00:00:00
Cham who are you giggling with are you
00:00:01
with your kids what's going on here
00:00:03
you're you're making goog I got I got a
00:00:05
message from that she's like is it hel
00:00:08
no no I got a message from that she's so
00:00:10
funny so you know we I was like blah
00:00:14
blah blah she did a Uranus joke
00:00:17
into my chat and I lost it it is like a
00:00:21
Freo says your anus is right there at
00:00:24
the edge of the UN she's like no she's
00:00:27
like are you hungry I could eat a Uranus
00:00:30
oh no we could go to a great restaurant
00:00:32
Urus they serve great meals there they
00:00:36
have a great chocolate
00:00:39
lava let your winners
00:00:46
ride and instead we open sources to the
00:00:48
fans and they've just gone crazy with
00:00:51
iten
00:00:54
of all right everybody welcome back to
00:00:56
your favorite podcast of all time it's
00:01:00
episode 169 of Allin with me again the
00:01:04
chairman dictator jamath ptia David
00:01:07
freeberg your Sultan of Science and the
00:01:10
Rainman yeah definitely David Sachs has
00:01:13
his Monclair hat back again I guess
00:01:15
they're back in stock how we doing boys
00:01:17
welcome back to the show let's get to
00:01:19
the docket here issue one Elon has sued
00:01:22
open AI begun the meme Wars have after
00:01:26
we finished the recording last week the
00:01:28
memes are incredible
00:01:30
after we finished recording last week
00:01:31
Elon sued Sam Waltman Greg Brockman and
00:01:34
the open AI organization he's suing for
00:01:36
breach of contract fiduciary duty and
00:01:39
unfair competition and his argument is
00:01:41
basically open AI started as an open
00:01:43
source nonprofit as you know he he gave
00:01:45
them something like 50 or 75 million if
00:01:47
both those numbers quoted and then of
00:01:49
course everybody knows open AI turned
00:01:51
into closed AI they became a closed
00:01:53
Source for-profit Venture after the tech
00:01:56
was
00:01:57
deployed this enabled them obviously to
00:01:59
to benefit from tax uh nonprofit tax
00:02:02
breaks while building the tech but they
00:02:05
now have two corporate entities this is
00:02:07
nonprofit called open AI there's a
00:02:09
for-profit it's called open AI Global
00:02:11
LLC that was created in
00:02:12
2019 and this all this funky
00:02:15
relationship between the two we'll get
00:02:16
into that because it's kind of
00:02:17
interesting actually Elon said that if
00:02:19
open AI is allowed to do this then it
00:02:21
should be the standard for every company
00:02:22
going forward that's an interesting
00:02:23
point you can start by donating money to
00:02:25
a nonprofit and then make a for-profit
00:02:27
seems pretty C Capital efficient so what
00:02:29
do Elon want to get out of this
00:02:30
according to the lawsuit he wants open
00:02:32
AI to open source their models by the
00:02:34
way that's what Facebook and Appo are
00:02:35
doing right now so that that seems more
00:02:37
than reasonable since that's how the
00:02:39
company was formed and he wants to make
00:02:41
sure that shareholders receive no
00:02:43
Financial benefit from open
00:02:46
Ai and we can get more into a bunch of
00:02:49
the open AI nonprofit status and and
00:02:51
their structure it's super convoluted
00:02:54
but I want to just get your initial
00:02:56
reaction sacks I haven't read the legal
00:02:59
filings Jason my knowledge of the case
00:03:01
is just coming from following what's on
00:03:02
Twitter and the last time I commented on
00:03:04
one of
00:03:05
elon's lawsuits on this pod I ended up
00:03:08
in six hours of deposition
00:03:11
so remember that was the the whole
00:03:13
Twitter lawsuit in any event my
00:03:16
understanding of the case from
00:03:18
the tweets that are flying back and
00:03:21
forth is I think Elon is making two
00:03:23
points one is that when open AI was set
00:03:27
up it was set up to to be a nonprofit
00:03:31
and to promote AI as an open source
00:03:34
technology that everyone could benefit
00:03:36
from and no one large tech company would
00:03:39
control in that case Elon was primarily
00:03:42
concerned about Google I think now he's
00:03:44
more concerned about Microsoft
00:03:45
nonetheless the idea was that this would
00:03:47
be open
00:03:48
source so I think Point number one is
00:03:50
Elon feels like the rug was pulled out
00:03:53
from under him after he donated 40
00:03:56
something million dollars to this they
00:03:58
completely changed what it was going to
00:03:59
be
00:04:00
and I think he's used the word swindled
00:04:02
before he feels like he was cheated and
00:04:05
I think usually when there's a lawsuit
00:04:06
like this um it's usually because
00:04:08
somebody does feel phenomenally cheated
00:04:10
by what happened so I think that's the
00:04:13
first point the second is that I think
00:04:15
that elon's making is well wait a second
00:04:17
if you can start a company as a
00:04:20
nonprofit using pre-tax dollars and then
00:04:23
all a sudden convert to for-profit then
00:04:25
why wouldn't everybody do this in order
00:04:27
to circumvent paying taxes and I think
00:04:29
that's another I'd say valid and
00:04:31
interesting point that that he's made
00:04:33
now openai has responded to this by
00:04:35
publishing some of elon's emails to them
00:04:38
and they think that they have a Smoking
00:04:39
Gun here because
00:04:41
apparently Elon told them they need to
00:04:43
raise far more money in order to have a
00:04:45
chance of taking on Google deepmind I
00:04:49
just don't know whether that's the
00:04:50
Smoking Gun they claim it is but that's
00:04:52
that's basically what they're putting
00:04:53
out there and then to add fuel to the
00:04:56
fire you've got VCS coming over the top
00:04:59
debating different aspects of this
00:05:00
you've got sort of venod Defending open
00:05:03
AI he was the first VC investor in it
00:05:06
and so he's been out there attacking
00:05:08
Elon and then I think Mark Andre is
00:05:12
responding to venod and may I know if
00:05:15
he's exactly defending Elon anyway it's
00:05:17
turned into a whole melstrom on X and
00:05:20
then everyone else is start starting to
00:05:23
put out a whole bunch of crazy memes so
00:05:25
I don't know who's gonna win this case
00:05:27
in court but the memes are
00:05:30
definitely lit and U maybe the funniest
00:05:32
one is that Elon has said that uh if
00:05:36
open AI will simply changed their name
00:05:38
to closed AI he'll drop the lawsuit I
00:05:40
think he's serious about that I think he
00:05:42
is serious about it but I think he's
00:05:44
making his point which is you ended up
00:05:47
doing something different than what you
00:05:49
told me when you got me to write this
00:05:51
big donation and when we co-founded this
00:05:54
thing together so I think that sort of
00:05:56
sums up elon's position on this
00:06:00
Jam a year from now where do you think
00:06:02
we'll be with this well I think I think
00:06:04
there's one more thing which is
00:06:07
that they used his name pretty
00:06:11
aggressively to get more money and he
00:06:13
was very instrumental in getting some of
00:06:16
the early critical hires particularly
00:06:18
Ilia which has been I think well
00:06:20
documented SS to leave Google right so
00:06:25
that's probably sacks where also some of
00:06:27
this
00:06:28
feeling that he was Bamboozled comes
00:06:32
from but I think that all of those
00:06:35
emotions matter less than the rule of
00:06:38
law which is his second point which is
00:06:40
the important Point irrespective of
00:06:42
whatever one email says or another email
00:06:44
it's not great for the US tax system if
00:06:47
all of a sudden a big gaping loophole is
00:06:50
identified and taken advantage of I
00:06:52
think that there's a huge economic
00:06:55
incentive for the State of California
00:06:58
every other state where these open AI
00:06:59
employees lived and have gotten equity
00:07:01
and now have gotten paid there's an
00:07:04
incentive for treasury there's an
00:07:06
incentive for the
00:07:07
IRS it touches a lot of aspects of very
00:07:11
complicated tax law and so I think
00:07:14
that's why there will be a resolution to
00:07:16
this case because I think that that
00:07:19
answer is very important and I think it
00:07:21
will as he has correctly
00:07:23
identified motivate a lot of people's
00:07:26
actions going forward so independent of
00:07:29
what the emotional situation is amongst
00:07:31
all the all the actors in this play the
00:07:34
reality is he's identified a loophole
00:07:36
and that loophole needs to get fixed
00:07:37
I'll give you a a different example of
00:07:39
this that's much more benign but it
00:07:41
dragged on for eight or nine years at
00:07:44
Facebook there was a period where and
00:07:46
this was public so I can I can tell you
00:07:50
we transferred a lot of our IP to
00:07:52
Ireland at one moment and there was a
00:07:54
transfer payment and whatnot and then a
00:07:56
few years afterwards when when every
00:07:59
everybody realized including us and we
00:08:01
had miscalculated but when everybody
00:08:02
realized the value of Facebook that
00:08:05
transfer payment did not seem correct
00:08:07
and it was a huge tax ARB that we had
00:08:10
facilitated right because all of that IP
00:08:13
sitting inside of Ireland gets taxed at
00:08:16
you know a very different rate than that
00:08:18
IP would have gotten taxed in the United
00:08:20
States and other companies had copied
00:08:21
this long story short the IRS Su you
00:08:24
know similar to you David I was in years
00:08:27
of depositions and interviews and all of
00:08:29
this stuff so the point is that the
00:08:31
government really cares about these
00:08:33
kinds of things because so much money is
00:08:35
on the line and if open aai turns out to
00:08:38
be this multi hundred billion dollar
00:08:40
Behemoth this will get figured out in
00:08:43
court because there's just too much
00:08:45
money at stick freeberg any thoughts and
00:08:48
maybe Steelman if if you feel like it
00:08:51
open eyes position I think we're all
00:08:55
ignoring the documents that open AI put
00:08:58
out yesterday showing emails and
00:08:59
interactions with Elon where Elon
00:09:02
acknowledged and recognized the
00:09:04
necessity of having a for-profit
00:09:07
subsidiary that could pursue the
00:09:10
interests of the
00:09:11
foundation by raising billions of
00:09:13
dollars of outside
00:09:15
capital I think it's a really
00:09:18
interesting set of facts that provides a
00:09:22
different light on the story and it was
00:09:23
really important that open AI released
00:09:25
it I'm going to share a hold on fre I
00:09:28
did mention that freeberg I mentioned
00:09:30
that too my point is it doesn't allow
00:09:31
you to break the law yeah so let me just
00:09:33
um tell you guys uh a little bit about
00:09:36
another example my sister works at a
00:09:39
nonprofit called the Cystic Fibrosis
00:09:40
Foundation cystic fibrosis is an
00:09:43
inherited disorder affects the lungs and
00:09:45
digestive
00:09:46
tract it's debilitating really affects
00:09:51
children there was a nonprofit called
00:09:53
the Cystic Fibrosis Foundation
00:09:54
established in 1955 they did tons of
00:09:56
work on drug development research and it
00:09:58
was a nonprofit profit for years until
00:10:00
they realized that there needed to be a
00:10:01
market-based system to create the
00:10:03
necessary incentives to drive the
00:10:04
capital needed to find a drug that could
00:10:07
cure cystic
00:10:08
fibrosis and in the year 2000 they
00:10:11
invested out of their Foundation
00:10:13
endowment $40 million in a company
00:10:16
called Aurora biosciences subsequently
00:10:18
it was acquired by vertex and they
00:10:20
continued to invest another $60 million
00:10:22
so total of 100 million bucks in the
00:10:24
lifetime of the development of a drug
00:10:26
that could cure this
00:10:27
disease in 2005 a drug was discovered
00:10:32
and in 2012 the FDA approved it the
00:10:35
nonprofit then sold their rights their
00:10:37
interest in this for-profit
00:10:40
entity for $3.3
00:10:42
billion it was an incredible return it
00:10:44
was the largest wow investment return by
00:10:47
a nonprofit in history who did they sell
00:10:50
it to back to the company I actually
00:10:53
think they sold the rights to royalty
00:10:54
Pharma who you and I know well and that
00:10:57
$3.3 billion continued uh opened up this
00:11:00
ability for them to continue to make
00:11:02
investments they now fund biotech VCS
00:11:04
and they make direct Investments and
00:11:05
other things but it really set the
00:11:07
Benchmark for this concept of what's
00:11:08
called Venture philanthropy where a
00:11:10
nonprofit parent company can make
00:11:12
investments in for-profits that can
00:11:15
raise additional Capital that's needed
00:11:17
to pursue the broad difficult interests
00:11:20
of the nonprofit and I think that this
00:11:22
argument really kind of ties into what
00:11:24
happened with open Ai and you can see it
00:11:26
in the email exchanges with Elon where
00:11:28
he was so impression that I I give Elon
00:11:30
extraordinary credit for this that he
00:11:31
saw that this is going to take billions
00:11:33
of dollars a year of investment to
00:11:35
realize the pursuit that open AI was
00:11:36
going after and there was no way that
00:11:38
Elon was going to be able to generate
00:11:39
that cash himself or that Reed or others
00:11:41
were going to just be able to Pony up
00:11:43
that money they needed to have some sort
00:11:45
of for-profit vehicle that would allow
00:11:47
the market to work and allow capitalists
00:11:49
to find their Capital into this
00:11:51
organization to make this this
00:11:52
investment interest so I think the real
00:11:55
question with respect to is open AI in
00:11:57
trouble as a foundation
00:11:59
is does the nonprofit own a meaningful
00:12:02
piece of the for-profit entity and I
00:12:04
don't know the answer to that I don't
00:12:05
know if you guys do sounds like it I
00:12:07
mean yeah must right so if they do then
00:12:09
they have an investment interest the
00:12:11
second question is does the nonprofit
00:12:13
parent still do charitable stuff because
00:12:16
if it doesn't then there's a real
00:12:18
question on the nonprofit status I think
00:12:20
you have to have a certain amount of
00:12:21
your assets deployed every year in order
00:12:23
to qualify for
00:12:24
501c3 so I think the test in the courts
00:12:27
will likely end up being look it's
00:12:28
totally reasonable to have a for-profit
00:12:30
entity to fund a for-profit entity other
00:12:32
nonprofits have done it particularly
00:12:34
when you need to attract billions of
00:12:35
dollars of outside Capital to make it
00:12:37
work the real question is does the
00:12:38
nonprofit still do nonprofit stuff
00:12:40
that's an interesting question because
00:12:42
freeberg do you think that when they
00:12:44
booted off the uh the nonprofit experts
00:12:50
from the board that they they may have
00:12:52
crossed a tripwire there yes yeah I'm no
00:12:54
lawyer I'm no lawyer that's a great
00:12:56
question I did a little journalism and I
00:12:58
talked to
00:13:00
a couple of people who are in and around
00:13:02
two other of these scenarios one of the
00:13:03
key issues here is that the IP in your
00:13:06
example freeberg it wasn't like that
00:13:09
cystic fibrosis organization did the
00:13:11
drug Discovery right they didn't do it
00:13:14
correct well they I think they they
00:13:17
handed over some some research that was
00:13:18
open and public at the time and they but
00:13:21
they funded an independent forit and so
00:13:23
what happened here was you had the IP
00:13:26
and the employees of the nonprofit given
00:13:28
gied they were transferred over to the
00:13:32
for profit organization and they when
00:13:34
the Mozilla Foundation did this and
00:13:36
another company called SAS source and
00:13:39
these both have two structures a
00:13:43
nonprofit and a for-profit but in both
00:13:45
of their cases they set up separate
00:13:46
boards and the two different boards you
00:13:49
know answer to the for-profit concerns
00:13:51
and the nonprofit concerns they also
00:13:53
didn't give huge chunks of equity to all
00:13:56
of the employees which then sold them in
00:13:59
secondary so what's happened here is all
00:14:01
of the IP in all likelihood has been and
00:14:04
the wealth transfer has gone to all
00:14:06
these employees they've now done a
00:14:08
secondary and this actually triggers a
00:14:10
lot of
00:14:12
IRS investigations the Mozilla
00:14:14
Foundation which was making the Netscape
00:14:16
browser which many of you probably have
00:14:18
used over the years they were making
00:14:19
hundreds of millions of dollars a year
00:14:21
David from advertising with Google the
00:14:23
IRS said hey wait a second you look like
00:14:25
a for-profit to us so they made a
00:14:27
for-profit entity all of the and
00:14:30
Mitchell Baker and the team over there
00:14:33
set this up so all the money went back
00:14:35
into the Mozilla foundation in their
00:14:38
nonprofit efforts and they just paid
00:14:42
above Market salaries but it was very
00:14:45
important that these things be separated
00:14:47
and that the employees not have equity
00:14:48
and that all that money flowed back up
00:14:50
and I think this is where open AI is
00:14:51
going to get super tripped up and I
00:14:53
think the IRS is going to have a field
00:14:54
day here here's the Nick can you please
00:14:56
throw the image up this is from the open
00:14:58
AI
00:14:59
website that explains their structure so
00:15:03
let's just speculate for a second
00:15:04
because we're we just guessing but it
00:15:07
has been reported that Microsoft owns
00:15:10
49% right that's the number of the
00:15:13
for-profit in this example Microsoft
00:15:16
where it says minority economic interest
00:15:18
that number there would be 49 and that's
00:15:20
what they own of this open AI Global
00:15:23
LLC and the majority owner is this
00:15:26
holding company okay so that means 51%
00:15:28
is there but it says here that that 51%
00:15:32
is owned by a combination of the
00:15:34
nonprofit employees and investors so the
00:15:37
question is back to Sax's question we
00:15:40
can know very precisely how much the the
00:15:43
nonprofit owns by just exing out the
00:15:47
investors and employees and I'm going to
00:15:49
guess it's at least 30 or 40% I think
00:15:53
that that's probably a pretty reasonable
00:15:54
guess so it means that the open AI
00:15:57
Foundation probably own somewhere
00:15:59
between five to 20% is that fairly
00:16:02
reasonable
00:16:04
guess at the maximum that's seems right
00:16:07
yeah and so I think that they're going
00:16:10
to have to show that ownership structure
00:16:12
and and decompose all these entities and
00:16:14
show the the actual
00:16:16
percentages if it's a lot less than that
00:16:18
if it's say like a 1% thing then it's a
00:16:21
little bit more sketch I think but if
00:16:23
it's really like 15 or
00:16:25
20% they're still going to have to prove
00:16:28
that all of this was done in um in a
00:16:31
clean way because the the big thing that
00:16:33
these guys did was because by creating
00:16:35
this lpgp structure and these Capp
00:16:37
profits it's clear that they were trying
00:16:40
to avoid something and the question is
00:16:42
in the discovery will everybody learn
00:16:46
what it is that they were trying to
00:16:47
avoid because I've been through this I
00:16:49
think saak you've been through it
00:16:50
freeberg you were through
00:16:52
it when you set up these large vehicles
00:16:55
and pulled amounts of money folks come
00:16:57
out of the woodwork with all kinds of
00:16:59
convoluted structures and the discipline
00:17:01
is always to tell these lawyers and
00:17:02
accountants
00:17:03
no and they'll be like let's set up a
00:17:05
master feeder and you'll go through
00:17:07
Bermuda and this and that and and it's
00:17:10
so easy to say yes because it's very
00:17:12
seductive what they're selling you but
00:17:14
it's always about trying to avoid
00:17:15
something so the real question is why
00:17:18
this convoluted structure what was it
00:17:19
trying to avoid and where did this IP
00:17:23
come from like all that IP sacks came
00:17:26
from a nonprofit all those employees
00:17:27
worked for the nonprofit
00:17:29
then they decided to go from an open
00:17:30
source model to closed and then by doing
00:17:33
that they capture all the value and then
00:17:36
it's gifted to these employees and this
00:17:39
investor base that's where I think but
00:17:42
remember well it's important to remember
00:17:44
the majority of the IP that exists at
00:17:46
open AI today was generated after the
00:17:49
for-profit was set up the real question
00:17:51
was there a fair transfer at the time
00:17:53
when it was set up of value into this
00:17:55
for-profit entity and my guess is that
00:17:57
GP box that you just looked at the
00:18:00
compensation that they're earning
00:18:01
through that GP box that the nonprofit
00:18:03
earns is likely meant to be kind of
00:18:06
a a fair amount relative to what was
00:18:08
contributed at the time the real
00:18:10
question for me Still Remains What other
00:18:13
nonprofit and charitable work does the
00:18:15
open AI 501c3 actually do and if there's
00:18:19
the answer is nothing and it's just a
00:18:21
shell nonprofit and under it the only
00:18:23
thing that it owns is a for-profit
00:18:25
interest then I think there's a real
00:18:27
question on what's the activity and I I
00:18:29
don't mean to speak out of turn on all
00:18:30
this stuff I don't know enough but I
00:18:32
think that's the only sort of lens I
00:18:34
would kind of look at this stuff through
00:18:35
S going to say something well a couple
00:18:37
of points one is one of my contentions
00:18:40
has been never innovate on structure all
00:18:43
you do is create legal problems there's
00:18:45
a tried andrue way of creating a startup
00:18:48
which is a a C Corp when you try to
00:18:50
innovate on legal instead of product it
00:18:53
almost always
00:18:54
backfires the second Point here is I
00:18:57
think open AI is in a little bit of a
00:18:59
damned if you do damned if you don't
00:19:01
situation to go back to freeberg point
00:19:03
about the nonprofit sitting at the top
00:19:06
there remember what created all the
00:19:08
problems when Sam briefly got fired as
00:19:10
CEO and then came back is they had these
00:19:13
directors from the nonprofit world who
00:19:16
didn't really seem to understand how
00:19:17
startups worked and so either they fired
00:19:21
Sam for no reason which was kind of
00:19:23
incompetent or they had a really good
00:19:25
reason but didn't communicate it
00:19:27
properly which was Al incompetent either
00:19:29
way that whole thing was a it was a
00:19:31
spectacle and it just kind of showed
00:19:34
that there was this like culture Clash
00:19:35
going on between the for-profit company
00:19:39
and like the standard way of running a
00:19:41
Silicon Valley startup to maximize the
00:19:43
outcome and then this nonprofit board
00:19:45
that was sitting on top of the whole
00:19:46
thing so I think that now they're in the
00:19:50
situation where they've changed that
00:19:53
nonprofit board they've booted off the
00:19:56
nonprofit people
00:19:59
and that may have been the right thing
00:20:01
for the for-profit entity but now it
00:20:03
might get them in trouble because it
00:20:06
lends Credence to elon's lawsuit that
00:20:09
they've completely
00:20:10
changed the original Mission of of this
00:20:13
organization it was supposed to be it
00:20:16
was supposed to be open source yes and
00:20:18
it was supposed to be nonprofit and now
00:20:20
it's for-profit and closed Source think
00:20:22
about this you know and again we are
00:20:24
speculating here but if it if the
00:20:26
mission was to be open sourced
00:20:28
and then you realize you've got
00:20:29
something super valuable and you close
00:20:32
Source it create a for-profit and then
00:20:34
take all the employees and all the IP
00:20:37
and put it into the for-profit lock it
00:20:39
down break the original Mission that hey
00:20:41
this was going to help Humanity because
00:20:43
we're going to make it open source so if
00:20:44
you donate money here and we get the tax
00:20:46
exemption Humanity benefits because
00:20:49
everybody gets to look at that code at
00:20:50
the same time and work on it and iterate
00:20:53
now they they've just basically stolen
00:20:55
all that in order to enrich themselves
00:20:57
and I think think then you add to it Sam
00:20:59
maybe doing some deal making which seems
00:21:01
to be one of the triggers according to
00:21:03
reports and this open AI Venture fund
00:21:05
Etc maybe that deal making made the
00:21:07
nonprofit people say Hey listen you're
00:21:09
doing even more for-profit stuff with
00:21:10
the open AI name we're supposed to be a
00:21:12
nonprofit here what's the Venture fund
00:21:14
thing he started a venture fund to
00:21:16
invest in companies called the open AI
00:21:18
Ventures fund and he was the sole owner
00:21:21
of it which they're saying now it's a
00:21:22
clerical mistake or something but he's
00:21:24
invested in a bunch of startups that
00:21:26
have unique access ACC to the open AI
00:21:30
you know I think he's the he's the GP on
00:21:33
behalf of open the open. fund he's a no
00:21:36
but he said GP on behalf of himself or
00:21:38
on behalf of open the way the paperwork
00:21:40
was done it was on behalf of himself
00:21:42
they're now saying that that was done in
00:21:44
Era so okay fine maybe it was maybe it
00:21:47
wasn't who knows actually we're gonna
00:21:49
wait wait the open AI fund yes whose
00:21:53
money is
00:21:54
that uh Sam raised the money to create
00:21:57
an open ey fund to to invest in
00:21:59
companies using open AI software okay so
00:22:03
that's a different kind of issue that's
00:22:05
called a corporate opportunity issue
00:22:08
where open AI
00:22:10
fund it's basically a a conflict of
00:22:13
interest where the Economic Opportunity
00:22:16
of the open AI starup fund should belong
00:22:17
to open AI if Sam created a separate
00:22:20
fund with separate LPS that he's the GP
00:22:22
of and gets economics in that then
00:22:24
that's a potential usurpation of a
00:22:26
corporate opportunity I don't I I don't
00:22:29
know the truth of that matter I'm just
00:22:30
saying that's a slightly different issue
00:22:33
to go back to uses the open a name yeah
00:22:36
that that's where like the corporate
00:22:37
opportunity becomes really explicit to
00:22:40
go back to the point you were making a
00:22:42
minute ago about the employees I just
00:22:44
think and you use the word stealing I
00:22:46
think it's like a really strong word I
00:22:48
mean I think we have to complement the
00:22:50
employees of this company if it is a
00:22:53
company including Sam including Greg
00:22:56
including Ilia for creating an amazing
00:22:59
product right incred they've created
00:23:01
something incredible absolutely and
00:23:03
they've created I think an amazing
00:23:04
ecosystem and there's a lot of
00:23:06
developers building on top of this and I
00:23:08
think no one is attacking their work and
00:23:11
I and I think it's a little too harsh to
00:23:13
say that anyone's stealing anything
00:23:14
because they have created all the value
00:23:16
the question and and I think you have to
00:23:18
include Sam in that and say say that
00:23:19
he's done a great job okay as CEO
00:23:21
because they've built something amazing
00:23:23
and saxs to your point right it's not
00:23:24
stealing because very smart investors
00:23:27
are putting in money
00:23:28
Eyes Wide Open yeah I think that the
00:23:31
criticism here is related to structure
00:23:33
from my standpoint it's related to
00:23:35
structure and whether Elon was told
00:23:38
something at the beginning that's
00:23:39
different it got changed and clearly it
00:23:42
got changed in a way that was not
00:23:45
consistent with the terms under which he
00:23:47
initially contributed all this money
00:23:50
yeah I mine is I I'm stating
00:23:53
what is the most cynical interpretation
00:23:57
of what happened here there is the most
00:23:59
you know benign and benevolent
00:24:01
interpretation you could have as well
00:24:02
and and maybe the truth is in between
00:24:03
the two but the series of events that
00:24:06
occur does not look good when you have a
00:24:08
mission to give this intellectual
00:24:10
property to the world so no one person
00:24:13
benefits from it all of humanity is
00:24:15
supposed to benefit from it that was the
00:24:17
point of being open source right then
00:24:20
you close it now who gets the benefit if
00:24:22
it's closed right the employes right
00:24:26
that's a good point and that's going to
00:24:27
be the of this the EMP and and if you
00:24:29
really want the investors the investors
00:24:31
the investors that's how they were able
00:24:33
to attract billions of dollars because
00:24:34
you can't get billions of dollars great
00:24:36
to invest in something that's going to
00:24:37
be open sourced and and then the
00:24:39
employees sell two billion in
00:24:42
secondary so now you if you really want
00:24:44
to take the most cynical approach to
00:24:46
this or interpretation again this is
00:24:49
just one interpretation they took an
00:24:51
open source project they closed it they
00:24:54
raised money and then within the next
00:24:56
two years on this incredible Innovation
00:24:59
they sold $2 billion and put that in
00:25:02
their pockets now those employees if it
00:25:03
was a for-profit obviously no problem
00:25:06
with that but they took a the stated
00:25:09
mission of giving this to all of
00:25:10
humanity and then took three or four per
00:25:13
of that gain and put it in their pockets
00:25:14
that's where I think this whole thing is
00:25:16
going to get really crazy and this is
00:25:18
where the IRS Point yeah and the IRS
00:25:21
also
00:25:22
interpreted Mozilla which was just $100
00:25:24
million and none of the employees took
00:25:26
any of it the 100 million that came out
00:25:29
of Mozilla they were just saying like I
00:25:30
think you should pay tax on that should
00:25:32
you pay tax on that and then they were
00:25:34
investigated for years so I think the
00:25:36
IRS is going to be on this like crazy
00:25:38
based upon what happened to Mozilla
00:25:41
which was Jason do you think do you
00:25:42
think individual employee sellers are
00:25:45
going to get audited by the IRS because
00:25:47
of this I I don't I mean I'm know we're
00:25:50
in Uncharted Territory here when I did
00:25:52
some tweets and said hey can everybody
00:25:53
contact me who has information on these
00:25:54
structures I was able to find the some
00:25:57
Source example and they did it very
00:26:00
clean and there was no like IP transfer
00:26:03
or employees enriching themselves or you
00:26:06
know a god king like Sam doing all kinds
00:26:08
of deals and enriching himself it was a
00:26:11
very clean structure and then Mitchell
00:26:13
Baker who set up Mozilla also did the
00:26:14
same thing hey we's pay the employees an
00:26:16
extra 50k on their salaries and they'll
00:26:18
be a little overpaid in Silicon Valley
00:26:20
but they're not getting billions of
00:26:21
dollars in equity which again I have no
00:26:24
problem with people getting billion
00:26:25
dollar billions of dollars in equity in
00:26:26
a for-profit but this was supposed to
00:26:29
benefit Humanity it was supposed to be
00:26:31
open if this was open source and they
00:26:33
took the 2 billion I would actually not
00:26:35
have much of a problem with it because
00:26:36
we could all be looking at that mission
00:26:38
and then saaks and myself and Friedberg
00:26:40
and anybody who wanted to use that code
00:26:42
could go in there and adapt it and you
00:26:44
know what there are four profits apple
00:26:47
and meta which are producing open-
00:26:49
Source software so if they're producing
00:26:51
Apple's pretty competent and meta is
00:26:53
super competent in this respect they're
00:26:56
able to do open source but Sam is
00:26:58
claiming or open AI is claiming it's too
00:27:00
dangerous to show us the open source
00:27:02
code I call on that their code
00:27:04
is not too dangerous for us to see
00:27:06
that's complete utter so you
00:27:08
you think that there is a motivation for
00:27:10
profit basically what you're saying is
00:27:12
essentially that this was highly
00:27:13
motivated by
00:27:16
profit versus like even like like some
00:27:18
of the emails the
00:27:21
emails try to paint a picture of very
00:27:25
industrious people on both sides who are
00:27:27
trying to solve a very hard problem and
00:27:30
dealing with a conundrum that's really
00:27:31
around Capital right yeah we need to we
00:27:34
need to do all this training it's going
00:27:35
to be so expensive but I think what
00:27:37
you're saying which is a a deviation
00:27:40
from that is actually come on guys these
00:27:41
are all very smart people and those
00:27:43
folks saw an opportunity to make a ton
00:27:44
of money and they took
00:27:46
it I mean that seems like what happened
00:27:49
here to me I mean it's aam's Razer kind
00:27:51
of situation I think they if you look at
00:27:54
it I think they probably regretted
00:27:55
making this a nonprofit and then try to
00:27:57
figure out a way to reverse it that's
00:27:59
actually what I think is going on here
00:28:00
and I do think there is part of it chth
00:28:02
you're right that they needed servers
00:28:04
and they needed capacity but they could
00:28:06
have done that without giving the
00:28:08
employees tons of equity without selling
00:28:10
billions of dollars keeping it Ops and
00:28:11
source and then telling Microsoft hey if
00:28:14
you want access to this open source or
00:28:16
whatever and you know very you're making
00:28:19
a very very good point in and and the
00:28:22
reason is that if it was open source
00:28:26
Microsoft could have just taken it and
00:28:28
done its own training and just paid for
00:28:30
it yeah or make a donation to the
00:28:33
foundation this is interesting because
00:28:35
it's a very good point because open AI
00:28:38
yeah open AI dropped a few of the emails
00:28:40
that they had with Elon obviously those
00:28:42
are Cherry Picked to help their case the
00:28:44
most what we don't know is what do the
00:28:46
other emails show like what were their
00:28:48
alternatives to this particular
00:28:50
structure could they have raised the
00:28:51
necessary funds in a way that was more
00:28:53
consistent with the original Mission of
00:28:54
the company could they have remained
00:28:56
open source for example
00:28:58
example also are there any emails where
00:29:01
employees talk about the potential
00:29:03
benefit to them of going private going
00:29:07
for-profit right those emails have
00:29:10
already been leaked there was this thing
00:29:11
where in the open AI emails that were
00:29:14
leaked they talked about being able to
00:29:16
swap this Phantom equity and open AI
00:29:19
into either YC equity and then Elon said
00:29:21
oh maybe SpaceX to but I'll have to
00:29:23
figure that out well then also like
00:29:25
Microsoft coming in and buying 4 % and
00:29:28
then it's close this is the other piece
00:29:30
that I would want to see in Discovery as
00:29:32
well and again I don't have any problem
00:29:34
with any of the people I love Sam M I
00:29:35
think he's great I think all the people
00:29:37
there are great I think what they've
00:29:38
done for Humanity is great I just think
00:29:40
they should have kept this thing open
00:29:41
source which was the mission but then
00:29:43
they close Source at chth and then give
00:29:46
49% of it all the weights all the source
00:29:49
code to Microsoft so that to me was like
00:29:52
a really like you you want to talk about
00:29:55
taking this nonprofits IP and then it
00:29:58
some amount of that bag gets given to
00:30:00
the employees for billions of dollars
00:30:01
and then Microsoft gets 49% of all that
00:30:04
nonprofits effort to then go
00:30:05
commercialize and Microsoft has added
00:30:08
what $500 billion dollar in market cap
00:30:10
since this whole thing has been
00:30:11
announced so now right all that profit
00:30:14
has been aggregated into Microsoft stock
00:30:16
and I think this is where you got to
00:30:18
understand why Elon feels swindled is
00:30:20
because not only are we going from
00:30:23
nonprofit to for-profit and open source
00:30:25
to closed Source he was specifically
00:30:27
concerned about all the benefits of AI
00:30:30
crew to one powerful big tech company
00:30:33
now at that time he thought it was
00:30:34
Google now it's Microsoft there's not
00:30:37
really that much of a difference he
00:30:38
never wanted all the benefits of AI in
00:30:41
the hands of one really powerful tech
00:30:45
company right and Microsoft is the most
00:30:47
it's now what the biggest company in the
00:30:49
world by market cap so this is like the
00:30:51
opposite of what he intended they gifted
00:30:52
a trillion dollars to Microsoft probably
00:30:54
there's an easy solution if they are
00:30:57
people of good faith and they're doing
00:30:59
this for the right reasons open source
00:31:01
it just go back and open source it Jason
00:31:04
but to your earlier point that may solve
00:31:06
the lawsuit and Elon may drop the
00:31:07
lawsuit but it's opened a can of worms
00:31:10
with respect to tax and structuring
00:31:12
that's much bigger than just open AI
00:31:14
yeah the entire nonprofit industry it's
00:31:16
a tax problem for every new company it's
00:31:19
every other entrepreneur that studies
00:31:21
this model and tries to replicate it for
00:31:23
their own personal gain even if that
00:31:24
wasn't intentional here so there's whole
00:31:27
set of issues that we've we've really
00:31:31
cracked the egg here we got we got to
00:31:32
figure out how to unscramble there
00:31:34
there's the IRS issues then there's
00:31:36
what's morally right and then there's
00:31:37
elon's beef and if it is a for-profit
00:31:39
company and Elon put in 50 million when
00:31:41
it was a seed round what would he own
00:31:43
saxs what would his ownership in this
00:31:46
for profit be oh my God I mean probably
00:31:49
98% okay and so yeah we we need know the
00:31:53
total size of that round he put in the
00:31:54
first 40 million do we know what other
00:31:56
people put in it was most of the money
00:31:58
right yeah it was most of the money so I
00:31:59
mean if you put let's just put a crazy
00:32:02
valuation on it 500 million okay he owns
00:32:05
whatever 10% at that point at that point
00:32:08
it it would AI was not what it is today
00:32:11
so you would have raised that like you
00:32:13
would have raised 150 on 50 pre okay I
00:32:16
mean that's that's the way that that
00:32:17
that a very capex intensive deal in a
00:32:19
space that wasn't thought to yield big
00:32:22
outcomes that's unfortunate half the
00:32:24
company at least at least right so give
00:32:26
him 20% of the compan is $20 billion
00:32:29
that's even half of what you're saying
00:32:30
sack right we talked about on the show
00:32:32
and that whole fracus with these
00:32:34
nonprofit directors went down is we said
00:32:37
go back and and restructure the whole
00:32:40
thing to make it what it always should
00:32:41
have been which is just a clean
00:32:43
for-profit entity give Elon his Equity
00:32:46
give Sam his Equity because it never
00:32:48
made sense that have Sam has no equity
00:32:50
but he's got a venture firm it's it's
00:32:52
all so convoluted it's a weird form of
00:32:53
compensation in which they're giving him
00:32:57
corporate
00:32:58
opportunities in effect as like a type
00:33:00
of compensation when really he should
00:33:02
just have compensation in the
00:33:04
corporation and then the corporation
00:33:06
should own all of its opportunities oh
00:33:08
that's okay that's a really interesting
00:33:10
thing you just said so basically yeah
00:33:11
it's like he famously has no equity but
00:33:14
then he has this retained optionality to
00:33:17
monetize the ecosystem so even though
00:33:18
he's not monetizing the thing he's he
00:33:21
gets to monetize everything around and
00:33:23
it seems like the the board or at least
00:33:25
the new board's okay with that because
00:33:27
he is being undercompensated with
00:33:29
respect to the main thing so then he
00:33:30
gets the side things but that's not
00:33:32
really the way it should work either you
00:33:34
know Bill Gates said to us famously when
00:33:36
we were at when I was at Facebook the
00:33:38
value of an ecosystem is when the
00:33:40
economic value generated by the
00:33:42
ecosystem exceeds that of the platform
00:33:44
now in this case you'd actually rather
00:33:47
have 50 basis points of the ecosystem
00:33:50
than 5% of open AI if that's true
00:33:52
because if this thing could be so
00:33:53
revolutionary you're talking 10 2030
00:33:56
trillion dollars
00:33:58
what a mess oo Sam should just be given
00:34:01
like a a huge option Grant an open AI
00:34:04
but then open AI should own its own
00:34:06
Venture
00:34:07
fund yeah and the SEC is looking into
00:34:10
all this stuff you know they look into a
00:34:11
lot of things in fairness but they're
00:34:13
looking into it what a mess and uh we'll
00:34:15
keep track of it the other thing that's
00:34:17
crazy I don't know if you guys know this
00:34:19
but Nick just as we close here the most
00:34:22
insane part of open's LP investment
00:34:24
agreement which is on their website you
00:34:25
can just search for open LP agreement is
00:34:28
this part the partnership exists to
00:34:30
advance opening ey inc's mission of
00:34:31
ensuring that safe artificial general
00:34:33
intelligence is deployed and benefits
00:34:35
all of humanity the general partner duty
00:34:36
to this Mission and principles advance
00:34:38
in open aing Charter y yada y take
00:34:40
precedent over the obligation to
00:34:42
generate a profit the partnership may
00:34:44
never make a profit and the general
00:34:45
partner is under no obligation to do so
00:34:46
the general partner is free to reinvest
00:34:48
any and all of the operating entities
00:34:51
cash flow into research and development
00:34:52
activities Andor related expenses
00:34:54
without any obligation to the limited
00:34:56
partners and so they basically told
00:34:59
everybody venod and employees or
00:35:01
whatever we can just basically make wipe
00:35:03
your Equity out and we can do whatever
00:35:04
we want with the profits and you're
00:35:06
probably going to lose your money so
00:35:08
this this structure is weird do you guys
00:35:10
think that when the investors came in
00:35:13
especially in this latest round the $86
00:35:15
billion round do you think they
00:35:17
underwrote the legal risk inherit in the
00:35:19
structure or they just sort of handwaved
00:35:23
over it and said no typically what
00:35:25
happens in these deals
00:35:29
is you hire somebody like KPMG or deoe
00:35:32
or ernston young to do like the full
00:35:34
Financial diligence packet right and
00:35:37
that typically tends to be how a lot of
00:35:38
late stage
00:35:40
organizations document that they've done
00:35:43
and manage their fiduciary
00:35:44
responsibilities on behalf of their
00:35:45
limited partners so what happens is you
00:35:48
will do a deal you'll sign a term sheet
00:35:50
you'll turn it over to deoe you'll turn
00:35:52
it over to KPMG and say please go and
00:35:55
run this down and and then what they do
00:35:57
is they will furnish a report that says
00:35:59
yes this meets all the customary
00:36:01
expectations I suspect that if these
00:36:04
folks were
00:36:05
doing a decent job of of running late
00:36:08
stage money they probably sent it to
00:36:11
those folks and those folks probably
00:36:13
produced something that said this looks
00:36:15
fine now if you read Buffett's letter
00:36:17
this year he has a really great
00:36:20
commentary on deoe and KPMG and these
00:36:23
sorts of letters which is not exactly
00:36:26
the most supportive is the best way to
00:36:29
say it you can find that mention in in
00:36:31
his case it's deot and touch but
00:36:33
typically Sachs that's what they do they
00:36:34
go to a KPMG or a deloid or a nurse and
00:36:36
young and say and and by and by the way
00:36:39
99% of the time they're I mean they do
00:36:42
really good work but it's a standard
00:36:46
structure and they just want to make
00:36:47
sure that nothing nefarious is a muck in
00:36:49
this case I doubt anything was a muck
00:36:51
anyways I doubt though that they looked
00:36:53
at the structure and then also elevated
00:36:56
the the litigation risk of this getting
00:36:58
Unwound I just don't see that that
00:37:00
diligence report and I've seen enough of
00:37:01
them typically has
00:37:04
that section in it the other thing
00:37:07
that's completely hypocritical here is
00:37:09
they said when they hit AGI and they're
00:37:11
going to be like a sentient artificial
00:37:14
intelligent going on here Friedberg that
00:37:17
they would wrap up shop and they're
00:37:18
going to no longer be a non nonprofit
00:37:20
Etc but they haven't and but they're
00:37:22
claiming they haven't hit
00:37:24
that but they closed the soft so it
00:37:27
should be open source if they haven't
00:37:28
hit AGI
00:37:30
and you don't think they've hit general
00:37:33
intelligence right free barg or anything
00:37:35
close to it maybe you could educate the
00:37:37
audience on what that is and and that
00:37:39
claim that they have to shut off the for
00:37:41
profit yeah they haven't but I think
00:37:45
the we keep repeating this concept of
00:37:48
the models should be open source versus
00:37:51
closed Source making AI for the benefit
00:37:53
of humanity can be interpreted in a lot
00:37:55
of ways there may have been some
00:37:56
anecdotal conversation at some point
00:37:58
with Elon or others about we're going to
00:38:01
make the models open
00:38:02
source uh but there was a reason that
00:38:05
that change was made along the way which
00:38:08
was to attract dollars and those dollars
00:38:12
need to have some return of capital
00:38:15
available to them because they're
00:38:17
private investor dollars and so I don't
00:38:19
think that that was necessarily and
00:38:20
correct me if I'm wrong I don't think
00:38:22
that's in the mission that the open AI
00:38:24
software models will be open source
00:38:27
making AI for the benefit of humanity
00:38:28
could probably be interpreted in a lot
00:38:30
of different ways and we'll see but no I
00:38:33
I don't think that anyone has this
00:38:36
achieved
00:38:38
this Holy Grail of general intelligence
00:38:42
yeah I think one of the the one of the
00:38:45
more interesting and kind of wacky
00:38:47
things about open AI is that their
00:38:50
mission is explicitly to create
00:38:52
AGI which most people would associate
00:38:56
with some sort of sci-fi dystopian
00:38:58
outcome and I think this has like raised
00:39:01
the Fear Factor around AI because
00:39:04
they're explicitly trying to create the
00:39:07
sence that's going to replace Humanity
00:39:09
now I think they defined AGI in a
00:39:11
different way they say it's something
00:39:13
that can replace 80% of the jobs but I
00:39:16
think we all kind of know what it really
00:39:18
is so I just wonder if I think I think
00:39:21
that that's like that assumes a steady
00:39:23
state in the world so if you end up with
00:39:27
a system and the system has all the
00:39:32
capabilities of a bunch of really highly
00:39:34
qualified knowledge workers and I can
00:39:36
sit in front of a computer terminal and
00:39:39
I can say let's design a mission to Mars
00:39:41
a mission to Mars could be a 20-year
00:39:43
engineering project with hundreds of
00:39:45
people involved to design the buildings
00:39:47
to design the flight path to figure out
00:39:49
the fuel needs to figure out how you
00:39:50
would then be able to terraform ours and
00:39:53
what if one person could interact with a
00:39:54
computer and design a plan to go and
00:39:57
inhabit Mars all of the technical detail
00:40:00
docks could be produced all of the
00:40:01
engineering specifications could be
00:40:03
generated all of the operating plans all
00:40:05
of the dates the amount of Labor needed
00:40:07
the amount of production needed the
00:40:08
amount of capital needed what would
00:40:10
otherwise take NASA or some you know
00:40:13
International or well-funded private
00:40:15
company many many decades to do a piece
00:40:17
of software could do in a very short
00:40:19
order I think that's like a really like
00:40:22
for me poignant example of the potential
00:40:25
of having these tools broadly available
00:40:27
that the potential of humanity starts to
00:40:30
become much broader we could say I want
00:40:32
to develop a city underneath the ocean
00:40:34
because I want to explore more of the
00:40:36
earth I I think humans need to go go
00:40:38
solve cancer figure out the biologic
00:40:40
drugs and the combination of biologic
00:40:42
drugs that would be needed to solve
00:40:43
cancer based on this this patient's
00:40:46
genotype the extensibility of Highly
00:40:49
knowledgeable or what other people might
00:40:51
call general intelligence type tooling
00:40:53
is extraordinary that one individual
00:40:55
starts to have an
00:40:57
entire cohort of knowledge workers
00:41:00
available at their disposal to do things
00:41:02
that we can't even imagine today so I
00:41:04
don't think that it is nefarious it's
00:41:06
nefarious because we assume a steady
00:41:07
state of the world today that nothing
00:41:08
changes therefore a piece of software
00:41:11
replaces all of us but the potential of
00:41:13
humanity starts to stretch into a new
00:41:14
era that we're not really comfortable
00:41:17
with because we don't really know know
00:41:18
it or understand it yet I'm not saying
00:41:19
it's nefarious to uh want to develop
00:41:22
AI because I agree with you about all
00:41:25
the extraordinary potential of but I'm
00:41:27
saying there's something a little bit
00:41:28
cultish and weird about explicitly
00:41:31
devoting yourself to
00:41:33
AGI which I think in common parlance
00:41:36
means
00:41:37
Skynet yeah it mean
00:41:40
you maybe that that parlance is what
00:41:43
needs to be addressed which is Agi
00:41:45
effectively enables equivalence to a a
00:41:47
human knowledge worker and that you know
00:41:50
that can kind of unleash a new kind of
00:41:52
set of opportunities you think that's
00:41:54
what it is my definition for AGI is
00:41:56
smarter than the smartest human being
00:41:57
who ever lived yeah I was talking to
00:41:59
somebody this week who's in a position
00:42:01
who said the definition of AGI is very
00:42:03
fuzzy that there isn't a clear
00:42:05
definition and therefore it allows every
00:42:08
side to kind of anchor on their
00:42:09
interpretation of what that term means
00:42:11
and therefore kind of justifies their
00:42:13
position so you know I don't really feel
00:42:15
great about like just saying are we at
00:42:17
AGI what is it we don't have a clear
00:42:19
sense of what it means I do think if you
00:42:20
look at some of the work that was done
00:42:22
by anthropic and published in the cloud
00:42:24
3 Model this week did any of you guys
00:42:26
see the demos that were done of the
00:42:27
output of that model there was a guy who
00:42:30
wrote a thesis in quantum physics on a
00:42:32
very esoteric complicated problem set
00:42:34
and he asked Claude 3 to solve this
00:42:36
problem set and it came up with his
00:42:38
thesis it was really like extraordinary
00:42:40
and this is something he's like no one
00:42:41
in the world knows this stuff and he's
00:42:43
like I can't believe this model like
00:42:45
came up with my thesis and that's the
00:42:48
sort of thing that very few people on
00:42:49
Earth even read or understand and the
00:42:52
cloud 3 Model was able to kind of
00:42:53
recreate the basis the buildup and then
00:42:56
the the the output of his desent was
00:42:59
somebody writing a screenplay and then
00:43:00
giving it the first two acts and say
00:43:02
guess the third act and it's like oh
00:43:03
yeah this is the third act here's what
00:43:04
happens like it's pretty
00:43:06
impressive comp ask you like deep down
00:43:09
when these guys say they're going to
00:43:10
create AGI what do you think they really
00:43:14
mean in their heart of hearts oh they
00:43:16
mean the Terminator yeah they mean the
00:43:18
sentient God you guys are you guys are
00:43:20
propagating some bad you guys shouldn't
00:43:22
be I think that's what they
00:43:25
think said it remember what Larry paage
00:43:27
said to Elon don't be
00:43:29
speciest yeah I think there's like I
00:43:32
think there's a meaningful number of
00:43:33
people in The Tech Community who
00:43:36
deliberately want to give rise to the
00:43:38
super intelligence there's another point
00:43:40
of view of super intelligence where
00:43:41
super intelligence means that the
00:43:44
software is now more intelligent than
00:43:45
all humans and as a result the software
00:43:48
may have its own motivations to figure
00:43:50
out how to supersede humans on earth now
00:43:54
the the Larry Page statement which which
00:43:56
I I don't know firsthand I read the same
00:43:58
article you did is one that a group of
00:44:00
people might say evolution is
00:44:03
evolution you know people that's right
00:44:07
and I know elon's taken this point of
00:44:09
view that like you know we need to
00:44:10
maintain
00:44:11
human human Supremacy but uh my my
00:44:15
favorite test is you know there was
00:44:17
there's the Turing test which like you
00:44:19
can't tell if it's a human or if it's a
00:44:20
robot but Mustafa came up with the um
00:44:23
with the modern touring test which is an
00:44:25
AI mod is given $100,000 and has to
00:44:28
obtain 1 million go it's kind of
00:44:30
interesting then the other other the
00:44:32
other interesting one was Gary Marcus he
00:44:35
said uh the Ikea test the flat pack
00:44:38
Furniture test an AI views the parts and
00:44:40
instructions of an Ikea flat pack
00:44:42
product that controls a robot to
00:44:44
assemble the furniture correctly oh look
00:44:47
what's up sunny it's now time to do a
00:44:49
just a quick little congratulations to
00:44:52
our dear friend Sund deep mandra we call
00:44:54
him Sunny that's his nickname and he's
00:44:56
one of our poker buddies he keeps
00:44:58
building great
00:44:59
companies and in this case sandep is the
00:45:02
first person to collect all four besties
00:45:04
we all invested in his company
00:45:05
definitive and definitive uh was working
00:45:08
in AI but we got some great news this
00:45:11
week and we thought we would give him
00:45:13
his flowers Sunny you want to tell us
00:45:15
what happened this week with our
00:45:17
investment in your company definitive
00:45:19
intelligence at definitive doio I
00:45:21
believe is your demand yeah well you
00:45:24
know with your guys support and you know
00:45:26
we've been growing our company and we
00:45:27
saw a really great opportunity to work
00:45:30
together with grock and we've been
00:45:31
working with them for a couple of months
00:45:32
and all the hype that you've seen has
00:45:35
been built on the collaboration that
00:45:36
we've done building the cloud offering
00:45:39
the API offerings and so you know we've
00:45:41
decided to merge with them and we're
00:45:43
super excited and all the besties are
00:45:45
now not only shareholders in definitive
00:45:48
previously but now shareholders in Gro
00:45:50
this is this is the first investment I
00:45:52
think where we're all on the cap table
00:45:54
right so we can we're all rooting in the
00:45:55
same Direction what's happening in the
00:45:57
developer site how's the momentum going
00:45:59
well the the momentum is incredible you
00:46:00
know we have now 16,000 plus developers
00:46:04
in our you know self- served playground
00:46:07
there's well over like a thousand apps
00:46:09
that people have developed using the API
00:46:12
and all kinds of new functionality you
00:46:14
know the API allows people to get a
00:46:16
higher rate of throughput on tokens and
00:46:18
low latency so there all kinds of new
00:46:20
applications from voice to real-time
00:46:22
translation of web pages you know we're
00:46:25
collecting on our Discord we have you
00:46:27
know 3,000 people in the Discord as well
00:46:30
it is a real Community that's come
00:46:32
together building around Gro and what I
00:46:33
will say is you know sort of the same
00:46:35
jump that developers saw when we went
00:46:37
from dialup internet to broadband
00:46:38
they're seeing that now and from using
00:46:40
like traditional apis for llms to using
00:46:43
you know the ones that we offer you guys
00:46:46
support the latest anthropic models that
00:46:48
they just launched that seem pretty
00:46:50
kickass no we don't have those yet but
00:46:52
we're we're in you know we're having
00:46:54
discussions with everyone out there
00:46:55
there and we want to support their
00:46:57
models right now what we've done given
00:46:59
all the demand is we've kind of limited
00:47:00
it to llama 270b and mix trol and we
00:47:03
actually have a bunch of other models
00:47:04
that we make available in private mode
00:47:06
for folks we're pretty excited but if
00:47:08
there's anyone out there that wants to
00:47:09
have us you know give us a call and
00:47:11
we'll we'll get you going on our systems
00:47:13
well Sunny congrats to
00:47:16
you you're an incredible entrepreneur
00:47:19
and it's always fun to kind of be on the
00:47:22
journey with you this is my fourth
00:47:24
business that I've done with sunny madra
00:47:27
that's amazing extreme lands then there
00:47:29
was a company between no no no there was
00:47:32
before you have to understand sunny and
00:47:33
I met because Sunny went to school where
00:47:35
I grew up he went to the University of
00:47:36
Ottawa I grew up there and we met
00:47:38
through a mutual friend and his first
00:47:40
company was called sponge
00:47:42
fish which I backed I think in 2006
00:47:46
maybe I mean I had no money I may have
00:47:48
written a
00:47:49
0,000 10K maybe maybe less May 5K yeah I
00:47:54
mean whatever I had I didn't have much I
00:47:55
like it's scraped together and it's it's
00:47:57
our fourth business Sunny's a veritable
00:47:59
rabbit's foot you're a rabbit's foot sun
00:48:02
he is he is a lot Lu prefers the
00:48:06
prepared well I mean I've been in two of
00:48:08
Sund Deep's companies and uh I thank him
00:48:11
for that and then extreme Labs actually
00:48:13
to to thank you again you sponsored a
00:48:15
lot of my events 10 years ago when I was
00:48:17
coming up so I appreciate that so it's
00:48:18
just great to see a nice guy win as
00:48:21
opposed to the three miserable people
00:48:23
who I work with on this podcast keep
00:48:24
winning so it's good now finally a good
00:48:26
guy wins as opposed to the four of us I
00:48:28
appreciate all your guys support it
00:48:30
means a lot to us you guys getting the
00:48:32
developers and keep keep putting it out
00:48:35
there and and keep us honest as well so
00:48:37
if we're not doing something right let
00:48:38
us know really appreciate you guys all s
00:48:40
you want to say something motivational
00:48:43
to sunny here I know you always have a
00:48:44
great motivational word for your friends
00:48:47
something you say that just gives people
00:48:49
that thrill of being in the game go
00:48:51
ahead sax you always have something say
00:48:53
we're going to we're going to ship uh 20
00:48:55
million into your a safe note I know you
00:48:56
closed it but we're going to pry it open
00:48:58
and uh get get some get some allocation
00:49:01
hey wait a second if you're prying open
00:49:03
that's safe if the safe if there's a
00:49:05
wedge I know a guy with a podcast who
00:49:08
could got 800,000 followers maybe I
00:49:09
could slip in a quick five hundy is
00:49:11
there possible what do you think Cy let
00:49:13
us know right now I don't know we we did
00:49:14
a really big announcement with the
00:49:15
Saudis this week so that price might be
00:49:18
up let us ship something into that safe
00:49:20
note so we actually have some real skin
00:49:21
in the game David text me all right
00:49:23
we'll do well no but seriously hey s
00:49:25
wait a second if he I got to get a
00:49:27
second Jason wants in for 500 by that he
00:49:30
means $500 so we'll let you no I got a
00:49:32
gift card for you from Starbucks I got a
00:49:34
$500 gift card for you and the team I
00:49:37
think I got
00:49:38
likeon that that Starbucks gift card you
00:49:42
sell for 20 cents in the dollar and then
00:49:43
ship let me zip it in let me zip it in
00:49:45
maybe I can 100 Exit no seriously 500k
00:49:47
from from your boy jcal all right
00:49:48
everybody thanks uh to Sunny for jumping
00:49:51
on congratulations on the mder thanks
00:49:53
guys all right issue two issue two apple
00:49:56
is battling two major iOS developers and
00:49:59
Regulators are siding with the devs you
00:50:02
may know about the Apple versus epic gam
00:50:04
Saga we've talked about it here epic
00:50:06
games was planning to create a custom
00:50:08
App Store on iOS because Europe's dma
00:50:12
the the digital markets act has said
00:50:14
that Apple now has to allow thirdparty
00:50:16
app stores in the EU so epic created a
00:50:20
developer account based in Sweden and
00:50:22
apple actually approved the account two
00:50:23
weeks ago then on Wednesday Apple
00:50:25
flipped
00:50:26
terminated epic EU developer account and
00:50:28
app said one of the reasons they
00:50:30
terminated the account was because epic
00:50:31
CEO publicly criticized their dma
00:50:34
compliance plan additionally on Monday
00:50:37
Apple was fined two billion by the eu's
00:50:39
antitrust regulators and was forced to
00:50:41
remove its anti- steering rules uh from
00:50:44
music apps like Spotify basically Apple
00:50:46
has been restricting music apps from
00:50:47
informing users about pricing and
00:50:49
discounts and the European commission
00:50:52
considered this anti-competitive since
00:50:53
Apple runs Apple music
00:50:56
and they want Spotify to pay 30% Daniel
00:50:59
e friend of the
00:51:00
Pod you know basically did a whole video
00:51:03
on this about how they can't charge 30%
00:51:05
more yada yada sax you've uh spoken
00:51:08
about Apple's Monopoly before your
00:51:10
thoughts on what's happening in the EU
00:51:13
and then we'll get into Apple's wider
00:51:15
problems I mean did you just say that
00:51:18
Apple booted Epic from their app store
00:51:21
because they didn't like what epic said
00:51:23
about them I think
00:51:26
well like they're violating epic's Free
00:51:29
Speech because they don't like what epic
00:51:30
is saying I mean according to Epic yes
00:51:33
this is crazy heavy-handed by apple
00:51:36
apple heavy-handed with developers yeah
00:51:38
have they lost their minds I mean this
00:51:39
is right out of power corrupts and
00:51:41
absolute power corrupts absolutely yeah
00:51:43
I mean whatever dispute you have with
00:51:45
epic you don't boot them out of your app
00:51:47
store because you don't like their
00:51:48
criticism of you I mean this is
00:51:51
basically proving exactly what
00:51:53
everyone's been saying about Apple which
00:51:55
is they're too powerful and heavy-handed
00:51:58
and Apple's coming along and saying let
00:52:00
me confirm it for you guys by acting
00:52:02
tyrannically against epic exactly I mean
00:52:05
talk about a backfire this seems insane
00:52:07
to me yeah it's super nuts chamat any
00:52:10
that'ss on this before we get into
00:52:12
Apple's other problems I think it's the
00:52:15
beginning of of the decay of Apple Peak
00:52:19
Apple well let's let's get into that um
00:52:21
there's tons of headwinds facing Apple
00:52:22
you can add a bunch of other things to
00:52:24
the list as well Jason yeah the thing is
00:52:26
Apple for the last couple of years has
00:52:28
been what is effectively what we call a
00:52:30
GDP plus Growth Company which means that
00:52:34
take GDP 2 3 4% maybe they can grow by a
00:52:37
couple of percentage points more than
00:52:38
that but they are effectively levered to
00:52:41
GDP meaning when you look at a Facebook
00:52:44
or an Nvidia they're growing at 50% 200%
00:52:47
2,000% whatever it is that's not tied to
00:52:50
GDP they're just taking share but Apple
00:52:53
is is a company now that grows as the as
00:52:55
the economy of the world growth so
00:52:56
that's not super great for its future
00:52:58
prospects unless it can expand the
00:53:01
surface area of where they operate and
00:53:03
then on that Dimension the there are a
00:53:06
few trillion dollar markets that they
00:53:08
can really penetrate and they just
00:53:10
announced that they've killed a project
00:53:11
in one of those areas which is Autos
00:53:13
right project Titan which was1 billion
00:53:16
turned out to be a
00:53:17
failure so all of these things I think
00:53:19
mean to me that it is effectively
00:53:21
becoming a cyclical rate sensitive stock
00:53:27
and then the Cuda is Warren
00:53:29
Buffett and Nick and I were talking
00:53:31
about it this week and what was
00:53:33
interesting about Buffett's letter is
00:53:35
that you can tell when Buffett has
00:53:39
gotten disengaged with a company based
00:53:42
on the number of times he mentions it in
00:53:45
his annual letter so in this example
00:53:48
this is the number of times Apple was
00:53:49
mentioned and just to be clear what I
00:53:51
mean by mentions is not when it's
00:53:52
included in a chart or part of a
00:53:54
disclosure what I mean is when Warren
00:53:56
actually explicitly mentions it in a
00:53:58
positive or even negative way or he
00:54:01
doesn't mention it at all which I think
00:54:03
Rings very loudly he went from basically
00:54:06
saying Apple was the absolute endall and
00:54:10
Beall and now what you can start to see
00:54:13
is this shrinking and it's gone from
00:54:15
basically a bunch of times to almost
00:54:17
none he did mention it once but he
00:54:19
mentioned it in the context of talking
00:54:21
positively about Coca-Cola and AMX and
00:54:23
he Wasing these two positions but just
00:54:26
mentioning that they were not as large
00:54:28
in comparison to Apple that's the only
00:54:30
mention in this year's annual letter
00:54:33
what's interesting about that is the
00:54:34
last time that that happened was with
00:54:36
drum roll Wells Fargo over 15 and 20
00:54:39
years Buffett built up a huge
00:54:43
position I think he was able to weather
00:54:46
the vicissitudes of the market so even
00:54:48
when the markets would contract he knew
00:54:50
when to hold on to that company until he
00:54:52
realized that that company was not
00:54:54
really one of his forever stocks and he
00:54:57
got out of it and the number of times it
00:55:00
basically was mentioned in his letter
00:55:02
went to zero so interestingly I think
00:55:05
this Buffet index is a really important
00:55:08
one for Apple which is it went from a
00:55:09
forever holding that he said he would
00:55:12
own forever to barely getting mentioned
00:55:15
and above him on above apple on that
00:55:17
list were all the Japanese trading
00:55:19
companies that Buffett owns American
00:55:22
Express Coca-Cola
00:55:24
so that is a person that understands the
00:55:26
economy I think better than anybody else
00:55:28
in the world and so if you're basically
00:55:31
taking a lever bet to the economy as a
00:55:33
reason to own apple and the person that
00:55:35
understands the economy the most has now
00:55:37
started to Pivot away he started to sell
00:55:40
in quarter four and then you see all of
00:55:42
these things Anti-Trust rules killing
00:55:44
projects in in trillion dollar Tams
00:55:47
unfortunately it speaks for a very bad
00:55:50
next 5 to 10 years for this company
00:55:51
unless they figure something out I think
00:55:53
it's just such a great insight with the
00:55:55
Warren Buffett mentions and we should do
00:55:57
that forward looking like who is he
00:55:59
talking about now and let the monitor
00:56:01
when he stops talking about them I have
00:56:02
a theory about this I think this is
00:56:04
really just about Peak iPhone if you
00:56:07
look at the majority of their revenue
00:56:08
it's obviously from the iPhone which has
00:56:11
become massively profitable over time
00:56:13
but the iPhone Revenue has been flat for
00:56:15
a couple of years now and they're
00:56:18
starting to make their money from
00:56:19
Services everybody knows Apple 1 iCloud
00:56:22
storage Apple arcade Apple music apple
00:56:24
plus TV they've got a great collection
00:56:27
of services and services revenues
00:56:28
growing but I don't think that they are
00:56:31
filling in this growth problem with the
00:56:34
iPhone and this happened to me for the
00:56:35
first time and I don't know if you
00:56:37
gentlemen have had the same experience
00:56:38
but I was buying every Apple phone and
00:56:40
then I would buy every like M medium
00:56:43
upgrade like when they would do like a
00:56:45
12 and then a 12s or an 11 11s and when
00:56:48
I got to iPhone 13 I was like this
00:56:50
thing's kind of peaked and I just forgot
00:56:52
to buy the 14 didn't need it and then I
00:56:54
bought the 15 and I'm sitting here with
00:56:55
my 15 and my 13 I really couldn't tell
00:56:57
the difference between them and so now
00:56:58
I'm a technologist who would buy it
00:57:00
every year the latest one and I don't
00:57:03
feel the need to upgrade it and I think
00:57:05
that a lot of people and I I have family
00:57:07
members who would take two generations
00:57:09
off now they take three or four
00:57:10
generations off so I don't even know
00:57:12
what number I'm on you know I don't even
00:57:14
know either it all looks the same to me
00:57:16
okay so I mean that's the problem here
00:57:18
and then if you look at their road map
00:57:21
what's what is the device post the Steve
00:57:24
Jobs era that they have launched how do
00:57:26
you know what device you have you go to
00:57:29
General in your settings and about I
00:57:31
just checked it I'm iPhone 13 Pro what
00:57:33
version are they on what is it General
00:57:36
they're on 15 and I'm on 13 Pro I don't
00:57:37
even know yeah and you're not price
00:57:40
sensitive yeah and these things have
00:57:42
gotten absurdly expensive so they're
00:57:43
trying to really extract what is the
00:57:45
what is the latest 15 15 I'm on a 14 I
00:57:49
don't like the hassle of having to like
00:57:51
reset everything when I get a new phone
00:57:53
and it's easy to do now it's like all in
00:57:55
the cloud but even that going to the
00:57:57
store buying it like who cares it's not
00:57:59
my apps I have to relog in and freeberg
00:58:02
what uh what what Android phone are you
00:58:04
on are you you're a googler didn't the
00:58:06
googlers all have Android phones or are
00:58:08
you an your team
00:58:10
iPhone I got the Macintosh original in
00:58:12
1984 I've only been on MacIntosh and
00:58:15
Apple product since then
00:58:17
H what iPhone are you on be honest 15
00:58:21
Pro okay have you ever tried a pixel
00:58:24
freeb
00:58:26
I don't like
00:58:27
it but I just I'm not used to the OS I
00:58:30
just what was the rule at Google if you
00:58:31
worked at Google did were you kind of
00:58:33
forced to have an Android phone or did
00:58:35
you just keep your iPhone in your pocket
00:58:37
and not take out there weren't a lot of
00:58:39
Androids on the market when I was at
00:58:41
Google because we acquired the company
00:58:43
when I was there in O we acquired Andy
00:58:46
Ruben's company in
00:58:50
04 and I think the first major devices
00:58:53
started to roll out in
00:58:55
7 and so this was all post my era I left
00:58:58
at the end of
00:58:59
06 wow freeberg that's an amazing chart
00:59:02
holy so pull up this chart now here's
00:59:05
the global here's the global statistics
00:59:07
the US market share is 57% for iOS but
00:59:10
on a global basis iOS only has a 27%
00:59:14
market share in uh mobile operating
00:59:17
systems Android is 72% and all other is
00:59:21
less than 1% but look at this it's all
00:59:23
the future GDP is on Android Brazil
00:59:26
Nigeria India holy but I think this is
00:59:30
one counter to your point shamat that
00:59:32
the macro driver is as um the economic
00:59:36
uh position the GDP per capita scales in
00:59:39
these bricks Nations they can start to
00:59:42
afford to buy iPhones which are
00:59:44
generally some multiple of the Android
00:59:46
devices in these markets that you see
00:59:48
here so while today Android is 72% of
00:59:51
the market if the Emerging Markets
00:59:53
continue to grow GDP capita and iOS
00:59:55
continues to be the superior product
00:59:58
you'll see apple able to to steal into
01:00:00
more share over time how do you but how
01:00:02
do you do that when you have messaging
01:00:04
groups on Android when you have photos
01:00:06
in Google photos and is it does the
01:00:09
switching cost stop that do you think I
01:00:11
don't know the answer to that it's a
01:00:12
good question they're down to5 to $25
01:00:16
for these Android phones in India
01:00:17
there's a famous that's the difference
01:00:20
right so these Android devices I think
01:00:22
apple is levered to GDP Global GDP so
01:00:24
they need to figure out some way to grow
01:00:26
Superior to that is the market what
01:00:29
could they do Services a company that
01:00:32
generally has a bunch of cash flow being
01:00:36
generated by some set of products today
01:00:39
and you assume stasis over time the
01:00:42
market share for those products could be
01:00:43
eaten away you'll see the profits per
01:00:45
year go down and a company like that
01:00:47
will trade anywhere from you know 7 to
01:00:49
20 times the technology value arises
01:00:51
from the value of the brand that you can
01:00:53
launch new products lever your brand
01:00:55
leverage your distribution leverage the
01:00:56
sale of new services and new products I
01:01:00
think to your point shim off the
01:01:01
challenge apple is facing is that the
01:01:04
pool of options the portfolio of call
01:01:07
options that you would get a new product
01:01:09
coming out of apple is shrinking very
01:01:10
Lim with the you know with the car being
01:01:13
taken out of that pool now Apple Vision
01:01:15
Pro a lot of question marks on you know
01:01:17
how much it can scale as you guys know I
01:01:19
I feel like there's going to be a market
01:01:20
for that device but it's a it's a
01:01:23
high-end device today it has to become
01:01:25
more cheap to be more ubiquitous the
01:01:27
thing that a lot of these companies
01:01:28
confront though is that you can also
01:01:30
grow inorganically right you don't
01:01:31
necessarily have to incubate these
01:01:34
projects we can remember the moment
01:01:36
Apple had a chance to buy
01:01:38
Tesla right didn't then they didn't take
01:01:41
it Apple still has a chance to buy pick
01:01:44
your company that that would take a
01:01:47
sweet acquisition offer rivan Lucid
01:01:50
postar whatever it is I'm not saying
01:01:51
that these companies are good or not
01:01:53
good I'm just saying that Apple has the
01:01:55
one tool that they've never used in
01:01:57
their toolbox is the large inorganic
01:02:00
acquisition and at this point if they
01:02:03
are proving that they can't execute
01:02:05
internally the market is going to demand
01:02:07
that they prove they can use that
01:02:09
balance sheet as a cudgel to go and
01:02:11
execute it externally yeah I think the
01:02:12
challenge do that they're GNA discount
01:02:15
that cash to zero the challenge is that
01:02:18
there's a certain discipline and quality
01:02:21
to the products and the businesses that
01:02:23
apple produc es and runs and it's very
01:02:26
hard to see that in other markets I mean
01:02:28
why would they go buy a money losing low
01:02:30
margin Hardware uh car hardware company
01:02:33
when they are proven to make you know
01:02:36
quality on the reasons project tipan
01:02:38
fil yeah the article in Bloomberg
01:02:42
basically said that the reason that they
01:02:43
abandoned the project was in part that
01:02:47
they did not know how to go from where
01:02:49
they were to a full production vehicle
01:02:52
that could be level five and then when
01:02:54
it was proposed that they step down and
01:02:56
just launch a level three autonomous
01:02:57
vehicle everybody said no that it wasn't
01:03:00
disruptive enough well if that's the
01:03:02
case the the thing that they made a
01:03:04
decision about there was not going into
01:03:06
a market because of Regulation not
01:03:07
because of technical
01:03:09
capability and I don't think that that's
01:03:11
necessarily a smart decision they would
01:03:13
have been better off going into the car
01:03:15
market launching a level three vehicle
01:03:17
and just letting the market play out
01:03:18
they probably just like they were able
01:03:20
to do music have the influence to change
01:03:22
the laws yeah especially if they stepped
01:03:25
in there with their rigor right and they
01:03:27
they're not doing that so it's a little
01:03:29
bit of a head scratch of what's going on
01:03:30
over there they need some new products I
01:03:32
mean maybe Apple Vision Pro someday
01:03:34
becomes the platform of the future but
01:03:36
they've squeezed as much revenue as they
01:03:38
can out of the iPhone Services seems
01:03:40
like the best place for them to make
01:03:41
money but as we see my idea under
01:03:44
assault yeah my idea was just that they
01:03:47
should launch a huge competitor to AWS
01:03:49
as Ur and gcp so if you look every
01:03:52
developer that Apple has theoretically
01:03:54
could have been running on an Apple not
01:03:57
just on Apple's SDK and apis but they
01:03:59
should be running in an Apple Cloud you
01:04:00
could have made that claim and it would
01:04:02
have made a ton of sense for app
01:04:04
developers to have TurnKey access to
01:04:06
that right and they could have
01:04:07
subsidized it and by the way to Sax's
01:04:09
point that would have been an incredible
01:04:11
way to defend a 30% rev share okay
01:04:13
listen I'm taking 30% but here's a bunch
01:04:15
of subsidized access to hardware and the
01:04:17
market would have loved it and in this
01:04:20
AI shift they can still do that where
01:04:23
now people are chipping away at the 30%
01:04:25
right people are saying well I can just
01:04:27
build around you they need to do
01:04:29
something and service is Jason the most
01:04:33
valuable thing they could do would be to
01:04:34
launch a big cloud I think yeah I mean
01:04:37
they have all the app developers on test
01:04:38
flight ready to go they just email them
01:04:40
and say hey here's your free storage I
01:04:42
mean they could just slowly add features
01:04:44
right and they're they're doing this uh
01:04:46
open source I think it's called Maggie
01:04:49
or Magpie or something but they're doing
01:04:51
an image one and uh well look if they
01:04:54
could if they could just make Siri work
01:04:56
like the way it was supposed to using an
01:04:58
llm and have it be like Snappy that'd be
01:05:01
a major upgrade to the iPhone I don't
01:05:04
think you need to buy a new iPhone for
01:05:05
that because it would just be a software
01:05:06
upgrade but just getting Siri to work
01:05:09
would be a big win I think that's where
01:05:12
they're going with their silicon and
01:05:13
they just announced the M3 on the
01:05:15
MacBook airs like they're making their
01:05:17
own silicon and I think it's going to
01:05:19
power llms locally on the devices and
01:05:21
that's so powerful I was using I photo
01:05:24
this week and in apple photos there's
01:05:28
now on certain images if you swipe
01:05:29
through them you'll see the little AI
01:05:31
icon when you click it it tells you
01:05:34
things in the photo like that's a
01:05:36
bulldog that's pasta and so they're
01:05:38
already subtly adding these features now
01:05:40
that's a feature Google's had in Google
01:05:42
photos for five years but go ahead and
01:05:45
open your phone right now and search for
01:05:46
a dog or a bulldog or whatever type of
01:05:48
dog you got you know uh Labrador and
01:05:51
then watch golden retriever that it
01:05:53
actually knows how to do it they never
01:05:54
announced it it's just subtly being put
01:05:56
in there uh so there's so many
01:05:58
opportunities if they required a
01:06:00
hardware upgrade to get like actually
01:06:02
good AI built into the software then
01:06:05
everyone's going to have to upgrade for
01:06:06
that that's a pretty big motivation to
01:06:09
upgrade there you have it Tim Cook
01:06:11
somebody send this clip to Tim Cook put
01:06:13
put AI chips on your uh phone this isn't
01:06:15
hard right I mean all they got to do is
01:06:17
just take the latest open source models
01:06:20
and figure out how to customize them for
01:06:21
their own products yeah and they're
01:06:23
doing it doing it right now all right
01:06:25
let's go to Issue four Tik Tock
01:06:27
bipartisan ban will the CCP agree a
01:06:31
bipartisan group of a dozen plus
01:06:33
lawmakers introduced a bill that would
01:06:35
effectively ban Tik Tock in the house
01:06:37
this week the bill is officially called
01:06:38
the protecting Americans from foreign
01:06:40
adversary controlled applications act
01:06:43
gives bite dance 165 days to divest from
01:06:46
Tik Tok that's the parent company the
01:06:47
Chinese company that owns Tik Tok the
01:06:49
app that's very popular in the west
01:06:51
especially here in America it would make
01:06:53
it legal for companies like Apple and
01:06:55
Google to show Tik Tok in their app
01:06:57
stores as you know Tik Tok 170 million
01:07:00
us users and they claim that they're
01:07:03
headquartered headquartered in Singapore
01:07:05
I know people who have worked at Tik Tok
01:07:06
or do work at Tik Tok and they said
01:07:09
that's nonsense to me the company said
01:07:12
it has not and will not share user data
01:07:14
with the CCP in my mind it's an obvious
01:07:17
lie in 20121 the CCP took a board seat
01:07:19
on bite dance Beijing based subsidiary
01:07:22
that's according to Reuters and then in
01:07:23
2022 by dance admitted that it accessed
01:07:26
IP addresses and data by journalists
01:07:28
covering Tik Tok to see if they'd been
01:07:30
in the same places as B Dan employees
01:07:32
obviously to find leakers and by Dan
01:07:34
claims they fired the people involved y
01:07:36
yada last year a former head of
01:07:37
engineering at by Dan us said CCP
01:07:39
members had god mode access to user data
01:07:42
in 2018 sax um I can go explain more of
01:07:45
this but I just got to go to you here if
01:07:48
this if bite dance is not spying on
01:07:51
Americans and the CCP is on the board
01:07:54
that would be make no logical sense to
01:07:56
me and why are they fighting the
01:07:58
divestiture if it's just a financial
01:08:00
reason why would they take the CCP off
01:08:02
their board do you think it's spyware do
01:08:03
you think the US is crazy for allowing
01:08:05
this product here in the United States
01:08:07
when we're not allowed to put Twitter
01:08:10
Facebook Instagram pick your social
01:08:12
network in China we have zero
01:08:14
reciprocity here look if it's true that
01:08:19
Tik Tok is sharing data with the CCP
01:08:21
then I think the United States is well
01:08:23
within rights to either ban it or cause
01:08:26
it to be divested and I personally like
01:08:29
the divesture option I mean that's what
01:08:31
Trump was suggesting during his term
01:08:33
because we don't I think in the United
01:08:34
States like to
01:08:36
just essentially confiscate or destroy
01:08:40
people's property but I think we're
01:08:41
within our rights to require to be
01:08:43
divested to an entity that we know is
01:08:46
completely separate from and won't
01:08:48
cooperate with the
01:08:50
CCP this bill is a better is better than
01:08:53
the last bill we talked about that was
01:08:55
targeted at Tik Tok I don't know if you
01:08:57
remember that one but it was weirdly
01:09:00
prohibiting Americans from using VPN and
01:09:03
gave the government the right to go
01:09:04
after Americans who are using VPN so
01:09:07
this one seems cleaner and better and
01:09:09
more narrowly targeted at divesture now
01:09:12
at the beginning of my response to you I
01:09:15
did say if you know I know that
01:09:18
everyone's just assuming that it's true
01:09:20
that Tik Tok is sharing data with the
01:09:24
CCP but I just want to confirm that that
01:09:26
is the case because they are denying it
01:09:29
and I can understand why people think it
01:09:31
and why it might even be likely but
01:09:33
since we do have a concept of due
01:09:35
process in America I do think some
01:09:38
evidence should be provided that that
01:09:39
actually is taking place well we've had
01:09:43
you know whistleblowers inside there but
01:09:45
chth if the CCP is spying on their own
01:09:47
citizens what are the chances that they
01:09:49
wouldn't take the opportunity to spy on
01:09:52
government officials who have Tik Tok on
01:09:53
their phone or their kids and get
01:09:56
comprom mod on them possibly or no
01:09:58
locations of people what does your gut
01:10:01
tell you is a CCP is this too dangerous
01:10:03
for us to have here in America under CCP
01:10:05
control or this kind of influence being
01:10:07
on their board I think you're asking the
01:10:10
exact right
01:10:12
question I have two comments to make one
01:10:16
is Nick if you can bring up this article
01:10:19
about the Google AI IP case basically
01:10:23
what happened was that
01:10:25
doj filed an
01:10:28
indictment actually I think I sent I I
01:10:30
think I sent you guys the actual
01:10:33
indictment but essentially what happened
01:10:35
is there was an engineer at Google that
01:10:37
has been charged crazy that has been
01:10:39
charged with stealing AI secrets for
01:10:42
China and I don't know if whether he's
01:10:44
back in China now or not but the whole
01:10:47
point is that if it's happened at Google
01:10:50
right where there's a motivation for the
01:10:53
Chinese intelligence apparatus and
01:10:55
frankly every intelligence apparatus to
01:10:57
infiltrate that organization and get
01:10:59
access to all kinds of
01:11:02
data I think we should presume by
01:11:05
default that all of these organizations
01:11:07
are infiltrated and I think that that's
01:11:10
probably a more conservative and
01:11:11
reasonable posture so Facebook is
01:11:13
infiltrated Google is infiltrated apple
01:11:15
is infiltrated Tik Tok is infiltrated so
01:11:19
on that Dimension I think that it should
01:11:22
be considered 100% certainty that this
01:11:25
data is getting back to not just the
01:11:27
Chinese but multiple state sponsored
01:11:29
actors so the question about Tik Tok
01:11:33
then I think should be one of business
01:11:35
and I think Paul mer lucky did a very
01:11:37
good job of simplifying this down to its
01:11:40
Essence which is essentially what he
01:11:42
called The Law of Equivalent Exchange if
01:11:43
you want to just play this it's like
01:11:45
just a few seconds I was kind of
01:11:46
frustrated that people made Tik talk
01:11:48
into a cultural issue by the way I'm
01:11:50
totally on the culture War side of it uh
01:11:53
but but but but I was saying practically
01:11:55
speaking you should not make this a
01:11:56
culture War issue don't talk about how
01:11:59
it's ruining our you's ideals just say
01:12:01
strictly on a trade basis yes we cannot
01:12:04
allow them to sell this thing to us if
01:12:08
we can't sell the same thing to them
01:12:10
like that that should be totally fair
01:12:11
that's reciprocity that's reciprocity me
01:12:14
and Palmer lucky and Sy right so this is
01:12:16
what he calls the Law of Equivalent
01:12:17
Exchange and I think it just makes a lot
01:12:19
of sense so on the face what I would say
01:12:21
is Jal my response is I think that the
01:12:23
CCP but also other intelligence
01:12:26
organizations have infiltrated all of
01:12:28
these big companies and all of our data
01:12:31
is accessible by them I'm not going to
01:12:34
say on a whim but I think it's
01:12:36
accessible I think you have to deal with
01:12:38
Tik Tok as a business issue and I agree
01:12:40
with Palmer lucky which is they should
01:12:42
not be able to sell to us what we cannot
01:12:44
sell to them and I think that that's a
01:12:45
fair principle that we can live on Resto
01:12:47
is a very simple position freedberg let
01:12:51
me use your creativity
01:12:53
Your Love of Cinema if you were to use
01:12:56
this tool let's take the most cynical
01:12:58
approach here or interpretation CCP has
01:13:01
complete access to the algorithms and
01:13:03
they want to do maximum damage let's say
01:13:05
during the election let's say in a
01:13:07
conflict like the one going on in
01:13:10
Ukraine with Russia or in
01:13:14
Gaza what could they do using the
01:13:17
algorithm using videos what would be the
01:13:20
Doomsday scenario for America as in the
01:13:23
C CP comes in and influences the
01:13:24
management of this company and tells
01:13:26
them what to tweak and how and why yes
01:13:29
content what would they do I think we
01:13:31
saw this
01:13:33
during after October
01:13:35
7th that there was a significant surge
01:13:40
in Pro Hamas videos relative to Israel
01:13:45
support videos that's the sort of thing
01:13:47
where you could kind of see something
01:13:49
that sets an
01:13:50
opinion that maybe disruptive to the
01:13:54
social fabric to the election cycle that
01:13:57
starts to get shared more frequently and
01:13:58
show shows up in feeds more frequently
01:14:01
unlike Facebook and other places where
01:14:03
there's a linear feed where you can
01:14:04
scroll up and select what you want to
01:14:05
watch as you know Tik Tok has already
01:14:08
lined the videos up so when you scroll
01:14:09
up they've they automatically play the
01:14:12
next thing for you so the ranking really
01:14:14
matters in terms of viewership uh on Tik
01:14:16
Tok unlike a lot of other kind of Select
01:14:19
to playay social media type sh an
01:14:22
election
01:14:23
hearts and Minds in a war well I've
01:14:26
always said this I think it's the
01:14:27
craziest thing in the
01:14:28
world that someone can spend advertising
01:14:31
dollars and change someone's vote like
01:14:34
that like just think about that fact for
01:14:35
a second I mean I've said this before
01:14:37
and people have told me I'm an idiot for
01:14:39
saying it but meaning that you can or
01:14:41
that it does that you that both that
01:14:44
that
01:14:45
that think about it like like people
01:14:47
don't individually go and gather data
01:14:49
and then make an informed opinion about
01:14:51
who they're going to vote for their opin
01:14:53
changes based on seeing an ad it is so
01:14:56
crazy to me that that's the truth I
01:14:58
think the I think the country knew that
01:14:59
it does that's why they prevented it
01:15:01
from happening
01:15:02
until yeah exactly but I mean I think
01:15:05
that's what's so nuts is that there's no
01:15:08
longer a forced discourse that kind of
01:15:12
makes people go out and choose what
01:15:14
content they want to consume what they
01:15:16
want to hear debate stages Etc that now
01:15:18
it's about who spends the most money to
01:15:20
get the most views in front of someone
01:15:22
and that that actually influences
01:15:23
someone's decision on who to vote for is
01:15:25
what's so compelling to me about why all
01:15:29
of these systems have such extraordinary
01:15:30
power it's just so amazing to me that
01:15:32
the more frequently someone sees an ad
01:15:34
the more likely they are to buy
01:15:35
something or do
01:15:37
something here's the bottom line this
01:15:39
the more frequently you show someone a
01:15:40
AIC on Tik Tok the more likely they are
01:15:43
to vote something differently this thing
01:15:44
is far too powerful for the CCP to have
01:15:47
any kind of access to it for the Chinese
01:15:49
government to have any kind of access to
01:15:50
it it has to be divested if you look at
01:15:53
what we went through in the last couple
01:15:54
of election Cycles not being partisan
01:15:56
here at all but we've said and we've
01:15:59
talked about Hunter Biden's laptop as
01:16:00
but one example you know that was
01:16:02
suppressed obviously on social networks
01:16:05
that could have been Amplified in social
01:16:06
networks and it could have had the
01:16:08
opposite effect Hillary Clinton's emails
01:16:10
you know Trump this Hillary this Biden
01:16:13
that you you could really sway an
01:16:15
election by putting in specific subtle
01:16:19
information let alone taking a hack and
01:16:23
relasing it like they did with Hillary
01:16:24
Clinton's email so you know those things
01:16:26
might not have swayed an election but
01:16:28
sax if Putin had access to this like
01:16:30
somebody who's super capable and he had
01:16:32
access always where it goes no I just
01:16:36
secretly pulling the strings of our
01:16:38
election okay Iran this is like the
01:16:40
biggest threat inflation ever okay look
01:16:44
if you ask if you ask people out there
01:16:47
do you think other people's votes are
01:16:49
influenced by social media they'll say
01:16:50
yeah of course people are brainwashed if
01:16:51
you say to them is your own vote
01:16:54
influenced by social media do you make
01:16:56
up your own mind based on all the
01:16:58
information you have they'll say yeah of
01:17:00
course I'm not brainwashed everybody
01:17:02
else is and I I believe that people are
01:17:04
closer to telling the truth when they're
01:17:06
talking about themselves because they
01:17:08
understand their own situation better
01:17:10
than they understand everyone else's
01:17:11
situation the fact of the matter is that
01:17:13
all of us are constantly bombarded with
01:17:16
information 24 hours a day 7 days a week
01:17:19
through all of the channels both online
01:17:21
and offline where we get information
01:17:23
some of those data points come from
01:17:26
advertisements but I don't think we take
01:17:28
ads very seriously we're trained to kind
01:17:30
of even just block out the ads when I
01:17:33
see banner ads or even ads in my stream
01:17:35
I just like scroll past them they don't
01:17:39
even factor in my Consciousness I am
01:17:42
influenced by accounts that I follow but
01:17:45
there are accounts I've chosen to follow
01:17:46
because I think they have signal over
01:17:48
noise and the more noisy those accounts
01:17:50
are the more I disregard them and and
01:17:52
take them them less seriously with
01:17:54
advertisements being the most noisy and
01:17:57
least useful channels so look I think at
01:17:59
the end of the day this idea that we're
01:18:00
all being brainwashed and secretly
01:18:02
influenced by malign foreign actors I
01:18:04
think is at a minimum threat inflation
01:18:07
and that entire narrative might just be
01:18:09
completely
01:18:11
bogus
01:18:13
nonetheless I do agree that for data
01:18:16
collection reasons and reciprocity
01:18:18
reasons I think it's like I said we're
01:18:20
within our rights to require the
01:18:23
divestment of of Tik Tok I you know s
01:18:26
you don't make this into more than it is
01:18:29
again I think this whole disinformation
01:18:31
narrative by the way you want to know
01:18:32
why they push it so hard it's because
01:18:34
our own intelligence Community wants to
01:18:38
be involved and there are political
01:18:41
actors in the US who want to regulate
01:18:43
quote disinformation on our own social
01:18:46
networks and I'm not talking about Tik
01:18:47
Tok I'm talking about X I'm talking
01:18:49
about Facebook insta and so on and we
01:18:51
saw this in the Twitter files we saw
01:18:53
what a cozy relationship the
01:18:55
intelligence Community had with Twitter
01:18:57
they were all up in there trying to
01:19:00
control what Legacy Twitter was
01:19:04
allowing URL they blocked the New York
01:19:06
Post URL and you've said on this program
01:19:08
that you believe that could have swung
01:19:10
the election well I think it was that I
01:19:12
think that actually was genuine election
01:19:14
interference I actually said I didn't
01:19:16
know whether it could swing the election
01:19:17
I think that you think could
01:19:20
have I think it was a story that deserve
01:19:23
be publ published so that and
01:19:25
distributed online so that the public
01:19:27
could take that into account when they
01:19:29
voted I mean and so if it had been
01:19:31
widely received do you think I'm asking
01:19:33
a question not answering it okay you're
01:19:37
you're asking me a question that I
01:19:38
personally think is irrelevant how would
01:19:39
I know whether the
01:19:42
suppression of that story swung the
01:19:44
election how could I know that the point
01:19:46
is that the American people were
01:19:49
deprived of information that they had
01:19:53
Believe come it would have swung the
01:19:55
election well I can't prove I can't
01:19:58
prove that it swung the election what
01:19:59
I'm saying is that it was a type of
01:20:01
election interference why was that story
01:20:03
suppressed because 51 former
01:20:05
intelligence officials who still
01:20:07
maintain close relationships with the
01:20:08
intelligence Community publish a bogus
01:20:10
letter saying that it was Russian
01:20:12
disinformation which it wasn't and then
01:20:14
that caused our social media sites to
01:20:16
suppress it so that to me is as
01:20:20
concerning if not more concerning than
01:20:23
whatever it is that Tik tok's accused of
01:20:25
so I don't want social media companies
01:20:28
being used by the intelligence
01:20:30
communities of either China or the
01:20:33
United States to swing or to influence
01:20:35
our elections and we need to be equally
01:20:38
concerned about that as we are about
01:20:41
supposed Chinese influence just to go
01:20:43
back in time you probably remember the
01:20:44
Willie Horton ad I mean that kind of
01:20:46
sunk dakus if you remember that I mean
01:20:48
media and these ads can really have a
01:20:50
big impact so people should just look
01:20:52
historically there have been many
01:20:53
moments where whether it's Nixon
01:20:55
sweating on TV or the wiie Hort Ed there
01:20:57
have been many moments where video can
01:20:59
do this and I think the the they could
01:21:01
be done even more subtly by the Chinese
01:21:03
by just you know promoting certain
01:21:06
videos all right let's move on to
01:21:07
bitcoin issue five bitcoin's back baby
01:21:10
it hits an all-time high
01:21:13
69,000 interesting number on Tuesday
01:21:15
69,000 before dropping sitting around
01:21:18
68k as we're taping it be 75 by the time
01:21:22
you hear this or 50 it's been ripping in
01:21:25
2024 of
01:21:27
70% since January 25th uh there's two
01:21:31
things going on here uh that you've
01:21:32
probably heard about the Bitcoin ETFs
01:21:34
finally arrived in the US they were
01:21:36
approved by the SEC on January 10th
01:21:38
we'll get into that in a minute
01:21:40
moth obviously it's an ETF it's super
01:21:43
easy to buy them and sell them they've
01:21:44
been a total hit black rocks Bitcoin ETF
01:21:47
became the fastest ETF to ever reach 10
01:21:50
billion in assets also as the technical
01:21:53
crypto heads in the audience know there
01:21:55
is a having happening in April this
01:21:58
happens about every four years at the
01:22:00
current Pace last one was May of 2020
01:22:02
when these
01:22:03
halfings happen the mining rewards are
01:22:06
cut in half this reduces the supply of
01:22:08
new Bitcoins entering circulation and
01:22:10
can cause some swings shath uh we had a
01:22:14
we released a clip I think um just about
01:22:16
your prediction and you nailed it again
01:22:20
you said this would be a big year for
01:22:21
Bitcoin so your thoughts on being
01:22:25
right I don't have much thoughts on that
01:22:27
but my two comments are that I talk to a
01:22:30
lot of Bitcoin Traders and and folks
01:22:33
that seem to have a very good pulse and
01:22:35
touch on this Market I don't say that I
01:22:38
I do because I don't really look at it
01:22:41
every day but they seem to think that
01:22:43
this thing is on a death march to 100K
01:22:45
I'm not sure whether that price is
01:22:48
realistic or not in the year but I will
01:22:49
say
01:22:51
that we're going to get to a Tipping
01:22:53
Point where everybody really talks about
01:22:55
this I still don't think we're there yet
01:22:57
I think we're just at the beginning but
01:22:59
when you see the inflows into these ETFs
01:23:02
jakal it's like a very big deal because
01:23:04
it just allows every mom and pop
01:23:07
individual to buy some to the extent
01:23:09
that they want to own it or they want to
01:23:11
speculate on it whatever it
01:23:14
is so I I think it's been a very big
01:23:16
year and I think that psychologically
01:23:19
it's proven a lot of folks wrong
01:23:23
and it's a setup for something really
01:23:25
constructive the other thing I'll say is
01:23:28
that it's not just Bitcoin but as goes
01:23:31
Bitcoin there are a handful of other
01:23:34
things people are now speculating that
01:23:36
there's going to be an ethereum ETF that
01:23:38
gets approved as well because you know
01:23:40
if you approve one there's probably
01:23:42
legitimate cause to approve a few others
01:23:44
so these things are becoming part of the
01:23:46
financial Fabric and I think that that
01:23:49
should not be underestimated and the SEC
01:23:51
is still taking action against certain
01:23:53
Bad actors in crypto there were a couple
01:23:55
of those this week but Bitcoin fredberg
01:23:57
is incredibly resilient just on a
01:23:59
technological basis the fact that it
01:24:02
hasn't broken down under stress it
01:24:04
hasn't had you know a denial of service
01:24:07
type of
01:24:08
attack or a government hasn't been able
01:24:11
to capture 51% of the mining or or some
01:24:15
great amount of it or just even be
01:24:17
hacked in any way you have to be
01:24:19
impressed by the the fundamental
01:24:21
technology here free maybe you could
01:24:22
speak to that level of success that this
01:24:25
thing is so stable and trustworthy and
01:24:28
reliable to date it's you know it's got
01:24:31
a great incentive model as long as
01:24:32
bitcoin price remains High the miners
01:24:34
will still be there and the system will
01:24:35
keep running if bitcoin price
01:24:39
drops transaction fees will decline the
01:24:42
value of mining will
01:24:43
Decline and you know it kind of goes the
01:24:47
other way as well I think the real
01:24:49
question is in the last couple of years
01:24:51
have we really seen seen a change in
01:24:53
Bitcoin being used for transactions or
01:24:56
for Commerce in any meaningful way I
01:24:59
think the answer is still likely not no
01:25:01
definitely not and it's really a stored
01:25:03
value system and it's become this kind
01:25:04
of stored value Asset freeberg this
01:25:07
microplastics thing we talked about it
01:25:09
on the show and uh since that time I
01:25:13
refused to open plastic bottles I'm
01:25:15
doing all glass I'm getting rid of all
01:25:17
this goddamn plastic I I already did
01:25:19
glass bottles in my house cuz I'm cheap
01:25:21
and I like to fill it from my water
01:25:22
filter
01:25:23
but we're uncovering more information
01:25:26
and then I saw this headline this week
01:25:28
that microplastics are in our blood
01:25:30
streams in some cases and what the heck
01:25:34
does that mean it's worse than that team
01:25:36
of scientists in Italy collected
01:25:39
samples from patients that had plaque
01:25:42
removed from their cored artery it's a
01:25:44
kind of common cardiac procedure where
01:25:48
you get plaque that blocks up in your
01:25:50
cored they go in they removed the plaque
01:25:53
so a total of 304 patients agreed to
01:25:56
have the plaque that was removed from
01:25:58
their artery submitted for analysis and
01:26:01
then what this team did is they took
01:26:02
that plaque and they studied it to
01:26:05
see how much plastic was found in that
01:26:10
plaque and they used a bunch of
01:26:12
measurement techniques to do this
01:26:14
including electron microscopy and mass
01:26:16
spec so they because it's really hard to
01:26:18
find these molecules and microplastics
01:26:21
or nanoplastics remember are less than 5
01:26:23
mm in size with a mean level of 21
01:26:28
microG per mg of plaque roughly 1 per 50
01:26:32
is the ratio of plastic to plaque that
01:26:35
they found which is really incredible
01:26:37
because it shows that Plastics these
01:26:39
little Nano and microplastics are
01:26:40
accumulating good or incredible bad
01:26:43
incredible bad that these micro Plastics
01:26:45
these nanoplastics are accumulating in
01:26:47
the human body now here's the the scary
01:26:49
part they then did a followup 34 months
01:26:52
later the patients that had plastic in
01:26:55
their blood had a four and a half times
01:26:57
higher Ratio or likelihood of having
01:27:00
heart attack stroke or death from any
01:27:03
cause so all of these major health
01:27:06
effects were four and a half times
01:27:08
elevated in patients that had plastic in
01:27:10
their blood this was published in the
01:27:12
New England Journal of Medicine if I
01:27:13
didn't say it it really indicates that
01:27:15
there is this kind of cumulative problem
01:27:18
and that the cumulative problem is
01:27:19
likely leading to really adverse health
01:27:22
outcomes and I'll just highlight one
01:27:24
other paper from a team in Germany and
01:27:26
Norway back in May of 2022 and this team
01:27:29
tried to figure out how Plastics are
01:27:31
causing adverse health effects in the
01:27:33
body and they had a theory like let's
01:27:35
put little microplastics or nanoplastics
01:27:37
together with all the human cells that
01:27:39
we know shake it up and see what happens
01:27:41
and what they found was that these
01:27:43
little plastic fragments were binding to
01:27:45
dendritic cells and monocytes key cells
01:27:47
in the immune system and when those
01:27:50
cells were um Bound by PL IC they
01:27:53
release these cyto and the
01:27:54
pro-inflammatory signals go through the
01:27:56
roof it causes the immune system to go
01:27:58
Highwire increases inflammation and the
01:28:01
cascading effects of that obviously can
01:28:03
ultimately lead to many of the events
01:28:05
that we're mentioning were measured in
01:28:06
this set of patients in Italy so um
01:28:09
again we're just starting to uncover
01:28:11
these effects this concept that
01:28:13
microplastics and nanoplastics that are
01:28:15
accumulating let me just say these
01:28:17
Plastics are mostly pet which is what we
01:28:19
use to make plastic bottles that we
01:28:21
drink water and drinks out and PVC or
01:28:23
polyvinyl chloride which is what a lot
01:28:25
of our plastic plumbing and piping is
01:28:27
made from and so as little tiny bits of
01:28:30
these plastic materials either are
01:28:31
exposed to sunlight and break off and
01:28:33
end up in our water and food supply and
01:28:35
we consume them they are slowly
01:28:37
accumulating in in bodies and they may
01:28:39
be driving inflammatory response they
01:28:41
may be driving adverse Health outcomes
01:28:43
we're really kind of tip of the iceberg
01:28:45
and really studying this understanding
01:28:46
and analyzing it but here's another
01:28:48
really interesting empirical data set
01:28:50
that highlights that this really is
01:28:52
um pretty significant half the patients
01:28:55
had it and of that half they had a four
01:28:57
and a half times higher chance of dying
01:28:59
or having a heart attack or a stroke in
01:29:02
the 34 months that followed yeah the
01:29:04
scary data the thing the thing that that
01:29:06
study said which was nuts is it it
01:29:08
looked like the the nanop particles the
01:29:10
nanoplastics and microplastics were
01:29:12
effectively acting as scaffolding for
01:29:14
plaque so in in almost like it was a it
01:29:18
was a shim that allowed it to grow the
01:29:20
question is would it grown faster than
01:29:22
it would have otherwise that's even
01:29:24
scarier
01:29:26
so I did not like reading that paper
01:29:30
that really freaked me
01:29:32
out really really freaked me out I was
01:29:35
drinking water from plastic water
01:29:36
bottles this week and time I drink water
01:29:39
out of a plastic bottle now I'm like
01:29:41
like nervous every time I take a Si
01:29:43
you're not supposed to double your the
01:29:45
risk of all cause mortality by drinking
01:29:47
yeah aiji water you know what I mean
01:29:49
four and a half x four and a half X it's
01:29:51
crazy in 34 months think about the
01:29:53
cumulative effect over time long perod
01:29:56
imagine drinking water out of a plastic
01:29:58
bottle thinking you're doing the right
01:30:00
thing and then trotting over to the
01:30:01
recycling bin you know yeah and you do
01:30:04
that for 20 years you may be killing
01:30:08
yourself well there it is so wait what's
01:30:10
the if you if you if the water glass
01:30:13
bottles you must use you cannot use
01:30:16
plastic you just can it's over no
01:30:19
plastic no plastic you got to stop it's
01:30:21
over we're drinking it's done glass cans
01:30:24
good plastic stainless steel stainless
01:30:27
steel is fine just like I mentioned when
01:30:29
we talked about this a few weeks ago the
01:30:32
carbon footprint the environmental cost
01:30:34
the cash cost is much higher with all
01:30:35
these alternatives to plastic so there
01:30:37
are big challenges with respect to
01:30:39
having some big massive response to
01:30:42
Plastics used in our supply but you know
01:30:45
living in the luxury world that we all
01:30:46
get to live in we get to have that
01:30:48
choice and we'll make that choice but
01:30:50
it's a real problem for Humanity because
01:30:52
Plastics are so ubiquitous in so many
01:30:54
things and they've they've enabled
01:30:56
they've enabled affordability of
01:30:57
consumer goods this is such
01:30:59
honestly like all you have to do is have
01:31:01
glass bottles or carry a water bottle
01:31:03
with you like I have a Contigo one I
01:31:06
like I carry it with me I empty it I it
01:31:10
I have a water filter in my house and we
01:31:12
fill water bottles and put them in the
01:31:14
fridge no but what if you like yogurt
01:31:15
yogurt comes in a plastic container
01:31:17
there's all kinds of stuff you can't
01:31:19
avoid plastic that's so scary we try to
01:31:21
that's so scary we we do have the French
01:31:23
yogurt that comes in glass bottles but
01:31:25
yes it is hard yes we do the French
01:31:26
yogurt in
01:31:28
glass where do you get it from there's a
01:31:31
French yogurt that comes in
01:31:34
glass it's called
01:31:36
like or something they been while no
01:31:38
water water you just install a filter
01:31:40
system which you have at the mum and
01:31:42
just fill huge glass bottles just have
01:31:46
your staff fill glass bottles and put
01:31:47
them in the fridge and don't throw them
01:31:49
away and give your kids like some of
01:31:51
these thermoses or whatever but don't
01:31:52
have the water bottles like you know
01:31:54
it's so funny it's been a disaster in
01:31:56
the poker game in some ways we got rid
01:31:58
of it and there's been way more broken
01:32:01
glass people knock over the you know the
01:32:04
the side tables I get it it's been a
01:32:06
huge pain but I will not go back no it's
01:32:09
the right thing to do absolutely the
01:32:10
right thing to do okay well if if jamath
01:32:13
is endorsing this I guess I'm gonna take
01:32:15
it seriously I think you got to do it my
01:32:18
opinion doesn't matter with sex I'm not
01:32:19
saying you're wrong I'm just saying that
01:32:21
your threshold for becoming concerned is
01:32:25
lower and then if Cham if it hits Chamas
01:32:27
threshold Which is higher I'm gonna take
01:32:30
it more seriously as your besti I would
01:32:33
like you to stop using plastic for the
01:32:34
rest of your life okay you're in a
01:32:36
position to do it I would ask you to not
01:32:38
do it all right everybody Welcome can
01:32:40
all you fers come back to the B so we
01:32:43
can see each other absolutely I miss you
01:32:44
guys all right love you guys I miss you
01:32:47
I miss you guys too for the Sultan of
01:32:50
science the chairman dictator
01:32:52
and the Rainman David Sachs I am the
01:32:55
world's greatest moderator okay love you
01:32:57
guys I got to go love see you next time
01:33:01
bye let your winners
01:33:08
ride and instead we open source it to
01:33:10
the fans and they've just gone crazy
01:33:12
with it love queen
01:33:15
[Music]
01:33:20
of besties
01:33:23
are that's my dog taking your
01:33:28
driveway oh
01:33:31
man we should all just get a room and
01:33:33
just have one big huge orgy cuz they're
01:33:35
all this useless it's like this like
01:33:36
sexual tension that we just need to
01:33:38
release
01:33:39
[Music]
01:33:44
somehow we need to get merch
01:33:50
our
01:33:55
I'm going all
01:33:56
[Music]
01:33:58
in
