# No One Realizes What is Happening

https://www.youtube.com/watch?v=S6pzO6hQkhQ

[00:00] What's the actual competitive edge in
[00:02] the AI era? You might say it's the model
[00:06] or the algorithm, or you might say it's
[00:08] the data, but Alex Carb would say it's
[00:10] something much more specific and
[00:13] something much harder to copy. I want to
[00:15] walk you through three clips from some
[00:17] of the sharpest voices in tech and
[00:20] markets right now. We've got Alex Karp
[00:22] on data sovereignty and why most AI
[00:24] deployments are secretly working against
[00:27] the companies buying them. We've then
[00:29] got Kathy Wood on SpaceX and their
[00:31] advantages and then legendary investor
[00:34] Bill Aman on the investing model that
[00:36] makes sense of all of this AI growth.
[00:40] Let's get into it. How's going through
[00:42] many phases? First, we didn't have
[00:43] videos and we couldn't do videos and it
[00:45] was dental music and now it's kind of
[00:47] rocking. So, I don't know. I don't know
[00:49] what that means, but it's definitely
[00:51] better than the waiting for the drill to
[00:53] hit your tooth. Um
[00:56] I think one of the interesting things
[00:58] about Silicon Valley or first of all it
[01:00] is Silicon Valley, it's still Silicon
[01:03] Valley as much as we would like it to
[01:07] go all over the world. It's very seems
[01:10] to be very specific and it's where we
[01:12] started and I love to come home. Um but
[01:15] then the obvious thing is every single
[01:18] company has some kind of secret that
[01:22] they're trying to eviscerate. Uh, and in
[01:25] enterprise sales, that's mostly been a
[01:27] sales secret. Like, you think you're
[01:29] buying a widget, but it's buying you.
[01:32] You think you're getting the steak
[01:33] dinner, but you're paying. Um, and then,
[01:36] so I was thinking, well, what is our
[01:38] current secret we're hiding? And, uh, I
[01:41] have one. Uh, our sales, obviously,
[01:44] everyone knows we barely have sales. And
[01:46] Samir and his partner run it. and
[01:50] they're that his wife was actually
[01:51] asking, "Does Karp really think I'm an
[01:54] idiot?" And it's like, "No, I don't."
[01:56] Uh, and we do have sales, but we
[01:58] actually have the biggest sales and most
[02:00] effective sales force in the world. Uh,
[02:02] that we never talk about. And that's one
[02:04] of the main reasons we're bringing here.
[02:05] We're hoping that before, during, and
[02:07] after, you'll go to a large language
[02:10] model company and learn that they don't
[02:13] care about you at all. And no matter how
[02:16] bad this conference is, we can never
[02:19] sell you like they're going to sell you,
[02:21] which is we're going to give you
[02:22] something that makes you feel smart
[02:24] while your business goes out of business
[02:26] that doesn't actually work unless you
[02:28] have an ontology, foundry, Apollo, or
[02:30] something useful. We're going to show
[02:32] you we're smarter than you, which they
[02:34] may not be. and you're going to go home
[02:37] feeling poorer and less safe and you buy
[02:40] the product and you're going to pay a
[02:42] lot in tokens and it's going to be very
[02:44] hard to understand how it helps you but
[02:46] investors are going to know you're
[02:48] you're smart.
[02:50] That's our sales secret. uh please feel
[02:53] free to go and learn about that. And
[02:56] then what you will find is there are a
[02:58] myriad of problems that these very very
[03:01] important models solve. And there are
[03:05] even bigger problems that they create.
[03:07] And what are we in the business of?
[03:10] We're in the business of giving you the
[03:12] ability to solve those problems for
[03:14] yourself and own the means of
[03:16] production.
[03:18] And what are those problems? Well, you
[03:20] can write code or cheaper, faster, but
[03:23] what does the code base look like? Who
[03:24] does it? How do you make your engineers
[03:26] into for deployed engineers as opposed
[03:28] to people just writing code? How do you
[03:30] deploy that code in a secure, safe, fast
[03:33] way? How do you manage that asset so
[03:35] it's actually an asset? Asset is defined
[03:37] by it's valuable. What is valuable?
[03:39] Something that you own, that you have,
[03:41] that no one else has, that can't be
[03:43] outsourced. How is your data yours? What
[03:46] part of the stack belongs to a public
[03:48] cloud? What part of the stack do you not
[03:50] share? I.e. you own the exact part of
[03:54] the data that creates the alpha that you
[03:56] do not share with your competitor or
[03:58] anyone else. Your problem is exactly the
[04:00] same problem the Ukrainians have, the
[04:03] French have, the D department of war
[04:05] has, which is you have personal,
[04:08] private, very specific data that will
[04:11] and must belong that must belong to you
[04:14] that things trained on that data have to
[04:16] belong to you while simultaneously using
[04:19] public sources. How do you manage the
[04:21] code inside that? Who manages it? Who
[04:23] owns that asset? How do you deploy? How
[04:25] do you divide? How do you define and
[04:28] understand your company? The parts that
[04:30] how do you find vulnerabilities in your
[04:32] company? Uh, of course models are very
[04:35] useful for that. But can they do it at
[04:37] the pace you want? Can they fix the
[04:39] vulnerabilities? Who owns that
[04:41] information? How how do you quantify it?
[04:43] Does you owning it mean the person down
[04:46] the street owns it? These are the
[04:48] problems that we solved for the
[04:51] intelligence community and the defense
[04:53] community globally from Ukraine to
[04:56] Israel to America.
[04:59] You can decide which ones you like in
[05:01] River. We're very proud of our clients.
[05:03] We're very proud that we stick up for
[05:04] America. I'm very proud of the people
[05:07] who hate Palenter and me. Very proud of
[05:10] it. Put a list of them up there, please.
[05:13] uh you would be proud most of you too to
[05:15] be on our side when you look at the
[05:16] people who don't like us may not love
[05:18] you'll love us a lot more after you
[05:20] visit the labs and you look at the
[05:21] people who hate us um and encourage you
[05:24] to do it but these structural challenges
[05:27] were exactly the structural challenges
[05:30] that we solved at scale and we're in the
[05:34] business of transferring that to you now
[05:35] why do we do that well I mean we think
[05:38] we're likable but we obviously do it
[05:40] because it's a paro optimal situation
[05:43] These problems are real. They've been
[05:44] exposed. We do very well by solving for
[05:47] you. We can do them natively, meaning we
[05:49] get to keep our culture, our way of
[05:51] being, our way of working, the way we
[05:54] like to live and work. And you get to
[05:56] keep the same thing too at scale. And
[05:59] that's essentially what Palanteer is
[06:01] doing. It's the reason why we're doing
[06:04] so well, especially in the US commercial
[06:07] context. And it's the reason why you are
[06:10] going to fly. And the flying part is
[06:12] really interesting because you fly
[06:15] precisely because everything I just said
[06:19] is both true and specific and not
[06:21] accepted and that's how you do very well
[06:24] and that's what we're transferring to
[06:26] you and so yeah and we also just by and
[06:29] large with exceptions but very few
[06:32] really actually like our customers
[06:36] which most people don't many of you guys
[06:39] I mean I don't mean people in in this
[06:41] room, but definitely the valley. Um, and
[06:45] that ends up being a compounding and
[06:47] very important asset. So, with that,
[06:49] welcome. You're going to see a lot of
[06:51] actual things that work. Uh, and we do
[06:53] very little kind of show but don't, you
[06:56] know, tell but don't show and you can
[06:58] trade notes here and it's always a lot
[07:00] of fun and I'm very glad you guys came.
[07:02] Thank you.
[07:02] >> Alex Karp is describing a trap that many
[07:05] enterprise software companies are
[07:07] springing onto their customers right
[07:09] now. So KB's argument is actually pretty
[07:11] savage when you pull it apart. He's
[07:14] saying that the big AI model companies,
[07:16] the ones selling you API access and chat
[07:19] interfaces, have built a product that
[07:20] makes you feel like you're doing
[07:22] something, it makes you feel smart. It
[07:24] makes you feel modern. And then when you
[07:27] actually try to operationalize it, try
[07:29] to build your real business on top of
[07:31] it, you realize the thing doesn't
[07:33] integrate with your actual data or your
[07:36] actual systems. Palanteer's pitch is the
[07:39] exact opposite of that. They're saying
[07:41] your data is the competitive edge. And
[07:44] the only way AI makes your business
[07:46] better is if the AI is trained on your
[07:48] data, deployed in your environment, and
[07:51] then owned by you. That's what
[07:53] Palanteer's been doing for intelligent
[07:55] agencies for many years, helping
[07:57] organizations with extremely sensitive,
[08:00] extremely specific data actually use
[08:02] that data without giving that data away.
[08:05] And now they're selling that same
[08:07] capability to commercial enterprises.
[08:09] Alex keeps using this phrase, own the
[08:12] means of production. And what he's
[08:14] saying is that if your AI runs on
[08:16] someone else's infrastructure, trained
[08:18] on someone else's data mix, the thing
[08:20] generating your alpha isn't really
[08:23] yours. Your competitor has access to the
[08:25] same model. The edge basically
[08:28] evaporates. You've automated yourself
[08:30] into a commodity. I think he's right
[08:33] about the dynamic. But the question
[08:34] worth asking is when we plug into this
[08:37] AI system, what are we actually giving
[08:40] away and who else benefits from what we
[08:42] hand over? Alex Karp's been shouting
[08:45] this for years. The difference is now
[08:47] that more customers are starting to feel
[08:49] the problem firsthand. Quick word on
[08:52] something. If you run a service
[08:53] business, the dynamic Alice Carp is
[08:56] describing applies directly to how you
[08:58] build your client's trust, too. Many
[09:00] service businesses are still relying on
[09:02] referrals and hoping their reputation
[09:04] carries them. Utilizing YouTube creates
[09:07] a different model. By the time a
[09:09] prospect reaches out, they've already
[09:10] watched an hour of how you think, how
[09:12] you analyze, and how you make decisions
[09:15] because they've seen your videos. So,
[09:17] the trust is built before the first
[09:19] call. This means businesses that have
[09:21] YouTube channels find it easy to close
[09:23] the warm leads that come flooding in
[09:25] from people who watch their videos. My
[09:27] team at DC Social Media runs the whole
[09:29] YouTube operation for service businesses
[09:32] end to end if you think this could be
[09:34] valuables for yourself. There's a link
[09:36] in the description to find out more. So,
[09:38] Alex Carps essentially arguing that most
[09:40] companies are buying the wrong thing and
[09:43] SpaceX is a company that clearly figured
[09:45] out what the right thing is. Gathy Woods
[09:47] just made the case for why the market is
[09:50] still pricing SP X wrong. So speaking of
[09:52] of screaming past and and ex ascending
[09:55] rapidly um one of the companies uh
[09:57] probably the most notable one here uh
[09:59] SpaceX is u you know filed we've all
[10:02] read it's filing it's preparing to list
[10:04] Kathy you've had a lot of experience
[10:07] investing alongside Elon in his various
[10:11] enterprises uh and uh and here we have
[10:14] another one about to go. How do you
[10:16] think about kind of this company that is
[10:19] um clearly has amazing ambitions and um
[10:24] it may be received into a public market
[10:26] that is that is looking at a revenue
[10:28] base versus a valuation base that is
[10:30] very different from what they're used
[10:32] to. How do you think it'll be received
[10:33] and how do you think people should think
[10:35] about it? Well, judging from the kind of
[10:38] demand for SpaceX that we've seen in the
[10:41] private markets, uh I think it well
[10:45] exceeds the what is it 75 85 billion uh
[10:50] offering. Uh and so just like we've seen
[10:53] with another with other offerings, I
[10:56] think we will see a burst out of the
[10:58] gate. uh especially if uh if anyone's
[11:01] tuning into our models and our thinking
[11:04] about uh the impact of uh uh Starlink
[11:08] and orbital data centers and the moon
[11:12] and then ultimately Mars and and so
[11:14] forth. Uh so we have that long-term uh
[11:18] time horizon. Uh and then it's going to
[11:21] be very interesting. At some point it
[11:24] will reach a level and we don't know
[11:26] what that is and it might happen very
[11:28] quickly where you will have more
[11:31] traditional players in uh in the markets
[11:35] uh uh using valuation as a reason to
[11:38] sort of bash uh bash the stock. Um we've
[11:42] already seen it. Uh Morning Star
[11:44] apparently came out and said it's worth
[11:46] half half as much as this. Uh and of
[11:48] course we've got uh you know a very
[11:51] different story. Uh and the reason we do
[11:53] is because we have five analysts working
[11:56] on this. This is the the convergence uh
[12:00] story of our lifetime. Even Elon nine
[12:03] months ago started using this concept of
[12:06] wait a minute my companies are
[12:07] converging uh uh more and more quickly
[12:11] than I thought they were uh because
[12:14] obviously he had this orbital data
[12:16] center in mind. So, I think you'll get
[12:18] that initial burst and we'll certainly
[12:21] be out there educating, educating,
[12:23] educating about and we'll see what
[12:26] happens. The fact that it's going to get
[12:28] in the indexes as quickly as as seems to
[12:32] be the case. Uh the index providers are
[12:35] are discarding their previous guidelines
[12:38] for for uh for SpaceX.
[12:41] uh it depends how big of the index uh
[12:45] indexes SpaceX will represent that will
[12:48] bring in natural institutional support
[12:52] especially from those uh institutions
[12:54] who don't get as much as they want at
[12:57] the outset. So I do think there's a lot
[13:00] of support that way but at some point uh
[13:04] you know we will have the bare stories
[13:06] out there just like we have with Tesla
[13:08] and all of the other stocks uh we own
[13:12] practically.
[13:13] >> Yeah. Yeah. Disruptive innovation is by
[13:15] its nature both volatile and easily
[13:18] misunderstood. That's why it's
[13:20] disruptive. Um and you can think of you
[13:23] can imagine
[13:24] >> the the way the market kind of like does
[13:27] companies. It's like this is the company
[13:28] that we can see in the rearview mirror.
[13:31] This is the next year which the company
[13:33] is telling us what it's going to do. And
[13:35] then after that next year basically we
[13:37] just I cause things to fall back down to
[13:40] like normal average you know and that
[13:43] approach to SpaceX would you know I'm
[13:46] sure people will take that approach and
[13:48] it will dramatically
[13:51] um reduce your expectations for value in
[13:54] SpaceX versus kind of like understanding
[13:57] that the company has as far as we can
[14:00] understand and model um the opportunity
[14:03] for a very very high return on invested
[14:06] capital. The reason they're coming to
[14:07] the capital market, it's like they need
[14:09] to build rockets. If they build the
[14:10] Starship rocket, launch costs can fall
[14:12] from, you know, a thousandish per
[14:14] kilogram to less than 100 per kilogram.
[14:16] If they have a $100 a kilogram rockets,
[14:19] then they can pack them full of
[14:20] satellites. One Starship full of
[14:22] satellites can generate them
[14:24] >> roughly a billion dollars of revenue
[14:27] >> per year for five years. uh you know and
[14:30] so kind of like they can repeat that
[14:32] action until kind of they saturate the
[14:34] low earth orbit constellation and and
[14:37] they own they own the market they they
[14:40] own that market right they have a 10year
[14:42] lead on everyone
[14:44] >> and that's and that's the reason why
[14:46] it's rational that people say well
[14:48] there's growth this year but then after
[14:50] that I'm just going to decay it is
[14:52] because in most circumstances if
[14:54] companies are printing very amazing
[14:57] gross profits and return on capital. It
[14:59] means a sea of competitors is going to
[15:01] come in and eat all that up. And so it
[15:03] when that happens, then naturally your
[15:06] metrics decay because you you have are
[15:08] fighting off all the competitors. But
[15:10] SpaceX has there's no other credible
[15:14] competitor for rocket launch services at
[15:16] the cost that they can do it. Uh and the
[15:18] and the most the you know closest thing
[15:21] they just exploded their rocket to a
[15:23] million billion smithetheriness just
[15:25] last week. that I don't think that was
[15:26] actually good for SpaceX, but I do think
[15:29] it's just a sign of how hard rocket
[15:30] engineering is. Um, and so it's a unique
[15:34] company at a unique time in the market.
[15:36] Um, and you know, we'll see, but I think
[15:39] it'll be a really exciting show.
[15:41] >> So, Brett Winton in that clip makes this
[15:43] point. If Starship works as designed,
[15:46] and the trajectory so far suggests it
[15:48] will, launch costs fall from roughly
[15:50] $1,000 per kilogram to something under a
[15:54] hundred bucks. That's a 10x plus cost
[15:57] reduction. And the implication is that a
[15:59] single Starship full of Starling
[16:01] satellites could generate something like
[16:03] a billion dollars a year in revenue for
[16:06] 5 years per launch. If you run the math
[16:09] on that, you start to see why Kathy Wood
[16:11] thinks the standard bare case misses the
[16:13] point entirely with SpaceX. The bare
[16:15] case is simple. SpaceX has great growth
[16:18] now, but eventually competitors will
[16:20] show up. Margins will get competed away
[16:22] and the stock will decay to something
[16:24] resembling a normal aerospace company.
[16:27] That's how markets usually work. Build
[16:29] something valuable, competitors copy it,
[16:31] and their edge disappears. Kathy's
[16:33] counterargument is that the barrier here
[16:35] isn't intellectual property or a clear
[16:37] product design. It's a decade of flight
[16:40] data, supply chain development, and
[16:42] launch infrastructure that you literally
[16:44] cannot buy your way into. Blue Origin,
[16:47] for example, has spent billions and is
[16:50] still way behind. Cathy also raised the
[16:52] orbital data center angle. If you've got
[16:55] cheap launch costs and a functioning
[16:57] Starlink constellation, you've basically
[16:59] got the infrastructure for compute in
[17:02] orbit already. It's a logical extension
[17:04] of everything SpaceX is building.
[17:06] question really is timing and timing is
[17:09] exactly what a traditional 5-year DCF
[17:12] model can't handle very well. So Kathy's
[17:14] argument is all about time horizon that
[17:17] SpaceX rewards investors who can think
[17:19] further out than the market typically
[17:21] does. Bill Aman actually has a specific
[17:23] approach for exactly this and he applied
[17:26] it directly to SpaceX in the next clip.
[17:29] Before I get into that, I want to get
[17:30] into how Aman actually underwrites an
[17:32] investment like this. But first, if you
[17:35] want to actually understand why the
[17:36] orbital infrastructure layer matters,
[17:38] not just for SpaceX, but for where AI
[17:40] compute goes next, I put together a
[17:42] course called the AI landscape. It maps
[17:44] the full stack from chips and data
[17:46] centers through to frontier models and
[17:49] physical AI systems. The SpaceX orbital
[17:52] data center thesis is one of those ideas
[17:54] that really only clicks when you
[17:56] understand how the infrastructure layers
[17:58] connect to each other. The course is
[18:00] designed to give you that full picture.
[18:02] links in the description if you're
[18:04] interested. So, in this third clip, Bill
[18:07] Aman breaks down how he actually
[18:08] underwrites an investment in a company
[18:10] like SpaceX and why he thinks you have
[18:12] to approach it the same way as you'd
[18:14] approach a venture capital bet. Is there
[18:17] any way to underrate and you know I
[18:19] don't want to pick on specific companies
[18:20] but we have the three that are going
[18:22] public and then you have like a
[18:23] palunteer let's say and these things
[18:25] have become very popular in pop culture
[18:29] in maxing on subreddits on you
[18:33] know the public's consciousness high
[18:35] netw worth individuals wanting to buy
[18:37] into SPVS that are double
[18:40] loaded and then getting wiped off the
[18:42] cap tables is there any way to
[18:44] underwrite a hundred times times
[18:46] revenue, 50 times revenue, 150 times
[18:49] revenue in these companies, or are these
[18:51] just tremendously overvalued because of
[18:52] the demand side?
[18:54] >> I think you underwrite a SpaceX the way
[18:56] you underwrite a venture capital
[18:59] investment.
[19:00] >> Interesting. Explain that. Unpack it.
[19:01] >> So, everyone here invests in venture,
[19:04] right? You know, you bet on, you know,
[19:06] who's running it, right? The talent is
[19:07] enormous. Um, it's people. They taught
[19:10] me, I had a professor business school.
[19:12] He said people, opportunity, context,
[19:14] deal. So on people, SpaceX,
[19:17] >> one of one.
[19:18] >> Yeah. Opportunity, one of one context,
[19:22] you know, incredible and actually, you
[19:24] know, feel bad for Blue Origin, but not
[19:27] harmful to SpaceX. The fact that, you
[19:29] know, they're their biggest way behind.
[19:31] >> Then you get to deal. Okay, that's the
[19:34] more complicated question for SpaceX.
[19:36] Again, we don't know what the valuation
[19:37] is going to be, but if it's a billion, a
[19:39] trillion, 750 billion, then you say,
[19:42] okay, well, let's think 5 years out.
[19:44] what does this company look like? You
[19:45] know, what is Starlink? What's the
[19:47] trajectory of Starlink? You know, SpaceX
[19:50] is, you know, near monopoly in terms of
[19:52] lowcost space launch. That's going to
[19:54] become increasingly important. And even
[19:56] Amazon is going to have to become an
[19:57] even bigger customer because they're
[19:58] not, you know, Blue Origins, you know,
[20:01] and and time, I would say, has become
[20:04] increasingly valuable in the AI era,
[20:06] right? You you delay a model. We were
[20:08] talking, David and I were talking about
[20:09] the administration and and his kind of
[20:11] stepping in for the president not to
[20:13] sign that executive order kind of slow
[20:14] us down.
[20:15] >> Allegedly,
[20:16] >> you lose a month, you lose a couple of
[20:18] months today and it means a lot. So, I
[20:19] think the only question I have and I
[20:21] haven't done the math. I, you know, I
[20:22] actually invested in X. I invested in
[20:24] XAI. I'm in an SPV. Ron Baron said,
[20:28] "Bill, you got to invest in SpaceX." So,
[20:30] so I'm I'm I'm in. So, now I have So,
[20:32] obviously, I'm rooting for kind of a
[20:33] good outcome. I just doesn't I haven't
[20:35] done the Yeah, you have to.
[20:37] >> What about
[20:40] sorry uh enthropic open AI Palunteer
[20:43] also fall into this category. Do you
[20:45] underwrite those as venture investments
[20:46] as well? And have you done the the work
[20:48] on those?
[20:48] >> They're venture investments that do what
[20:50] what's helpful is they're not seed or
[20:53] series A, right? They're, you know, DE,
[20:56] but they're still like venture
[20:58] investments. These companies have proven
[20:59] they can generate a lot of revenues. And
[21:01] actually I was just saying on Sarah I
[21:03] thought she had a very very thoughtful
[21:05] uh explanation on how they think about
[21:07] committing capital right and and that's
[21:10] the thing I haven't heard from on open
[21:12] AI which is why if I were open AI I
[21:14] would be getting that message out
[21:15] because you know from the outside you're
[21:17] like it's a pretty interesting business
[21:19] model you got a company that's spending
[21:20] making capital commitments they're
[21:21] massively in excess of you know revenues
[21:24] and how do you do that and get you know
[21:25] it's it's degree of difficulty I would
[21:27] say is hard
[21:28] >> your perch um on the boards of let's
[21:30] call it these more traditional Fortune
[21:32] 500 type businesses and your
[21:34] conversations with those CEOs, how are
[21:37] they thinking about AI? Is it something
[21:39] that they're tipping into into with
[21:41] pilots? Are they doing transformation
[21:44] initiatives? Do they think this doesn't
[21:46] really apply to us? We'll deal with it
[21:48] later. What What's your sense of how
[21:50] they're adopting or bracing AI?
[21:52] >> I would say every CEO in America today
[21:54] is like, how do I use AI? How does it
[21:57] apply to my business? How is it a
[21:58] threat? uh they got to find an internal
[22:00] champion. They maybe have to recruit
[22:02] someone from the outside. I would say
[22:03] it's on the hierarchy of things they
[22:04] worry about, it's probably number one as
[22:06] both an opportunity and a threat. So if
[22:08] you're not paying attention to it, you
[22:10] you'll I mean your board is going to be
[22:12] you know asking you first question every
[22:15] meeting about what you know what how are
[22:16] we dealing with the AI threat? How are
[22:18] we dealing with the AI opportunity? So
[22:20] it's absolutely top of mind. Bill
[22:21] reveals that the hardest part of
[22:23] investing in companies like SpaceX and
[22:25] Palanteer is the deal, not the vision,
[22:28] not the team, not the market. Aman lays
[22:31] out this four factor venture play,
[22:34] people, opportunity, context, and deal.
[22:37] And his point is that SpaceX scores
[22:39] basically perfect on the first three.
[22:41] Elon Musk is running it. He's a one of
[22:43] one. The opportunity in lowcost orbital
[22:46] launch and Starlink again is a one of
[22:49] one. and the competitive context with
[22:51] Blue Origin way behind and no other
[22:54] credible rival. But the four factor, the
[22:57] deal is where it gets a bit complicated.
[22:59] At whatever valuation SpaceX comes to
[23:02] market at, does the math work over a
[23:04] 5-year horizon, and that Bill Aman
[23:07] admits he hasn't fully figured out. The
[23:10] thing I find useful about this framework
[23:11] is that it forces you to disagregate the
[23:14] question. Many retail investors who get
[23:16] excited about SpaceX or Palanteer are
[23:18] really responding to people. The people
[23:21] opportunity, the Elon Musks, the Alex
[23:23] Carp, the incredible mission and the
[23:25] incredible technology. But it can also
[23:28] make you a bit sloppy about the deal.
[23:30] You could buy the right company at the
[23:32] wrong price and still have a bad
[23:34] outcome. The framework doesn't say
[23:37] ignore valuation. It says think about it
[23:39] last and after you've confirmed
[23:40] everything else holds up. Akaman's
[23:43] comment about AI, I think, is also worth
[23:45] digging into, too. He says, "Every CEO
[23:47] in America right now has AI as the
[23:49] number one thing on their priority list,
[23:51] both as an opportunity and as a threat.
[23:55] And that cuts both ways for companies
[23:56] like Palanteer and SpaceX. On the one
[23:59] hand, the urgency is pretty real, and on
[24:02] the other hand, urgency makes people
[24:04] sloppy buyers. When boards are demanding
[24:06] AI progress and CEOs are under pressure
[24:09] to show results, that's exactly the
[24:11] environment where Karp's trap lies. The
[24:13] demand for solutions that look good in a
[24:15] presentation is highest when the
[24:17] pressure to act is highest. If this
[24:19] video was worth your time, please
[24:21] consider subscribing. Most people
[24:23] watching right now aren't. And if you've
[24:25] just sat through three reaction segments
[24:27] on Carb, Kathy Wood, and Aman, this
[24:29] channel is probably built for you. and
[24:31] subscribing is the only reliable way to
[24:33] make sure you actually see the next
[24:35] video when it comes out. Thanks for your
[24:37] support. I'll see you in the next one.