them they're all
00:39:23
going circling all these media companies
00:39:25
are circling the training going into
00:39:26
their business but that's on the margins
00:39:27
I mean that's not just trying to start
00:39:29
Tech unions sure they're trying to start
00:39:30
them but when we think of unionized
00:39:32
workers you're thinking about Factory
00:39:33
workers and these people are not
00:39:35
affected okay listen this has been an
00:39:38
incredible debate this is why you tune
00:39:39
into the pod
00:39:40
a lot of things can be true at the same
00:39:42
time I really think the analogy of the
00:39:43
the atom bomb is really interesting
00:39:45
because what Elon is scared about with
00:39:48
General artificial intelligence is
00:39:49
nuclear Holocaust the whole planet blows
00:39:51
up between those two things are things
00:39:54
like Nagasaki and Hiroshima or a dirty
00:39:57
bomb and many other possibilities with
00:39:59
nuclear power you know Fukushima you
00:40:03
know Etc so less than the Hiroshima not
00:40:06
yet right and you know the question is
00:40:08
is a three mile island is a Fukushima is
00:40:12
a Nagasaki are those things probable and
00:40:15
I think we are all looking at this
00:40:16
saying there will be something bad that
00:40:18
will happen there will be the equivalent
00:40:21
strings together and these large
00:40:23
language models string together words in
00:40:25
really interesting ways and they give
00:40:26
computers the ability to have a natural
00:40:28
language interface that is so far from
00:40:30
AGI I think it's a company
00:40:33
hold on I think it's obviously the
00:40:35
ability to understand language and
00:40:37
communicate in a natural way is a
00:40:39
component of a future AGI but by itself
00:40:42
these are models for stringing together
00:40:44
language auto GPT where these things go
00:40:47
out and pursue things without any
00:40:49
interference I would be the first one to
00:40:51
say that if you wanted to scope
00:40:53
models to be able to just do human
00:40:56
language back and forth
00:40:58
on the broad open internet
00:41:01
you know there's probably a form David
00:41:04
where these chat GPT products can exist
00:41:06
I don't I think that those are quite
00:41:08
benign I agree with you but I think what
00:41:10
Jason is saying is that every week
00:41:12
you're you're taking a leap forward and
00:41:14
already with auto gpts you're talking
00:41:16
about code that runs in the background
00:41:18
without supervision it's not a human
00:41:20
interface that's like hey show me how to
00:41:22
color my
00:41:23
cookies green for St Patty's Day it's on
00:41:26
my trip to Italy yeah yeah it's not
00:41:28
doing that so I just think that there's
00:41:30
there's a place well beyond what you're
00:41:32
talking about and I think you're
00:41:33
minimizing the problem a little bit by
00:41:34
just kind of saying the whole class of
00:41:36
AI is just chat GPT and asking kid
00:41:39
asking it to help it with its homework
00:41:41
this example I hate to say it out loud
00:41:42
but somebody could say here is the
00:41:45
history of financial crimes that were
00:41:47
committed and other hacks please with
00:41:50
their own model on their own server say
00:41:52
please come up with other ideas for
00:41:54
hacks be as creative as possible and
00:41:56
steal as much money as possible and put
00:41:57
that in an auto GPT David and study all
00:42:00
hacks that occur in the history of
00:42:02
hacking and it could just create super
00:42:04
chaos around the world and your computer
00:42:06
history is going to regret buying into
00:42:08
this narrative because the members of
00:42:10
the Judiciary Committee are doing the
00:42:11
same Playbook they ran back in 2016
00:42:13
after that election they ran all these
00:42:15
hearings on disinformation claiming that
00:42:17
social networks been used to hack the
00:42:19
election it was all a phony narrative
00:42:20
hold on look what happened
00:42:24
the Russians hacked who had to write out
00:42:26
that stop they thought they got all
00:42:28
these Tech CEOs to buy into that phony
00:42:30
narrative why because it's a lot easier
00:42:32
for the tech CEOs to agree and tell the
00:42:35
centers what they want to hear to get
00:42:36
them off their backs and then what did
00:42:38
that lead to a whole censorship
00:42:40
industrial complex so we're going to do
00:42:43
the same thing here we're going to buy
00:42:44
into these phoning names the centers off
00:42:47
our backs and that's going to create
00:42:48
this giant AI industrial complex that's
00:42:52
going to slow down real Innovation and
00:42:54
be a burden on entrepreneurs okay
00:42:55
lightning round lightning round we got
00:42:57
to move on three more topics I want to
00:42:58
hit let's keep going if honor to become
00:43:00
an evil more evil yeah what is it called
00:43:03
an evil uh comic book character a super
00:43:06
villain a super villain if you wanted to
00:43:08
be an even more loathsome super villain
00:43:11
continue I would take every single
00:43:14
virus patch that's been developed and
00:43:17
publicized learn on them and then find
00:43:19
the next zero day exploit on a whole
00:43:21
bunch of stuff I mean it's just like
00:43:22
even publish that idea please don't I
00:43:23
mean I'm I'm worried about publishing
00:43:25
that idea that's not an intellectual
00:43:27
leap I mean you have to be a dollar it's
00:43:29
obvious okay let's move on another great
00:43:32
debate Elon hired a CEO for Twitter
00:43:35
Linda yakarino I'm hoping pronouncing
00:43:38
that correct was the head of AD sales
00:43:39
that NBC Universal she's a legend in the
00:43:41
advertising business
00:43:43
she worked at Turner for 15 years before
00:43:44
that she is a workaholic is what she
00:43:47
says she's going to take over everything
00:43:49
but product and CTO elon's going to
00:43:52
stick with that
00:43:53
she seems to be
00:43:55
very moderate and she follows people on
00:43:58
the left or right people are starting
00:43:59
the character assassination and trying
00:44:01
to figure out her politics
00:44:04
and that she was involved with the w
00:44:05
World economic Forum which anybody in
00:44:08
business basically does
00:44:10
but your take sacks on this choice for
00:44:13
CEO and what this means just broadly for
00:44:18
the next six months because we're
00:44:19
sitting here at six months almost
00:44:20
exactly since Elon took over obviously
00:44:22
you and I were involved in month one but
00:44:24
but not much after that what do you
00:44:26
think the next six months holds and what
00:44:27
do you think her role is going to be
00:44:28
obviously some precedent this for this
00:44:30
with you know why not as basic listen I
00:44:34
think this Choice makes sense on this
00:44:37
level Twitter's business model is
00:44:39
advertising Elon does not like selling
00:44:41
advertising she's really good at selling
00:44:43
advertising so he's chosen a CEO to work
00:44:46
with who's highly complimentary to him
00:44:49
in their skill sets and interests and I
00:44:52
think that makes sense I think there's a
00:44:54
lot of logic in that what Elon likes
00:44:56
doing is the technology and product side
00:44:58
of the business
00:44:59
he actually doesn't really like the
00:45:03
let's call it the standard business
00:45:05
chores and especially related to like we
00:45:08
said advertising stuff
00:45:11
with advertisers I think it's his
00:45:13
personal nightmare right so I think the
00:45:15
choice makes sense on that level now
00:45:16
yeah instantly you're right her hiring
00:45:19
led to attacks from both the left and
00:45:22
the right the right you know pointed out
00:45:24
her views on covet and vaccines and her
00:45:27
work with the wef and then on the left I
00:45:30
mean the attack is that she's following
00:45:32
Libs at Tick Tock which you're just not
00:45:34
allowed to do apparently a follow is not
00:45:36
an endorsement well if you're just
00:45:38
following lives of tick tock they want
00:45:39
to say you're some crazy right winger
00:45:40
now well she also follows David sacks so
00:45:43
that does mean that she's pretty that
00:45:45
that is a signal but the truth is if you
00:45:47
sax correct me if I'm wrong here or trim
00:45:49
off maybe I'll send it to you if you
00:45:50
pick somebody that both sides dislike or
00:45:53
are trying to take apart you probably
00:45:54
pick the right person yeah
00:45:56
here's what I think okay go ahead we're
00:45:58
not going to know how good she is for
00:46:01
six to nine months but here's what I
00:46:03
took a lot of
00:46:04
joy out of here's a guy who
00:46:08
gets attacked for all kinds of things
00:46:10
now right he's an anti-semite apparently
00:46:13
and then he had to be like I'm very Pro
00:46:15
semi he's a guy that all of a sudden
00:46:17
people think is a conspiracy theorist
00:46:18
he's a guy that people think is now on
00:46:21
the Raging right all these things that
00:46:23
are just like inaccuracies basically
00:46:25
fire bombs thrown by the left
00:46:28
but here's what I think is the most
00:46:30
interesting thing for a guy that
00:46:32
theoretically is supposed to be a troll
00:46:33
and everything else he has a female
00:46:36
chairman of Tesla a female CEO at
00:46:39
Twitter and a female president at SpaceX
00:46:42
of course it's a great Insight it's the
00:46:44
same insight
00:46:45
I think a lot of these virtual signaling
00:46:47
lunatics on the left virtue signaling
00:46:50
mids and you know what like they're all
00:46:53
new house Elizabeth Warren and and
00:46:55
Bernie Sanders giving you know the CEO
00:46:59
of Starbucks a hard time when he doubled
00:47:02
the pay of the minimum wage gave them
00:47:06
you love mid right that's a great term
00:47:09
isn't it
00:47:10
these [ __ ] mids and I paid for the
00:47:13
college tuition what gives you the right
00:47:16
at Starbucks to pay for college tuition
00:47:18
and double the minimum wage
00:47:20
I don't know why it's so funny to me
00:47:23
isn't it so great like just because you
00:47:25
can picture them when I say that these
00:47:26
are these mids feverishly typing on
00:47:29
their keyboards their virtue signaling
00:47:31
nonsense sex wrapping it up
00:47:33
yeah look like you said Elon has worked
00:47:35
extremely well with Gwen Shotwell who's
00:47:37
the president of SpaceX for a long time
00:47:39
and I think that relationship shows
00:47:43
the way to make it work here at Twitter
00:47:45
which is they have a very culinary skill
00:47:47
set I think my understanding is that
00:47:48
Gwen focuses on the business side and
00:47:50
the sales side of the operation Elon
00:47:52
focuses on product and Technology she
00:47:54
lets Elon be Elon I think if Linda tries
00:47:57
to Reign Elon in tell him not to tweet
00:48:00
or tries to meddle in the Free Speech
00:48:03
aspects of the business which is the
00:48:04
whole reason he bought Twitter
00:48:07
yeah that's right that's when it will
00:48:09
fall apart so my advice would be let
00:48:12
alone Be Elon you know he bought this
00:48:14
company to make it a free speech
00:48:16
platform don't mess with that and I
00:48:19
think it could work great and a free
00:48:21
speech platform it is when you are
00:48:23
saying anything about covid and I really
00:48:25
don't even want to say it here because I
00:48:26
don't want our to even say the word
00:48:28
covet or vaccine means that this could
00:48:30
get tagged by YouTube and BD you know
00:48:33
algor the algorithm could could uh D
00:48:37
um I don't know what they call it
00:48:38
deprecate this and when we don't show up
00:48:40
and people don't see us because we just
00:48:41
said the word covet I mean the
00:48:42
censorship built into these algorithms
00:48:44
is absurd but speaking of absurd Nina
00:48:47
Khan
00:48:48
who has been the least effective FTC
00:48:50
chair
00:48:51
I think
00:48:52
started out pretty promising with some
00:48:54
interesting ideas she's now moved a
00:48:56
block of major Pharma deal and December
00:48:59
Amgen agreed to acquire Dublin based
00:49:02
Horizon Therapeutics for 27.8 billion
00:49:05
this is the largest Pharma deal
00:49:06
announced in 2022. FTC has filed a
00:49:08
lawsuit in federal court seeking an
00:49:09
injunction that would prevent the deal
00:49:10
from closing the reasoning is the deal
00:49:13
would allow Amgen to entrench the
00:49:15
Monopoly positions of Horizon's eye and
00:49:16
gout drugs
00:49:18
the agency said that those treatments
00:49:21
don't face any competition today and
00:49:22
that Amgen would have a strong incentive
00:49:24
to prevent any potential Robbers from
00:49:27
introducing similar drugs
00:49:29
chamoth the pharmaceutical industry is a
00:49:31
little bit different than the tech
00:49:32
industry insanity
00:49:34
explain why and then sax will go to you
00:49:36
on the gout stuff because I know that
00:49:38
personally impacts you go ahead I think
00:49:41
that this is a little like
00:49:42
scientifically illiterate to be honest
00:49:45
unpack the thing is that you want drugs
00:49:48
that can get to Market quickly but at
00:49:50
the same time you want drugs to be safe
00:49:52
and you want drugs to be effective
00:49:54
and I think that the FDA has a pretty
00:49:56
reasonable process and one of the direct
00:50:00
byproducts of that process is that if
00:50:02
you have a large indication that you're
00:50:04
going after say diabetes you have to do
00:50:06
an enormous amount of work it has to be
00:50:09
run on effectively thousands of people
00:50:11
you have to stratify it by age you have
00:50:13
to stratified by gender you have to
00:50:15
stratify it by race you have to do it
00:50:17
across different geographies right the
00:50:19
bar is high but the reason the bar is
00:50:21
high is that if you do get approval
00:50:23
these all of a sudden become these
00:50:25
Blockbuster 10 20 30 billion dollar
00:50:27
drugs okay and they improve people's
00:50:30
lives and they allow people to live
00:50:33
etc etc
00:50:34
what has happened in the last 10 or 15
00:50:36
years because of wall Street's influence
00:50:39
inside of the Pharma companies
00:50:41
is that what Pharma has
00:50:44
done a very good job of doing is
00:50:47
actually pushing off a lot of this very
00:50:49
risky r d
00:50:51
to Young early stage biotech companies
00:50:54
and they typically do the first part of
00:50:57
the work they get through a phase one
00:50:59
they even may be able to sometimes go
00:51:01
and start a phase two trial a two-way
00:51:03
trial and then they typically
00:51:06
can get sold to Pharma and these are
00:51:09
like multi-billion dollar transactions
00:51:12
and the reason is that the private
00:51:14
markets just don't have the money
00:51:16
to support the risk for these companies
00:51:18
to be able to do all the way through a
00:51:19
phase three clinical trial because it
00:51:21
would cost in some cases five six seven
00:51:24
eight billion dollars you've never heard
00:51:26
of a tech company raising that much
00:51:27
money except in a few rare cases
00:51:30
in far in in biotech it just doesn't
00:51:32
happen so you need the m a machine
00:51:35
to be able to incentivize these young
00:51:37
companies to even get started in the
00:51:38
first place otherwise what literally
00:51:41
happens is you have a whole host of
00:51:42
diseases that just stagnate okay
00:51:47
and instead what happens is a younger
00:51:49
company can only raise money to go after
00:51:51
smaller diseases which have smaller
00:51:53
populations smaller Revenue potential
00:51:55
smaller costs because the trial
00:51:57
infrastructure is just less
00:51:59
so if you don't want industry to be in
00:52:03
this negative Loop where you only work
00:52:04
on the small diseases and you actually
00:52:06
go and tackle the big ones
00:52:08
you need to allow these kinds of
00:52:09
transactions to happen the last thing
00:52:11
I'll say is that even when these big
00:52:12
transactions happen half the time they
00:52:15
turn out to still not work there is
00:52:17
still huge risk so don't get caught up
00:52:19
in the dollar size you have to
00:52:21
understand the phase it's in and the
00:52:23
best example of this is the biggest
00:52:24
outcome in in biotech private investing
00:52:27
in Silicon Valley was this thing called
00:52:29
stem Centrics and that thing was a 10
00:52:31
billion dollar dud right but it allowed
00:52:34
all these other companies to get started
00:52:35
after semcentrics got bought for 10
00:52:37
billion freeberg I want to get your take
00:52:40
on this especially in light of maybe
00:52:42
something people don't understand which
00:52:43
is the the amount of time you get to
00:52:46
actually exclusively monetize a drug
00:52:49
because my understanding you correct me
00:52:52
if I'm wrong you get a 20-year patent
00:52:53
it's from the date you file it but then
00:52:55
you're working towards getting this drug
00:52:56
approved by the FDA so by the time the
00:52:58
FDA approves a drug this 20-year patent
00:53:01
window how many years do you actually
00:53:03
have exclusively to monetize that drug
00:53:05
and then your wider thoughts on this FTA
00:53:07
yeah I'm not going to answer that
00:53:09
question right now because I do want to
00:53:10
kind of push back on the point I'm
00:53:13
generally pretty negative on a lot of
00:53:15
the comments Lena Khan's made and her
00:53:17
positioning and obviously as you guys
00:53:18
know we've talked about it on the show
00:53:20
but I read the FTC uh filing it's in um
00:53:24
federal court
00:53:26
and if you read the filing let me just
00:53:28
start the company that Amgen is trying
00:53:29
to buy it's called Horizon Therapeutics
00:53:30
which is a company that's doing about 4
00:53:33
billion in Revenue a year about a
00:53:34
billion to a billion and a half in
00:53:35
ebitda so it's a it's a business that's
00:53:38
got a portfolio of orphaned drugs
00:53:40
meaning drugs that treat orphan
00:53:41
conditions that aren't very big
00:53:43
Blockbusters in the in the
00:53:44
pharmaceutical drug context
00:53:46
and so it's it's a nice portfolio of
00:53:48
cast generating drugs Amgen buying the
00:53:52
business gives them real Revenue really
00:53:54
but uh and helps bolster portfolio that
00:53:58
you know is aging and I think that's a
00:54:00
big part of the Strategic driver for
00:54:02
Amgen to make this massive 28 billion
00:54:05
dollar acquisition the ftc's claim in
00:54:08
the filing which I actually read and I
00:54:11
was like this is actually a pretty good
00:54:12
claim is that the way that Amgen sets
00:54:16
the prices for their pharmaceutical
00:54:17
drugs is they go to the insurance
00:54:19
companies the payers and the health
00:54:20
systems and they negotiate drug pricing
00:54:23
and they often do bulk multi-product
00:54:26
deals so they'll say hey we'll give you
00:54:28
access to those products at this price
00:54:31
point but we need you to pay this price
00:54:32
point for this product and over time
00:54:34
that drives price inflation it drives
00:54:36
costs up and it also makes it difficult
00:54:38
for new competitors to emerge because
00:54:40
they tell the insurance company you have
00:54:41
to pick our drug over other drugs in
00:54:44
order to get this discounted price
00:54:46
and so it's a big part of their
00:54:47
negotiating strategy that they do with
00:54:48
insurance companies so the ftc's claim
00:54:51
is that by giving Amgen this large
00:54:53
portfolio of drugs that they're buying
00:54:55
from Horizon it's going to give them
00:54:57
more negotiating leverage and the
00:54:58
ability to do more of this drug blocking
00:55:00
that they do with insurance companies
00:55:02
and other payers in the drug system so
00:55:04
they're trying to prevent pharmaceutical
00:55:06
drug price inflation and they're trying
00:55:08
to increase competition in their lawsuit
00:55:10
so I felt like it was a fairly kind of
00:55:12
compelling case I'm no lawyer on
00:55:14
Anti-Trust and monopolistic practices
00:55:17
and the Sherman Act but this was not
00:55:18
sorry let me just say this was not an
00:55:19
early stage biotech risky deal that
00:55:22
they're trying to block a mature company
00:55:24
with four billion in revenue and a
00:55:26
billion and a half in ebitda I
00:55:27
understand and yeah I read it too but
00:55:29
two comments
00:55:30
it is because the people that traffic in
00:55:32
these stocks are the same ones that fund
00:55:34
these early stage biotech companies and
00:55:36
I talked to a bunch of them and they're
00:55:38
like if these guys block this kind of
00:55:39
deal we're going to get out of this game
00:55:41
entirely so just from the horse's mouth
00:55:44
what I'm telling you is you're going to
00:55:45
see a Paul come over the early stage
00:55:47
Venture financing landscape because a
00:55:49
lot of these guys that are crossover
00:55:51
investors that own a lot of these public
00:55:52
biotech stocks that also fund the
00:55:54
private stocks will change their risk
00:55:57
posture if they can't make money that's
00:55:59
just the nature of capitalism
00:56:00
the second thing is Lena Khan did
00:56:02
something really good about what you're
00:56:04
talking about this week actually which
00:56:06
is she actually went after the pbms and
00:56:08
if you really care about drug inflation
00:56:10
and you follow the dollars the real
00:56:12
culprits around this are the pharmacy
00:56:14
benefit managers and she actually
00:56:16
launched a big investigation into them
00:56:18
but this is what speaks to the two
00:56:20
different approaches it seems that
00:56:22
unfortunately for the FTC every merger
00:56:25
just gets contested for the sake of it
00:56:26
being contested
00:56:28
because I think that if you wanted to
00:56:30
actually stop price inflation there are
00:56:32
totally different mechanisms because why
00:56:34
didn't you just Sue all the pbms well
00:56:37
there's no merger to be done but you can
00:56:39
investigate and then you could regulate
00:56:41
and I think that that's probably a more
00:56:43
effective way and the fact that she
00:56:44
targeted the PPM says that somebody in
00:56:46
there actually understands where the
00:56:48
price inflation is coming from but I
00:56:50
don't think something like an Amgen
00:56:52
Horizon because what I think will happen
00:56:54
is all the folks will then just
00:56:56
basically say well man if if these kinds
00:56:59
of things can't get bought then why am I
00:57:00
funding these other younger things yeah
00:57:02
but you know yeah we're just not seeing
00:57:04
a lot of the younger stuff get get
00:57:06
get blocked I don't think we've seen any
00:57:08
attempts at blocking speculative
00:57:10
portfolio Acquisitions or speculative
00:57:12
company Acquisitions so I think these
00:57:14
guys are getting caught up in the dollar
00:57:15
number you know yeah so I think the
00:57:18
problem is they see 28 billion they're
00:57:19
like we need to stop it you know what's
00:57:21
amazing I'll just wrap on this because
00:57:22
it's a good discussion but I think we
00:57:24
got to keep moving here is I took the
00:57:25
PDF that you shared and I put it into
00:57:29
chat GPT now oh wow and you don't need
00:57:32
to upload the PDF anymore you could just
00:57:34
say summarize this and put the link and
00:57:36
it did it instantly
00:57:39
are using the browsing Plugin or I just
00:57:42
used chat GP no it's not the browser
00:57:43
plugin I just did this is the 3.5 model
00:57:46
and it pulled the link in the GPT 3.5
00:57:49
model I don't know I could do that
00:57:52
that's new they must have added browse
00:57:54
me in the background yeah or just
00:57:55
pulling it they did today by the way
00:57:57
whoa they did it today yeah remember
00:58:00
last week we said that they had to build
00:58:02
browsing into the actual product like
00:58:04
Bard right otherwise they're moving I
00:58:07
gotta get it today yeah
00:58:08
closed AI is on the top of their game
00:58:10
the app is available in the App Store
00:58:12
now right as of today no they they had a
00:58:15
test app I was on the test flight no no
00:58:16
no they just launched the app they did
00:58:18
oh that's game over man if this thing is
00:58:20
an app form that's going to 10x the
00:58:21
number of users and it's going to 10x
00:58:23
the amount of usage by the way I tested
00:58:24
the same thing for Bard we should
00:58:26
compare the two but amazing pretty good
00:58:28
as well yeah yeah wow to actually
00:58:31
compare its summary with chat GPT
00:58:33
summary
00:58:35
okay tell us which one's better
00:58:37
some interesting news here you know we
00:58:40
speaking of platform shifts do I get to
00:58:42
get my view on the Lena con thing oh
00:58:45
well yes
00:58:49
I was getting a little personal here
00:58:51
David and I didn't I didn't want to
00:58:53
trigger you I know you've been
00:58:53
struggling with the gout
00:58:55
because of your lifestyle choices the
00:58:57
alcohol the the foie gras everything but
00:59:00
no you've lost a lot of way to give you
00:59:01
a lot of credit tell us what do you
00:59:03
think about the bundling we're seeing
00:59:04
here because it does seem Microsoft asks
00:59:06
with the operating system yeah it's very
00:59:07
similar and what I said in the context
00:59:09
of tech is that we should focus on the
00:59:11
anti-competitive tactics and stop those
00:59:13
rather than blocking all mergers and I
00:59:16
think the same thing is happening on the
00:59:17
Pharma space if bundling is the problem
00:59:19
focus on bundling the problem when you
00:59:22
just block M A is that you deny
00:59:25
early investors one of the biggest ways
00:59:28
that they can make a positive outcome
00:59:29
and what's the downstream effect of that
00:59:31
yeah exactly look it is hard enough to
00:59:34
make money as either a former investor
00:59:36
or as a VC that there's only two good
00:59:38
outcomes right there's IPOs there's M
00:59:40
A's everything else basically goes
00:59:42
everything else is a zero it goes
00:59:43
bankrupt so
00:59:45
if you take M A off the table you really
00:59:48
suppress the already challenge returns
00:59:51
of venture capital yeah well said well
00:59:53
said and you're right you mentioned
00:59:55
earlier that we were willing to give
00:59:56
Lena Khan a chance we thought that some
00:59:59
of our ideas were really interesting
01:00:00
because I think there are these huge
01:00:02
tech companies that do need to be
01:00:04
regulated these big Tech monopolies
01:00:06
basically that you have the mobile
01:00:08
operating system duopoly with Apple and
01:00:10
Google and you've got Amazon you've got
01:00:11
Microsoft and there is a huge risk of
01:00:14
those companies uh preferring their own
01:00:16
applications over Downstream
01:00:18
applications or using these bundling
01:00:20
tactics yes if you don't put some limits
01:00:23
around that that creates I think an
01:00:25
unhealthy Tech ecosystem this is the
01:00:27
insight and I think it's exactly correct
01:00:29
sex
01:00:30
ferlina Khan I know she listens to the
01:00:32
Pod hey Lena you want to go after
01:00:34
tactics not Acquisitions so if somebody
01:00:37
buys something and they lower prices and
01:00:40
increases consumer Choice that's great
01:00:43
if it encourages more people to invest
01:00:45
more money into Innovation that's great
01:00:48
but if the tactics are we're going to
01:00:50
bundle these drugs together to keep some
01:00:52
number of them artificially high or
01:00:53
reduce choice or if we're going to
01:00:55
bundle features into the you know Suite
01:00:57
of products and we do anti-competitive
01:00:59
stuff you have to look at the tactics on
01:01:01
the field are people cheating and are
01:01:04
they using the Monopoly power to force
01:01:06
you to use their App Store just make
01:01:08
apple have a second app store that's all
01:01:11
we're asking you to do there should be
01:01:12
an app store on iOS that doesn't charge
01:01:16
any fees or charges one percent fees
01:01:20
break the Monopoly on the App Store sax
01:01:22
is so right perfectly said she actually
01:01:24
did Issue compulsory orders to the pbms
01:01:28
so to your point sax the FTC has been
01:01:30
worried that what free Brook said had is
01:01:32
been happening but the real sort of
01:01:34
middlemen manipulator in this market are
01:01:36
the pharmacy benefit managers and so
01:01:38
this week she actually issued compulsory
01:01:40
orders to the pbms and said turn over
01:01:42
all your business records to me I'm
01:01:44
going to look into them that makes a ton
01:01:46
of sense but then on the same hand it's
01:01:48
like you see merger and you're like no
01:01:50
it can't happen it just doesn't speak to
01:01:52
a knowledge of the market we should have
01:01:54
Lena on hey Lena I know you listen to
01:01:56
the Pod I've heard the back Channel just
01:01:57
come on the Pod I should be a good guess
01:01:58
right we would have a good conversation
01:01:59
with everything yeah uh open invite
01:02:02
Nikki Haley's coming on the Pod by the
01:02:03
way you have homework to do for the
01:02:05
summit which is to see if you can get
01:02:07
Donald Trump to come to the summit okay
01:02:09
huge love J Cal love all in okay
01:02:20
because you your mannerisms are
01:02:22
unbelievable how did you practice it I
01:02:24
did I did a little bit only because I
01:02:25
like to troll people and Trigger them
01:02:27
I'm gonna really dial in my trump in the
01:02:29
coming weeks all right here we go
01:02:30
Apple's long anticipated AR headset that
01:02:33
stands for augmented reality which means
01:02:36
VR you can't see the real world you're
01:02:38
just
01:02:39
in uh in a virtual world AR lets you put
01:02:43
digital assets on the real world so you
01:02:45
can see what's happening in the real
01:02:46
world but you can put Graphics all
01:02:47
around that's expected to be revealed
01:02:50
as early as June the projected cost is
01:02:54
going to be around three thousand
01:02:56
dollars it won't ship into the fall this
01:02:57
is a break from Apple's typical
01:03:00
way of releasing products which is to
01:03:02
wait till it's perfect and to wait until
01:03:04
all consumers can afford it this is a
01:03:06
different approach they're going to give
01:03:07
this out to developers early
01:03:09
and Tim Cook is supposedly pushing this
01:03:11
there was another group of people inside
01:03:13
of Apple who did not want to release a
01:03:15
jamaat
01:03:16
but there is some sort of external
01:03:18
battery pack it seems like a bit of a
01:03:20
Franken product Frankenstein kind of
01:03:22
project here
01:03:24
that you know perhaps Steve Jobs
01:03:26
wouldn't have wanted to release but he
01:03:28
needs to get it out I think because
01:03:29
Oculus is making so much progress the
01:03:32
killer app supposedly is a FaceTime like
01:03:34
live chat experience that seems
01:03:36
interesting but they look like ski
01:03:38
goggles
01:03:40
on this as the next compute platform if
01:03:44
they can get it you know to work would
01:03:46
you wear these would they have to be
01:03:48
Prada what's the story here no and no
01:03:50
does this seem like a weird conversation
01:03:51
because none of us [ __ ] know none of
01:03:53
us have seen this product and none of us
01:03:54
have used it we were lucky friend of the
01:03:57
podpa lucky says it's incredible so what
01:03:59
that's just like commenting on one guy's
01:04:01
five letter five word tweets
01:04:03
Palmer knows I mean Palmer invented
01:04:05
Oculus great but what are we talking
01:04:07
about we have nothing to say about it
01:04:08
what do we know about okay I'll a form
01:04:09
of a really good question here do you
01:04:11
believe this is going to be a meaningful
01:04:13
compute platform in the coming years
01:04:14
because apple is so good at product how
01:04:17
do we know until we see it we gotta see
01:04:18
it
01:04:19
of course they're of course they're good
01:04:22
at product let's let's see the product
01:04:23
though like all right fine sax what are
01:04:25
your thoughts I think it's a good thing
01:04:26
that they're launching this like you
01:04:28
said it is a deviation for what they've
01:04:29
normally done they normally don't
01:04:31
release a product unless they believe
01:04:33
the entire world can use it so their
01:04:36
approach has been only to release mass
01:04:37
mass market products and have a very
01:04:39
small portfolio of those products but
01:04:42
when those products work they're you
01:04:44
know billion user home runs this
01:04:46
obviously can't be at a three thousand
01:04:47
dollar price point and it also seems
01:04:49
like it's a little bit of a early
01:04:51
prototype where the batteries are like
01:04:53
in a fanny pack around your waist and
01:04:55
there's a ways to go around this but I I
01:04:57
give them credit for launching what is
01:05:00
probably going to be more of an early
01:05:01
prototype so they could start iterating
01:05:03
on it I mean the reality is the Apple
01:05:05
watch the first version kind of sucked
01:05:06
and first five versions yeah now they're
01:05:09
on one that's pretty good I think so
01:05:10
look I think this is a cool new platform
01:05:12
they get knocked on for not innovating
01:05:14
enough I think good let them try
01:05:16
something new I think this will be good
01:05:18
for meta to have some competition yeah
01:05:21
it's great having two major players in
01:05:23
the race maybe it actually speeds up the
01:05:25
Innovation or maybe we get somewhere
01:05:27
down here I mean I I think they should
01:05:30
have
01:05:31
done something in cars
01:05:34
what should they do in cars if you were
01:05:36
going to talk about the car what would
01:05:38
it be tell me what you think would be
01:05:40
the right approach you you were going to
01:05:41
do the Facebook phone that could have
01:05:42
changed the entire Destiny of Facebook
01:05:43
they should have bought Tesla when they
01:05:45
could have the chance for four or five
01:05:46
billion they could have bought it for 10
01:05:48
billion 20 billion it's only when it got
01:05:50
to 50 60 that it got Out Of Reach
01:05:52
what do you think the car should be they
01:05:54
could have bought it at 100 billion they
01:05:55
could have bought 100 billion Tim Cook
01:05:56
famously wouldn't take the meeting Elon
01:05:58
said it he wouldn't he wouldn't be
01:05:59
bizarre bizarre maybe they missed an
01:06:02
opportunity there but I do think the end
01:06:03
game with the AR headsetter glasses
01:06:06
right yes where you get the screens and
01:06:08
you get the Terminator mode and in con
01:06:11
is that what is that these are just
01:06:13
glasses acting like they were like fancy
01:06:16
technology this this size glasses is
01:06:18
what you're talking about yeah when you
01:06:20
have like a little camera built in and
01:06:22
heads-up construction with AI
01:06:24
then it gets really interesting so that
01:06:26
that's the end game here I think give
01:06:27
the audience an example of what this
01:06:29
combination of AI plus AR could do when
01:06:31
you're walking around it could layer on
01:06:33
intelligence about the world you meet
01:06:35
with somebody and it can remind you of
01:06:37
their name and the last time you met
01:06:38
with them and give you a summary of what
01:06:39
you talked about
01:06:41
what action items there are
01:06:43
you could be walking in a city and it
01:06:45
could tell you it knows you like Peking
01:06:47
dock it could show you hey there's a
01:06:48
peking dog place over here some reviews
01:06:49
of it it just knows you and it's
01:06:51
customized in the world what about for
01:06:52
people that do the same routine 99 of
01:06:55
the time how does it gonna help you
01:06:56
though it could tell you your steps
01:06:58
every day could tell you incoming
01:07:00
messages so you don't have to take
01:07:01
yourself three thousand dollars on that
01:07:03
no but you would spend people spending
01:07:06
three years
01:07:07
to get your notifications on your wrists
01:07:10
why do you want it on your eyes for
01:07:11
three thousand I would love this maybe I
01:07:13
just do like a lot of meetings or I'm at
01:07:15
events where people are coming up to me
01:07:17
and I've met them like once a year
01:07:18
before like it would be really helpful
01:07:20
to kind of have the terminal let's be
01:07:22
honest though the Terminator mode for
01:07:24
you to be able to be present with your
01:07:27
family and friends but be playing chess
01:07:28
with Peter TL on those glasses that's
01:07:31
your dream come true you and Peter in AR
01:07:34
playing chess all day long
01:07:36
throw up the picture of sax beating
01:07:38
Peter Thiel I watched the clip from the
01:07:40
earlier all in episodes when we
01:07:42
discussed you beating Peter Thiel what a
01:07:44
great moment it was for you all right
01:07:46
listen let's wrap up with this Gallup
01:07:47
survey the number of Americans who say
01:07:50
it's a good time to buy a house has
01:07:52
never been lower 21 say it's a good time
01:07:55
to buy a house down nine percent
01:07:58
from the prior low of a year ago prior
01:08:00
to 2022 50 or more consistently though
01:08:03
it was a good time to buy significantly
01:08:04
significantly fewer expect local housing
01:08:07
prices to increase in the year
01:08:09
hey uh Sachs is this like a predictive
01:08:11
of a bottom and pure capitulation and
01:08:15
then that means maybe it is in fact a
01:08:16
good time how would you read this data I
01:08:18
don't see it as a bottom necessarily the
01:08:20
way I read the data is that the spike in
01:08:22
interest rates have made it very
01:08:24
unaffordable to buy a house right now
01:08:26
you've got you know the mortgages are
01:08:28
what like seven percent interest rate or
01:08:31
even slightly higher so people just
01:08:33
can't afford the same level of house
01:08:35
that they did before I mean mortgages
01:08:37
were at three three and a half percent
01:08:39
like a year and a half ago now I think
01:08:42
what's kind of interesting is that even
01:08:43
in the 1980s the early 1980s when
01:08:46
interest rates were at like 15 you still
01:08:50
had 50 thought it was an okay time to
01:08:52
buy a house or an attractive time to buy
01:08:53
a house so
01:08:54
for the number to be this low tells me
01:08:58
that it's not just about interest rates
01:09:00
I think consumer confidence is also
01:09:02
plummeting and people are feeling more
01:09:05
insecure so I think it's just another
01:09:09
economic indicator that things are
01:09:11
looking really shaky right now and I'll
01:09:13
tell you one of the the knock-on effects
01:09:15
of this is going to be that people can't
01:09:17
move because in order to move you have
01:09:19
to sell your current house and then buy
01:09:21
a new one and you're not going to want
01:09:23
to sell your current house when prices
01:09:25
are going down and then for the new one
01:09:27
you're going to lose your three percent
01:09:29
mortgage enough to get a new one at
01:09:30
seven percent so you're not going to buy
01:09:32
anything like the house so it freezes
01:09:35
the market it freezes Mobility I think
01:09:37
over the last few years during covid you
01:09:39
saw tremendous movement between states I
01:09:42
think that's going to slow down a lot
01:09:43
now because people just can't afford to
01:09:45
trade houses
01:09:46
so as a result of that I think
01:09:48
discontent is going to rise because I
01:09:51
think one of the ways that you create a
01:09:52
pressure valve is when people are
01:09:54
unhappy in a state they just move
01:09:56
somewhere else well now they're not able
01:09:57
to do that well and you can also move to
01:09:58
a better opportunity for you and your
01:10:00
family whether that schools taxes a job
01:10:02
lifestyle so yeah you can you can you're
01:10:05
going to reduce joy in the country and
01:10:07
it also it screws with price Discovery
01:10:09
doesn't it if you if you don't have a
01:10:11
fluid Market here then how does anybody
01:10:13
know what their house is worth and then
01:10:15
this just again creates more I think of
01:10:17
a frost I think Friedberg has said this
01:10:19
a couple times freeberg you can correct
01:10:21
me if I'm wrong but like the the home is
01:10:23
like the disproportionate majority of
01:10:26
most Americans wealth right mm-hmm
01:10:29
it's all their wealth all their wealth
01:10:31
yeah so
01:10:33
I mean
01:10:35
there's that factoid yeah and then what
01:10:37
does that do foreign
01:10:41
what's going on they're bringing you
01:10:42
lunch no I was looking I was looking at
01:10:45
uh
01:10:46
mansion that's for sale
01:10:51
175 million dollars but they just got
01:10:53
the price to 140 so I'm just taking a
01:10:55
little again I mean there's gonna be a
01:10:57
lot of distress in the market soon I'm
01:10:59
predicting a lot of distress actually
01:11:00
can we shift to the commercial side for
01:11:02
a second I just passed away yeah Sam
01:11:05
Zell passed away today oh wow rest in
01:11:07
peace yeah rest in peace uh Chicago uh
01:11:10
um yeah crazy but speaking of bombastic
01:11:13
interesting guy yeah but speaking of the
01:11:15
real estate market so I want to give an
01:11:16
update on San Francisco cre I was
01:11:19
talking to a broker the other day and so
01:11:23
here are the stats that they gave me so
01:11:25
it was a local broker and someone from
01:11:27
Blackstone and they're fans of the Pod
01:11:29
and just came up to me and we started
01:11:31
talking about what's happening in San
01:11:32
Francisco shout out shout out to them
01:11:34
didn't take a didn't take a photo but
01:11:36
but any event they're they're fans of
01:11:38
the pot of so we start talking about
01:11:39
what's happening in San Francisco real
01:11:41
estate
01:11:42
so the SF office Market is just a level
01:11:45
set is 90 million square feet they said
01:11:47
the vacancy rate is now 35 so that's
01:11:49
over 30 million square feet vacant and
01:11:52
vacancy still growing as Lisa's end and
01:11:55
companies shed space because some of
01:11:56
that space that they're not using is not
01:11:59
up for sublease everyone says what about
01:12:00
AI is AI going to be the Savior the
01:12:02
problem is that AI companies are only
01:12:05
that's only about a million square feet
01:12:07
of demand so one million out of 30
01:12:09
million is going to be absorbed by Ai
01:12:12
and you know maybe that number grows
01:12:14
over time over the next five ten years
01:12:15
as we create some really big AI
01:12:17
companies but it's just not going to
01:12:19
bail out San Francisco right now the
01:12:21
other thing is that you know VC backed
01:12:24
startups are very demanding in terms of
01:12:26
their tenant improvements and landlords
01:12:28
don't really have the capital right now
01:12:29
to put that into the buildings and
01:12:31
storms just are not the kind of credit
01:12:33
worthy tenants that landlords really
01:12:36
want so this is not going to bail
01:12:38
anybody out there they said there are a
01:12:40
ton of zombie office Towers especially
01:12:42
in Soma and all these office towers are
01:12:45
eventually going to be owned by the
01:12:46
Banks which you have to liquidate them
01:12:48
and then we're going to find out that
01:12:49
these loans that they made are gonna
01:12:51
have to be written off because the
01:12:53
collateral that they thought was Blue
01:12:55
Chip that was backing up those loans is
01:12:57
not so blue chip anymore
01:12:59
so I think we've got not just a huge
01:13:01
commercial real estate problem but it's
01:13:03
going to be a big banking problem as
01:13:05
basically people stop pretending you
01:13:07
know right now they're trying to
01:13:08
restructure loans it's called pretend
01:13:10
and extend you reduce the rate on the
01:13:12
loan but add term to it but that only
01:13:15
works for so long if this keeps going if
01:13:17
the market keeps looking like this I
01:13:19
think we're gonna have a real problem
01:13:20
and and that will be a problem in the
01:13:22
banking system now San Francisco is the
01:13:24
worst of the worst but they said that
01:13:25
New York is similar and all these other
01:13:27
big cities with empty office towers are
01:13:29
directionally I'm in New York right now
01:13:31
for the side connections conference and
01:13:33
uh
01:13:34
it is packed the city is packed getting
01:13:38
anywhere there's gridlock you can't walk
01:13:41
down the street you got to walk around
01:13:42
people every restaurant it is dynamic
01:13:45
and then I talk to people about offices
01:13:46
and they said people are staying in
01:13:48
their houses in their tiny little New
01:13:50
York apartments instead of going three
01:13:52
train stops to their office they go to
01:13:54
the office one or two days a week unless
01:13:55
you're like JP Morgan or some other
01:13:57
places that's that Drop the Boom
01:13:59
but there's a lot of people still
01:14:01
working from home the finance people
01:14:03
have all gone back media people are
01:14:04
starting to go back so there are three
01:14:06
to five days here and the city is
01:14:09
booming
01:14:10
contrast that I spent the last two weeks
01:14:12
in San Francisco walking from Soma to
01:14:14
the Embarcadero back
01:14:16
dead nobody in the city
01:14:19
it like literally a ghost town it's a
01:14:21
real shame it's a real real shame I
01:14:24
wonder if these this is the question I
01:14:27
have for you sacks can they cut a deal
01:14:30
can they go to like month to month rent
01:14:32
sublets you know loosey-goosey just give
01:14:35
people any dollar amount to convince
01:14:37
them to come back is there any dollar
01:14:39
amount because I'm looking for a space
01:14:40
for the incubator
01:14:41
in San Mateo I've been getting a ton of
01:14:43
inbound but the prices are still really
01:14:45
high and I'm like how do I cut a deal
01:14:47
here because shouldn't people be
01:14:49
lowering the prices dramatically or are
01:14:53
they all just pretending or will I get a
01:14:55
lower Ritz or definitely coming down big
01:14:57
time especially for space that sort of
01:14:59
commodity and not that desirable but
01:15:00
what's happening is according to the
01:15:02
people I talk to is that the demand the
01:15:05
people who actually are looking for new
01:15:06
space they only want to be in the best
01:15:08
areas and they want to be in the newest
01:15:10
buildings that have the best amenities
01:15:11
and so uh that sort of commodity office
01:15:14
tower where there's barely anybody ever
01:15:16
there like no one wants that um people
01:15:19
would rather pay a higher rent I mean
01:15:21
the rent will still be much lower
01:15:23
probably half the price of what it used
01:15:24
to be but they'd rather pay a little bit
01:15:26
more for that than get like a zombie
01:15:28
office tower we can't talk about all
01:15:29
this without talking about two cases
01:15:32
tragically
01:15:35
a shoplifter a criminal who was stealing
01:15:38
from a drugstore in San Francisco I got
01:15:41
shot
01:15:42
and the video was released I'm sure
01:15:44
you've seen in sacks and then here in
01:15:45
New York everybody's talking about this
01:15:47
one instance of a Marine
01:15:49
trying to subdue
01:15:52
a violent homeless person with two other
01:15:55
people and it's on everybody's Minds
01:15:57
here and Brooke Jenkins is not
01:16:00
Prosecuting in San Francisco the shooter
01:16:03
they look like
01:16:04
you know a clean shoot as they would say
01:16:06
in the police business an appropriate
01:16:09
and it's tragic to say it is but the
01:16:11
person did charge the security guard the
01:16:14
security guard did fear for their life
01:16:15
and shot him so Brook Jenkins is not
01:16:17
going to pursue anything but in New York
01:16:20
City they're pursuing manslaughter for
01:16:21
the person who did seem a bit excessive
01:16:23
from the video it's hard to tell what
01:16:25
the reality is in these situations any
01:16:27
thoughts on it David these two cases in
01:16:29
Two Cities
01:16:31
yeah look I mean the only time you can
01:16:33
get a source da excited about
01:16:35
Prosecuting someone is when they act in
01:16:37
self-defense or defensive others I mean
01:16:39
this Marine I guess Daniel penny is his
01:16:42
name he was acting in defense of others
01:16:44
the person who he stopped
01:16:47
was someone with an extensive criminal
01:16:49
record who had just recently engaged in
01:16:54
a attempted kidnapping who had punched
01:16:56
elderly people
01:16:57
had pretty gnarly dozens of arrests in
01:17:00
fact people on Reddit were talking about
01:17:03
how dangerous this person was apparently
01:17:05
a dozen years ago or so he was seen as
01:17:09
more of like a quirky like Michael
01:17:11
Jackson impersonator street performer
01:17:14
street performer but something happened
01:17:16
this is according to a Reddit post that
01:17:18
I saw where something happened and there
01:17:20
was some sort of psychological break and
01:17:22
then since then he's had dozens and
01:17:24
dozens of primes and they just keep
01:17:27
letting him loose through this revolving
01:17:30
door of a justice system we have and now
01:17:32
look no one likes to see him basically
01:17:35
dying and yeah it's too bad it's
01:17:38
horrible that that happened tragic I
01:17:40
don't know though that if you're trying
01:17:42
to stop someone I don't know how easy it
01:17:45
is to precisely control whether you use
01:17:48
too much force or not so I think Daniel
01:17:50
penny has a strong case that he was
01:17:53
acting in self-defense and defense of
01:17:55
others and there were two other people
01:17:56
by the way who were holding this person
01:17:57
down there were three of them
01:17:59
restraining him and what universally New
01:18:01
Yorker said to me of all different
01:18:04
backgrounds was this is not a race issue
01:18:06
the other I think one or two of the
01:18:08
other people were people of color it was
01:18:09
not a race issue and they're trying to
01:18:11
make it into a race issue in both these
01:18:12
cases and it's this is literally what
01:18:15
happens it's just having been through
01:18:17
this in New York in the 70s and 80s when
01:18:18
you do so who's they who's they when you
01:18:21
say trying to make it a bunch of
01:18:22
protests on the street both in San
01:18:24
Francisco
01:18:25
and New York people protesting these as
01:18:27
you know justice issues the fact is if
01:18:30
you do not
01:18:31
if you allow lawlessness for too long a
01:18:33
period of time you get a Bernie gets
01:18:36
situation and Bernie gets people can
01:18:38
look it up in the 80s I was a kid when
01:18:40
it happened but they tried to mug
01:18:42
somebody he had a gun he shot him and
01:18:45
like this is what happens if you allow
01:18:46
lawlessness for extended periods of time
01:18:49
it's just you're basically gambling and
01:18:52
what happened to Bernie Getz he got
01:18:54
knocked guilty the case he got not
01:18:56
guilty but I think he had an illegal gun
01:18:58
so he was guilty of that the Bernie gets
01:19:00
thing was really um crazy because at the
01:19:04
time the climate in New York in this
01:19:06
1984 shooting
01:19:10
there was a portion of people who I
01:19:12
don't want to say they made him a hero
01:19:14
but they made it a c this is what
01:19:17
happens if you allow us to be assaulted
01:19:19
forever we're going to fight back at
01:19:21
some point that was the vibe in New York
01:19:23
when I was a child that was 14 15 years
01:19:24
old when this happened he was charged
01:19:26
with attempted murder assault what was
01:19:28
the name of that vigilante group that
01:19:30
used to walk the streets to something
01:19:31
angels that was the um Guardian Angels
01:19:34
Guardian Angels so it was so bad in the
01:19:36
80s and I actually almost signed up for
01:19:38
the guardian angels I went to their
01:19:39
headquarters because I was practicing
01:19:40
martial arts and I thought I would check
01:19:42
it out and um they had their office in
01:19:44
Hell's Kitchen I didn't wind up joining
01:19:46
but what they would do is they would
01:19:47
just ride the subway they would wear
01:19:49
a certain type of hat
01:19:51
and wear a guardian angel shirt and all
01:19:53
they did was arrested the Subways a red
01:19:56
beret and they would just ride the
01:19:57
subways
01:19:58
and you felt kind of martial arts were
01:20:00
you taking Taekwondo I was in Taekwondo
01:20:02
yeah this is before a mixed martial arts
01:20:04
but they just rode the Subways and
01:20:06
honestly I've been on the Subways with
01:20:08
them many times you felt safe and it
01:20:10
wasn't Vigilantes they were garden
01:20:12
angels they used that term and many
01:20:14
times they would do exactly what this
01:20:16
Marine did which is try to subdue
01:20:18
somebody who was committing crime I was
01:20:20
I had
01:20:21
two distinct instances where people
01:20:23
tried to mug me you know riding the
01:20:25
Subways in New York in the 80s
01:20:27
two distinct times and one was a group
01:20:29
of people and one was one person like it
01:20:31
was pretty scary
01:20:33
both times I navigated it but it was
01:20:35
yeah not pleasant in the 80s in New York
01:20:38
can I say one more thing about this
01:20:39
Daniel Penny Jordan Neely case so look
01:20:42
at the end of the day this is going to
01:20:43
be litigated I don't know all the
01:20:45
details they're gonna have to litigate
01:20:46
whether Daniel Penny's use of force was
01:20:48
was excessive or not but but here's the
01:20:50
thing is that the media has been falsely
01:20:54
representing Jordan Neely by only
01:20:56
posting 10 year old photos of him and
01:20:58
leaving out crucial information this was
01:20:59
a press report so again this is why I
01:21:02
mentioned the whole Michael Jackson
01:21:03
impersonator thing is that the media
01:21:05
keeps portraying nearly as this innocent
01:21:06
harmless guy who is this like delightful
01:21:09
Michael Jackson impersonator in truth he
01:21:12
hasn't done that in more than a decade
01:21:14
because again he had some sort of mental
01:21:15
break and since then he's been arrested
01:21:18
over 40 times including for attempting
01:21:20
to kidnap a seven-year-old child
01:21:23
and so the media is not portraying this
01:21:27
case
01:21:28
I think in an accurate way and I think
01:21:30
as a result of that it leads to pressure
01:21:33
on the D.A to prosecute someone who has
01:21:36
I think a strong self-defense claim or
01:21:38
you know maybe the D.A just wants to do
01:21:39
this anyway and it gives the D.A cover
01:21:42
to do this Soros is I mean I know that
01:21:45
we have this back and forth with this
01:21:46
why is CNN being inaccurate do you think
01:21:48
sucks
01:21:49
they're basically cooperating with Alvin
01:21:51
Bragg's interpretation of the case and
01:21:52
they're trying to make the case against
01:21:54
penny look as damning as possible yeah
01:21:57
why don't they just take it straight
01:21:58
down the middle it's a tragedy we have a
01:22:00
screwed up situation here we got a
01:22:02
mental health crisis and it's a tragedy
01:22:04
for everybody involved on the Bernie get
01:22:05
stuff he served eight of a 12 month
01:22:08
sentence for the firearm charge and he
01:22:10
had a massive 43 million dollar civil uh
01:22:13
judgment against him
01:22:14
in 1996 decade later it's just this is a
01:22:18
little different than the the Getz thing
01:22:20
because pulling out a gun and shooting
01:22:22
somebody well yeah no that's deadly
01:22:23
intent yeah yeah that's that's the
01:22:25
congratulations Penny he's a he's a
01:22:28
trained Marine right he's trying to
01:22:30
immobilize him he has to believe that he
01:22:32
he's just trying to subdue yes Neely and
01:22:35
so using a Chokehold to kill him that's
01:22:37
an unfortunate consequence of what
01:22:39
happened but he was trying to restrain
01:22:41
the guy as far as we know right as far
01:22:44
as we know yeah I mean tragedies all
01:22:47
around we got to have Law and Order I I
01:22:49
tweeted like I don't know why we still
01:22:51
have the post office maybe we can make
01:22:52
that like once a week and redo all of
01:22:54
that space and allow every American
01:22:56
who's suffering from mental illness to
01:22:58
check in to what used to be the post
01:23:00
office
01:23:01
you know maybe like once a week and
01:23:03
obviously you can give those people very
01:23:05
gentle Landings but I don't think we
01:23:06
need Postal Service more than once or
01:23:07
twice a week and then let
01:23:09
you know let's reallocate some money
01:23:11
towards mental health in this country
01:23:13
where anybody who's sick who feels like
01:23:16
they're violent or feels like they're
01:23:17
suicidal can just go into a publicly
01:23:20
provided facility and say I'm a sick
01:23:21
person please help me this would solve a
01:23:23
lot of problems in society we've got a
01:23:26
mental health crisis we should provide
01:23:27
Mental Health Services to all Americans
01:23:29
and it's a obviously easy thing for us
01:23:31
to afford to do and if we had done that
01:23:33
then this never would have happened
01:23:34
exactly I mean literally you have sacks
01:23:37
who wants to balance the budget saying
01:23:39
hey this is something we're spending on
01:23:40
we can all agree on this compared to the
01:23:42
impact on society I don't think it would
01:23:44
be a huge expense we would save money
01:23:45
we'd save money because a city like San
01:23:48
Francisco could become quite livable or
01:23:50
New York if and then if God forbid these
01:23:52
terrible school shootings you know if
01:23:54
you avoid even one of them it's 30
01:23:55
people's lives or 10 people's lives
01:23:59
convert post offices what we need to do
01:24:01
is stand up scaled shelters and it
01:24:03
doesn't need to be done on the most
01:24:04
expensive land in a given city outside
01:24:07
of cities
01:24:08
there is no expectation in Europe for
01:24:11
like Paris or London to be affordable or
01:24:14
Hong Kong to be affordable there are
01:24:15
affordable places 30 minutes outside of
01:24:17
those places where you could put these
01:24:19
facilities I just want to ask one
01:24:21
question to sax because I don't know
01:24:23
and I know Saks is a little bit deeper
01:24:25
into the center what is George soros's
01:24:27
motivation for putting in these Lawless
01:24:31
insane DA's like I understand that he
01:24:34
was able to buy them they're low cost
01:24:36
there's not a lot of money in them okay
01:24:37
I understand that that's table Stakes
01:24:39
but what is his actual motivation for
01:24:41
causing chaos in cities listen we can't
01:24:43
know exactly what his motivation is but
01:24:45
what he did is he went into cities where
01:24:48
he doesn't live
01:24:50
and flooded The Zone with money to get
01:24:52
his preferred candidate elected his D.A
01:24:54
now the reason he did that was to change
01:24:55
the law and the way that he changed the
01:24:57
law was not through legislatures the way
01:24:59
you're supposed to operate but rather by
01:25:01
abusing prosecutorial discretion so in
01:25:03
other words once he gets his Soros D.A
01:25:05
elected they can change the law by
01:25:07
deciding what to prosecute and what not
01:25:09
to prosecute right and that's why there
01:25:11
is so much lawlessness in these cities
01:25:12
but that there's a better path you're
01:25:14
saying yeah this is not the only way
01:25:16
that Soros has
01:25:18
I'd say imposed his values on cities
01:25:23
that he doesn't live in where does he
01:25:25
live I think he's a New York guy but I'm
01:25:26
not sure but but he's gone far beyond
01:25:29
that obviously in these elections but
01:25:31
also he's done this across the world
01:25:34
Soros has this thing called the open
01:25:36
societies Foundation which sounds like
01:25:38
it's spreading democracy and liberal
01:25:39
values but in fact is fermenting regime
01:25:42
change all over the world and he's been
01:25:45
sponsoring and funding color revolutions
01:25:47
all over the world now if you like some
01:25:49
of the values he's spreading then maybe
01:25:51
you think that's a good thing but I can
01:25:53
tell you that the way this is perceived
01:25:55
by all these countries all over the
01:25:57
world is it creates tremendous
01:26:00
dissension and conflict and then they
01:26:03
look at America and they basically say
01:26:04
you know this American billionaire is
01:26:06
coming into our country and he's funding
01:26:08
regime change and it makes America look
01:26:12
bad now he's doing this I think with
01:26:14
the cooperation of our state department
01:26:16
a lot of cases and maybe the CIA I don't
01:26:19
know but this is why America frankly has
01:26:21
hated all over the world as we go
01:26:22
running around meddling in the in the
01:26:25
internal affairs of all these countries
01:26:27
too is this guy all there like that was
01:26:29
the other thing I heard is that he's not
01:26:30
all there and the people around him are
01:26:31
doing these kind of things in his
01:26:32
organizations I heard something similar
01:26:34
is that it's the idiot son Alexander
01:26:36
who's really now pulling the strings
01:26:38
would you would you allow source to
01:26:40
speak at all in Summit would you
01:26:41
interview yeah sure yeah let's have
01:26:43
Sorrows or a son and they could explain
01:26:44
themselves if they're so proud
01:26:45
apparently there is an article that
01:26:47
Alexander Soros has visited the White
01:26:49
House like two dozen times during the
01:26:51
Biden presidency this is an extremely
01:26:53
powerful and connected person I mean I'm
01:26:55
sure he listens to pod okay we'll see
01:26:57
you all next time this is episode one
01:26:59
two nine of all in we'll see you on
01:27:02
episode 130 bye bye we love you bye-bye
01:27:06
to let your winners ride
01:27:09
Rain Man
01:27:10
[Music]
01:27:13
we open source it to the fans and
01:27:16
they've just gone crazy with it
01:27:20
[Music]
01:27:22
besties
01:27:25
[Music]
01:27:30
release
01:27:33
[Music]
01:27:55
[Music]
