00:00:00
who's your number one Who's your number two if you had to could only bet on two here to achieve super intelligence AGI
00:00:06
let's just say win the AI re win the AI uh big prize the big prize super intelligence AGI you know in the midterm
00:00:15
five years five years from now we're sitting here Thomas give me your number one give me your number two look I I
00:00:20
think to me number one I I still think Nvidia right I don't see the GPU kind of getting displaced I see additional
00:00:26
architectures kind of coming on board right and growing the market but um at the end of the day all roads still lead
00:00:32
to the GPU for all of these models So I would kind of still put um kind of Nvidia on that My number two more of a
00:00:41
dark horse but I I would pick Tesla I do think it has the most potential for vertical integration right From all the
00:00:49
way the silicon to the model to actually the hardware right That might become super important not just in cars but in
00:00:55
Optimus So Nvidia 1 Tesla is my dark horse Wow stunning Chimath who's your number one and number two In the midterm
00:01:04
5 years from now we're sitting here on AllIn episode 700 Tesla's one and Google's two And the reason is because
00:01:12
they are the closest to having that vertically integrated stack that I spoke about I think that Tesla has the best
00:01:20
vision models Now with XAI they'll have one of the best LLMs and reasoning models and they'll be able to eventually
00:01:29
stick that on Dojo And then all of that will be in all of the physical AI that you will interact with in your daily
00:01:37
life whether it's a robot or whether it's a car or whether it's a robo taxi So that's number one And then number two
00:01:44
for many of the same reasons I think Google because you'll have the Gemini family of models which just absolutely
00:01:51
kickass like VO3 which we haven't really spoke about is going to destroy Hollywood like in the next year Like
00:01:58
Hollywood is done I think but they're landing model after model They have the TPU and the next generation TPU I think
00:02:06
is exceptional They're baking quantum and then they have an entire funnel of billions of people that they can direct
00:02:13
experiences to So Tesla one Google 2 Chimath quick followup on that I'm curious on Google This is the because I
00:02:21
I oscillate a lot on this particular name Can Google win if search declines Yes And I think that what probably has
00:02:29
to happen is bear with me when I say this but if you had to boil down Google's economic northstar metric right
00:02:37
not the value northstar the economic northstar metric would be price per click And I do think that Google is extremely
00:02:46
well positioned to pivot that to price per token And I think that they have some emerging classes of physical AI but
00:02:54
they have the largest pool of people where they can generate a price per token value framework through YouTube
00:03:01
through Gmail through Workspace I think through search but probably it's a different kind of model It just requires
00:03:08
them to rip the band-aid off at some point But yeah I think Google can do it I'm going to go with you Chimoth I'm one
00:03:15
uh my one and two are either Google uh or Elon And I I'll just say Elon because I uh like you I spent a day up at um XAI
00:03:24
and I saw what a magnet for talent he is I got to sit in some meetings and just he was interviewing people and he was
00:03:31
working with that talent 8:00 at night There's a lot of people there on a Saturday grinding it out It was nuts I
00:03:37
first went to XAI In the 15 minutes that I was in the parking lot finishing a call the kinds
00:03:42
of people that were walking in and out of there you could tell they were big brains Yeah I don't know how you know
00:03:48
what I mean Like from every walk of life they all just looked much smarter than the rest of us Yeah It some of them were
00:03:54
like chain smoking cigarettes and just like stressed out It was crazy I hit a couple of zins I'll be totally honest Um
00:04:00
but the reason I say Elon versus Google is I think Elon's in a unique position I
00:04:05
don't have any insider information here and and I haven't talked about this or I'm not back channeling from Elon lest
00:04:10
anybody aggregate this I think what Colossus has done and what Tesla has done both of these things Tesla with
00:04:17
their own stack of hardware to your point Shimoth hardware plus software plus the user application of FSD and
00:04:23
Optimus Then you put that together with the data the real-time data of X formerly known as Twitter plus um you
00:04:31
know what he's building with XAI and obviously those two companies merged I think Tesla board XAI board have to get
00:04:36
together put those two companies together One's worth a trillion one's worth 100 billion Put them and just have
00:04:42
all that brain power going in one direction as opposed to Elon T switching between the two You do that I think he
00:04:48
wins number one you don't do that I think he either gets one or two and then I think Google um is going to have a
00:04:55
better search product Thomas I think it's a really important point Do they lose search share Doesn't matter What I
00:05:01
think matters is are their ads more effective Is their ad network more effective And I think based on what they
00:05:05
know on you from your chat searches and your discussions and what they analyze in your email just analyzing your Gmail
00:05:13
and your surfing behavior and Chrome if they get to keep it your Android phone if you use it your YouTube list and what
00:05:19
you how when you drop off allin and when you start listening to another podcast whatever it is all that data all that
00:05:24
data is going to lead to an ad network that performs so much better that even if they lose search hair their ad
00:05:30
network is going to continue to grow and I think it will increase in velocity So
00:05:33
those are the my top two Freeberg I'm curious from your position which one you think is number one and
00:05:40
number two I saved you for last because you know what we do here We save the best for last Free part go ahead I think
00:05:45
there's a difference in how I would kind of lump them I I think that Tesla probably has the it is the best place to
00:05:55
invest if you want to have a shot at a massive new industry So they've got a baseline business in in obviously the
00:06:03
automobiles but I think this humanoid robot opportunity is absolutely mind-blowingly ginormous And I don't
00:06:12
think that there's a better company on Earth positioned to execute against this humanoid robotics opportunity than Tesla
00:06:20
So you know it's sort of like I would call it a low probability high upside sort of call option embedded within that
00:06:28
business And obviously you're paying a premium for that because it is still a very healthy premium you pay for that
00:06:32
business I think Nvidia to Thomas's point I think the common thesis is it is the most protected The durability of the
00:06:41
business is there But I would argue that there's actually a low probability but very high severity risk to Nvidia in
00:06:47
China There was just a demonstration last month of a 1 nanometer semiconductor manufacturing process out
00:06:53
of China I think the more that we continue to try and isolate China from a policy perspective the more we are
00:07:01
emboldening investment in China meaning from the government from private industry into China to create
00:07:07
alternatives to the chip stack where the United States companies particularly Nvidia have emote today So I do think
00:07:15
that there's going to be an emergent competitive threat coming out of China to Nvidia And just like we were knocked
00:07:20
over by DeepSeek I think we will be knocked over by some semiconductor manufacturing processes um coming out of
00:07:27
China in the near term But the overall kind of by the way Dave just on that point I think Sax's work on the
00:07:34
diffusion rule just generally I don't think has kind of gotten enough attention in the rescending of the
00:07:39
diffusion rule which essentially handicapped our ability to even arm our allies right with our sem with our
00:07:48
semiconductor technology um in my opinion was kind of a milestone and very important moment um to to try and offset
00:07:56
exactly what you were just describing That's exactly right I mean there there there was a report a few months ago and
00:08:01
I mentioned it on the show or maybe I didn't or maybe I sent it to Sax and we talked about it offline I I can't
00:08:06
remember but it was about a $40 billion investment being made in developing competitive semiconductor manufacturing
00:08:12
full stack solutions out of China So I I do think that the lithography IP moat is
00:08:19
being crossed in China I do think that China is developing actually new technology for uh DUV and EUV systems I
00:08:27
I do think that there's a risk uh to Nvidia's core Now look Nvidia is such a durable business There's great modes
00:08:33
great advantages but we're creating every incentive for an alternative to Nvidia to emerge from China And then my
00:08:39
my third kind of categorization would be what's the portfolio uh solution I think
00:08:44
that's Google I think that there's a diversification of high beta bets inside of Google of any one of which could have
00:08:52
call it a trillion dollar market cap outcome ranging from Whimo to quantum computing to the biologics work that uh
00:09:00
Demis is working on out of um isomorphic Uh there's a number of things that do not get a lot of attention at Google So
00:09:08
yes there's a there's a core business that that may be at risk Thomas but I think that there's a a portfolio of
00:09:14
options you get at Google and you just need any one of them to hit to kind of make up for the loss But I do think also
00:09:21
Sundar in my interview with him which we put out a couple of weeks ago is very thoughtful about where search evolves to
00:09:27
and he is being I think reasonably aggressive in in trying to evolve the search product architecture to meet the
00:09:34
market to meet the consumer I do give him credit for that So Google would be in a good place for me as an overall
00:09:40
kind of pick in that set of options So just to be clear Nvidia one Google 2 or Nvidia Tesla Like I said I think in
00:09:46
terms of like having the right sharp ratio is how I would think about it The alpha and the beta adjusted returns I
00:09:51
would put Google number one I would probably put Tesla Uh Tesla's valuation I think already has a premium associated
00:09:59
with those options So I don't know Yeah So I don't know if I would really pay that premium I think um aside from the
00:10:04
valuations let's take valuations out of it Just the the game here is who wins the AI prize 5 years That's how I
00:10:11
understood it as well Yeah So valuation irrelevant Valuation irrelevant Who wins
00:10:16
the AI prize One you're saying Google Two you're saying Tesla I think Google's in such a position I I mean look Deus uh
00:10:22
Demis I think has been fairly koi about where they are They obviously promote Gemini 2.5 but there's a lot still
00:10:30
coming And it's and and as Chimath pointed out it's not just LLMs There's a pretty sizable family of models
00:10:39
including a a lot of these um graph-based models that are being used in really novel applications that no one
00:10:45
else is even close to no one's spending time on I mean some of the weather forecasting it might seem small and
00:10:51
trivial but it's a demonstration of Google's competency in in core model development that shows an understanding
00:10:58
and a depth of research and work that goes well beyond LLM So I'm pretty bullish on the depth of talent the full
00:11:04
stack Yeah And whatever they learn there could apply to Gmail could apply to search could apply to ads could apply to
00:11:09
YouTube algorithm right It's just goes up and down Yeah Yeah From a product perspective I do think you see this kind
00:11:15
of multi-model emergence that that we're now seeing that no one talks about the single model that sits behind the
00:11:22
application There are multiple models that work together And obviously this agentic architecture unlocks another
00:11:30
layer of not just kind of solutions to complexity Sure And so there's there's quite a lot I think that's emergent here
00:11:37
um that Google will start to kind of benefit from uh in the year ahead