2026 Outlook: The AI → Physical World Inflection — Jordi Visser (4 enero 2026)

[0:00] for uh
[0:01] another year.
[0:05] Quite ending of the year, as always, but
[0:08] uh there was a lot of stuff that
[0:09] actually happened in the last uh couple
[0:11] weeks in the AI space uh both on the
[0:14] news front in terms of posts and X, and
[0:17] also uh
[0:19] some deals that were done. So,
[0:21] I'll go through all of that, as well as
[0:23] just kind of go through a wrap-up, go
[0:25] through some of the themes for this
[0:26] year, and I'll finish up with uh
[0:29] basically uh a recognition
[0:32] from my front in terms of where we are
[0:35] with AI. I think the progress that has
[0:38] been
[0:39] made in the last 3 months in terms of
[0:41] making it something that you can no
[0:43] longer wait for
[0:45] uh based on how much you can do with it.
[0:48] Uh and more importantly, I think for,
[0:50] you know, people my age,
[0:52] with your kids, uh I think it's a
[0:54] necessity to uh
[0:57] get them more involved, especially since
[0:59] the school system doesn't seem to
[1:00] know how to do uh deal with it yet. So,
[1:03] I'll go through all of that, and uh
[1:06] hopefully leave you guys with a lot to
[1:07] think about on multiple fronts. First of
[1:08] all, S&P finished the year so with all
[1:11] the bubble talk,
[1:13] basically, we ended up about the same as
[1:15] we did in 2020 in the COVID year, but
[1:18] you can see this was not some kind of
[1:20] freakish year. We finished up uh 16% in
[1:23] terms of price returns before dividends.
[1:25] For the NDX,
[1:27] the you know, the beta part of the
[1:30] market, it was uh
[1:32] the worst positive year we've seen uh
[1:36] over the last 7 years. So, uh definitely
[1:39] not showing uh bubble land. Uh Russell,
[1:44] minor positive, you know, 11%, nothing
[1:47] big. I expect bigger performance next
[1:49] year. The headline story was gold. Gold
[1:51] had its biggest performance since 1979,
[1:55] uh and not by a little bit. It was a a
[1:58] very, very, very good year for gold,
[2:00] especially vol adjusted.
[2:03] Bitcoin, uh
[2:05] sent the way sentiment ended, you'd
[2:06] think it was a crash, but after
[2:08] back-to-back 100-plus or 125% plus
[2:12] years,
[2:13] uh it finished down 6%.
[2:16] Everyone hates it now, sentiment is low,
[2:18] technicians hate it. I love it. Um I
[2:21] loved it last year, so take it for what
[2:23] it's worth. Uh but, I'll go through a
[2:25] lot of the the reasons why, at least
[2:27] from a a tactical basis. Forget the
[2:30] long-term story where I believe people
[2:33] are just not putting into context what's
[2:35] happening with the financial guardrails
[2:37] and the disruption from AI, but let's
[2:39] just focus on the tactical side.
[2:41] Sentiment ended up at the highest level
[2:43] of the year in terms of the AAII net, so
[2:46] it's no longer a
[2:47] situation where
[2:49] the market is negative, where we had
[2:52] been the entire year. Uh basically, if
[2:54] you look at the zero line, we spent most
[2:56] of the year below. If you take an
[2:57] average, it was one of the worst years.
[2:59] It was definitely the worst year
[3:01] adjusted for the fact that we were up.
[3:03] Uh I'm treating this as coming out,
[3:05] believe it or not, of a recession year,
[3:07] where the cyclical upswing is coming.
[3:10] Uh we'll we'll get more into that as we
[3:11] go on. I think the big news for the end
[3:14] and something that uh you know, you may
[3:16] not have seen, but GDP came out
[3:20] uh for Q3, and the
[3:24] result was 4.3%.
[3:27] Here it is out here. Only one economist
[3:30] uh basically was even close. Uh so, you
[3:34] had a very, very big move, and I think
[3:37] more importantly,
[3:38] this 4.3 comes after a 3.8.
[3:42] So, what I wanted to do in seeing that
[3:44] real GDP coming in that high, basically
[3:47] 4% over a two-quarter average, nominal
[3:51] GDP went back to the highest level since
[3:53] '23. And again, this is
[3:57] when inflation was on the higher end.
[3:59] So, we had nominal GDP go up, but
[4:01] inflation in Q3 was not uh accelerating.
[4:05] And so, basically, what I did is I went
[4:07] into Gemini and asked it to go find out
[4:10] how many times in history we've had
[4:11] back-to-back
[4:13] uh
[4:15] quarters of at least 3.8%
[4:18] since uh I guess this century. So, one
[4:21] of them, obviously, the COVID rebound,
[4:24] where we had a bunch of big numbers.
[4:25] Then, you had the mid-cycle boost in
[4:27] 2014.
[4:29] And before that, you got to go all the
[4:30] way back to the beginning of '03-'04
[4:32] coming out of the dot-com bubble, and
[4:34] then you had the pre-dot-com bubble. So,
[4:38] the most likely period, or the one that
[4:41] ChatGPT or Gemini gave us as the most
[4:43] likely period, uh it's most structurally
[4:45] similar to 203 uh 2003 to 2004. I'll go
[4:49] through the reasons why as we go through
[4:51] this, but one of the things that's
[4:53] unique
[4:54] uh about this is the fact that we just
[4:57] did this two-quarter run with no job
[5:00] creation.
[5:02] So, I want you to put that in a context.
[5:04] We grew 4% a quarter, which has really
[5:07] only happened uh three times this
[5:09] century,
[5:11] and we had zero job creation during the
[5:14] two quarters. So, if you're looking for
[5:16] something in there that
[5:18] tells you this time is different, and as
[5:20] investors and as macro people, you have
[5:22] to spend the time understanding how to
[5:25] put this into context and what it means.
[5:27] And there are not many times in history
[5:30] where you can find no job creation at
[5:32] the same time as nominal GDP coming in
[5:34] at an annualized 8% in the quarter, real
[5:37] GDP over two quarters coming in at 4%.
[5:39] So,
[5:40] uh you got that, and then you also have
[5:42] the fact that we've got for next year
[5:44] still two rate cuts
[5:47] built in. So, the Fed is still going to
[5:48] be cutting rates because of the job
[5:50] picture,
[5:52] but also because of the inflation
[5:53] picture, and gas at the pump is now down
[5:55] to 283, the lowest level since 2020. And
[5:58] basically, we're right in where we were
[6:00] from 2018 to 2020 in terms of the
[6:03] midpoint of the range. Without gas at
[6:05] the pump going higher, it's going to be
[6:07] really hard for the Fed to be raising
[6:09] rates, and anyone who's worried about
[6:10] inflation, I think should just
[6:13] sit there and say that we are probably
[6:15] going to have a reflationary boom, where
[6:17] inflation will be sticky,
[6:19] uh but if we're not going to see a
[6:21] change in the payroll numbers, then I
[6:22] don't see how they're going to get
[6:23] better with AI uh and AI agents coming
[6:26] in front of us. So, we'll talk more
[6:27] about that. Productivity, uh Dennis and
[6:30] the team at 22V basically highlighted
[6:32] well, we're starting to get more and
[6:35] more of the productivity conversation,
[6:37] and I think this is going to accelerate
[6:39] now. Marc Andreessen started to uh talk
[6:42] about it in the fact that we got 4.3,
[6:45] Elon Musk, double-digit growth is coming
[6:47] within 12 to 18 months.
[6:50] If applied intelligence is proxy for
[6:52] economic growth, which it should be,
[6:54] triple-digit is possible in 5 years.
[6:57] I just want you to take into context
[6:59] this whole double-digit growth and
[7:00] what's going on, even though I don't
[7:03] think that is coming, I do think the
[7:06] period you're looking at here with the
[7:08] ability of replacing labor or not hiring
[7:11] people as top-line
[7:14] revenue. In this case, nominal GDP grew
[7:17] 8% in the quarter.
[7:20] And yet, we have no hiring going on.
[7:24] So, think about that in terms of the
[7:25] productivity boom. If this compounds,
[7:28] then you can theoretically get to
[7:30] numbers that
[7:32] are insane. So, just keep it in the back
[7:34] of your mind, because it's not a joke.
[7:36] Uh copper moving higher. So, for anyone
[7:39] questioning whether where we are in a
[7:41] reflationary boom,
[7:43] copper
[7:44] accelerating into the end of the year.
[7:46] MSCI World ex the US, which is my proxy
[7:50] for global liquidity and kind of a
[7:53] global boom,
[7:54] closing on the highs of the year.
[7:57] The dollar, after, you know, a bounce
[7:59] from what were, you know, a very weak
[8:01] beginning part of the year, is in
[8:03] downtrend again,
[8:05] posting its worst year since 2017
[8:08] with Fed cuts coming. The yield curve
[8:10] steepening, all of these on suggesting
[8:13] the same thing. The Kospi,
[8:15] closing. Doctor Kospi, Doctor Copper,
[8:17] you've got everything kind of moving
[8:19] higher. In Korea, export growth posted
[8:23] back-to-back big quarter or big months
[8:26] to finish out the uh uh
[8:28] or sorry, quarters to finish out the
[8:30] year.
[8:32] And to show you the importance of South
[8:34] Korea from an export, this is
[8:35] year-over-year exports, the white line.
[8:37] This is the US PMI.
[8:40] We're going higher in terms of all of
[8:42] these fronts. You also had Taiwan
[8:45] exports, which again, another proxy for
[8:47] PMIs, blowing out to the upside.
[8:51] Capital goods, we got the durable latest
[8:53] durable goods report. Capital goods new
[8:55] orders, non-defense ex defen- ex air.
[9:01] At levels historically equivalent with
[9:04] close to 60 PMIs, we just continue to
[9:06] march higher. The diffusion index, this
[9:09] is from MacroMicro. I thought this was a
[9:11] good chart, just highlighting the global
[9:14] PMI diffusion. So again, diffusion's a
[9:17] diffusion on top of a diffusion. So,
[9:19] think of this again as a second
[9:20] derivative.
[9:21] Again, implying higher levels. This is
[9:24] using it verse
[9:25] global monetary policy. I've shown
[9:28] charts like this before. We are doing
[9:29] rate cuts. And so, not a surprise if
[9:32] you've watched my videos, uh next year
[9:34] to me, the number one theme
[9:37] is connecting AI into the physical
[9:39] world. So, my outlook paper for this
[9:42] year, long one with a lot of the themes
[9:45] that I think are going to be important
[9:46] to support PMIs not only going higher
[9:49] this year, but the beginning of a mega a
[9:51] mega trend
[9:53] on the back of commodity strength, on
[9:55] the back of enterprise using edge
[9:56] devices, edge device upgrades for cars,
[9:59] auto or cars, phones, computers, uh the
[10:03] MPUs that are going to be necessary for
[10:06] the embodied AI, humanoids coming behind
[10:09] it, the data center build out, the bring
[10:11] your own generation BYOG power trend
[10:14] that is in place. All of this stuff fits
[10:17] into the physical world. I'll be putting
[10:18] this out 22V. If you guys haven't
[10:21] reached out to 22V, get signed up.
[10:23] There's a lot of things I'll show at the
[10:24] end in terms of what I'm going to be
[10:25] doing on top of just the written work
[10:28] now, but giving you specific names in
[10:30] the space.
[10:32] Uh
[10:33] the thematic ideas within there, some of
[10:35] the ones I mentioned, but you get the
[10:36] picture. So, for all of these,
[10:39] including the BYOG side, uh which is
[10:42] becoming a bigger and bigger theme,
[10:44] solving most of the power problems, uh I
[10:47] think it's an important thing to get up
[10:49] to speed on. Colossus 2 is going to be
[10:51] uh up and running completely. There's
[10:53] more work going on on building out
[10:55] Colossus to get it to its eventual
[10:57] ending of a million GPU cluster.
[10:59] All of this stuff is going to be
[11:01] important. Uh one podcast to reference
[11:04] for this week, I did listen to Facts
[11:06] versus Feelings. I do from time to time.
[11:08] The reason I cared about this one were
[11:09] twofold. One is just wasn't a lot of
[11:12] podcasts out the last couple weeks of
[11:13] the year, but uh
[11:15] >> [snorts]
[11:15] >> Ryan Detrick did have on two
[11:16] technicians, uh and I thought they did a
[11:19] good job. I like technicians at the end
[11:21] of the year cuz I like to look for
[11:23] charts that are pointing a direction and
[11:25] then try to figure out what it means
[11:27] together. That has gotten a lot easier
[11:30] with artificial intelligence. Uh my
[11:32] number one thinking partner on this
[11:34] stuff, which again, I'll show you at the
[11:35] end. Uh this is going to be a big year
[11:37] for teaching whoever wants to learn how
[11:39] to be able to do this stuff I do, but
[11:41] also, I don't think you have time to
[11:42] wait anymore because I think we've
[11:44] reached a point where the capabilities
[11:46] of artificial intelligence have
[11:48] compounded at such a fast pace now that
[11:50] when I show you what I've been able to
[11:52] build in a matter of of hours uh on
[11:55] things that I had team of data
[11:57] scientists that were never able to
[11:59] finish for me, not because they weren't
[12:01] capable, but because of how difficult it
[12:02] was for me to give what my brain wanted
[12:05] and get them to be able to do it. That
[12:08] merging together of thought
[12:10] collaboration is what has happened with
[12:13] AI, which allows you to do this stuff
[12:14] and kind of get your answer, go back,
[12:16] get your answer, go back. So, with the
[12:18] podcast, what I did is I just said,
[12:20] "Take all of the charts that these two
[12:21] technicians talked about and let's break
[12:24] them down in terms of what went on." So,
[12:26] they talked about financials.
[12:28] Talked about long decade sideways
[12:30] assets. I've shown some stuff from John
[12:32] Roque. I'll go through a bunch of charts
[12:34] same way. These long-term bases we are
[12:36] breaking out of. This is the PMI side.
[12:39] Uh a lot of these long-term bases are
[12:40] even longer. They're commodity stuff
[12:42] going back. That's the hardware versus
[12:44] software. You have hardware versus
[12:46] software, all of those multi-decade
[12:49] decade sideways things, which really
[12:51] haven't worked since 2003 to 2004. Uh I
[12:55] think you're in the beginning stages of
[12:57] it. They talk about gold, what it means,
[12:59] especially when it's happening with
[13:00] stocks going higher. And they highlight
[13:03] another fact, which is the market
[13:04] breadth. So, when you take all of those
[13:06] together
[13:07] and you go look at gold having rising at
[13:09] the same time as cyclical strength, with
[13:11] breadth improving, with junk bonds
[13:13] strong,
[13:14] equities making new highs, these are the
[13:17] only time periods that you found where
[13:19] you could have gold going up alongside
[13:21] of risk-on assets, but in particular
[13:24] with inflationary type or reflationary
[13:27] assets, reflationary, not recessionary,
[13:29] global growth boom. So, these are the
[13:31] periods that's bringing up. Same thing
[13:33] he shows for the breadth. And again,
[13:37] usually
[13:38] another 12 to 24 months using history,
[13:41] even though they only have a couple of
[13:42] them. So, what regime are we most likely
[13:45] in?
[13:46] Well, financials breaking out, early
[13:48] cycle reflation theme, multi-decade
[13:50] sideways poised to break out, commodity
[13:51] super cycle beginnings.
[13:53] Global nominal GDP accelerates faster
[13:55] than expected. Real assets dramatically
[13:57] outperform financial lot. This is never
[13:59] a rechet recessionary regime. It is a
[14:02] global reflation regime.
[14:05] So, then you go through, what does it
[14:07] normally mean? And again, I just showed
[14:09] you, copper's going higher, nominal GDPs
[14:11] accelerating, equities are going higher,
[14:14] gold's going higher, financial
[14:16] conditions are easing. We've got the Fed
[14:17] cutting rates, the dollar is weakening.
[14:21] This is all part of what normally goes
[14:23] on during it, what we would expect in
[14:25] this. Inflation stabilizes, but remains
[14:28] above the pre-AI error.
[14:30] This is not late cycle stagflation. And
[14:33] again, I've talked about this. The
[14:34] amount of times I've heard stagflation.
[14:36] All of these things have kept people out
[14:37] of the market. Bubbles kept people out
[14:39] of the market. A full-scale reflationary
[14:41] expansion where tech and commodities
[14:43] rise together. This has only happened
[14:44] twice in 60 years, and both were
[14:46] extremely profitable multi-year cycles.
[14:49] Climb the wall of worry.
[14:52] So, Bitcoin had a horrible relative year
[14:54] and had a bad year overall. How does
[14:56] Bitcoin or how should it perform in this
[14:58] regime? AI-driven global reflation,
[15:00] commodity breakouts, financial
[15:02] strengthening, breadth expansion, and a
[15:03] weaker dollar, Bitcoin should be one of
[15:05] the strongest performing major assets in
[15:06] the world. I will say it will be the
[15:08] strongest. Thought the same thing last
[15:10] year. So, again,
[15:12] this is not advice. I'm putting my money
[15:14] where my mouth is, but I pretty much
[15:16] always do with Bitcoin because of my
[15:18] long-term view. The regime aligns almost
[15:20] perfectly with Bitcoin's historical
[15:21] environments.
[15:23] It loves a weaker dollar, steeper yield
[15:25] curve, it thrives when real assets
[15:26] outperform.
[15:28] Here is it versus PMIs. I've talked
[15:31] about this. When PMIs are rising, you
[15:33] get the best time of Bitcoin. Doesn't
[15:35] matter if it's a minor cycle or a bigger
[15:37] cycle. Anything going on, it is
[15:39] typically very sensitive to PMIs and
[15:42] altcoins,
[15:44] or let's just use higher beta
[15:46] uh crypto, and let's just take Ethereum,
[15:49] as I'll get into in a little bit, uh as
[15:51] another part. So, this is the perfect
[15:53] conditions for S-curve adoption. This is
[15:55] something critical. I've talked about
[15:57] this as my network effects. AI adoption
[15:59] and decentralization equal the Bitcoin
[16:01] narrative strengthening. Tokenization
[16:03] and real-world assets, a big theme for
[16:04] next year. More on-chain liquidity, more
[16:07] money inside
[16:09] the crypto world, the digital economy,
[16:11] good for Bitcoin. Institutional adoption
[16:12] is now structurally embedded, ETFs and
[16:14] custody rails. All of this stuff is the
[16:16] technology side of it. Should be one of
[16:18] the best performing assets.
[16:22] And again, reminder, this is after it
[16:25] had a quiet year. But again, if you take
[16:26] the three-year performance,
[16:29] nothing has beaten it.
[16:30] So, just use it for what it is. I still
[16:34] believe this is what went on last year.
[16:35] We continued into the end of the year to
[16:37] see OGs selling. This to me was just a
[16:41] representation again
[16:43] of what makes sense. Ideologically, the
[16:45] people that were early and went into it,
[16:47] you've got the government more involved
[16:48] in sponsoring it. You've got ETFs. This
[16:51] is not the true decentralized
[16:53] ideological breakaway from the fiat
[16:55] system. So, you've got those people
[16:57] bailing out at the same time,
[17:00] there's a new sheriff in town in terms
[17:02] of where you can find a five-bagger or a
[17:04] 10-bagger, and Bitcoin's not in the same
[17:06] class, clearly, as what's been going on
[17:09] in AI. Micron Technology, a very big
[17:12] asset,
[17:13] went from a low of 60-some-odd to over
[17:15] 300. You're getting moves like Palantir
[17:18] and things. So, it lost a little bit of
[17:20] its relative mojo as a beta asset, and I
[17:23] think that took a lot of people out,
[17:25] especially uh people that on the VC
[17:27] world went. We had some market maker
[17:28] issues towards the end of the year. I
[17:30] think all of this was a consolidation,
[17:33] and I love the fact that sentiment is
[17:35] dead, and I'm love the fact that I can't
[17:37] find one technician that likes it at
[17:39] this point. Now, I posted this
[17:42] on December 30th that we had closed
[17:44] below the 50-day moving average for
[17:48] 64 days. Uh it finished the year with
[17:51] another three on top of it. We did close
[17:53] above it on the first day of the year.
[17:56] So, it had been 64 straight days, or 65
[18:01] at that point,
[18:03] since 2020 that's only happened four
[18:04] other times. So, we tied this one from
[18:07] early this year. That was obviously a
[18:09] major bottom before it went higher. Not
[18:11] all of these ended up with a major
[18:14] bottom that ended up going higher, but I
[18:16] will show you the four points. One, two,
[18:20] three,
[18:21] four. All of them were at least
[18:22] short-term entry points. This one got
[18:24] you a good rally. This one got you a
[18:26] good rally. But none of these were kind
[18:27] of the beginning of this. This one here,
[18:30] you had to deal with some more basing.
[18:32] But my main point
[18:34] in terms of going through this was to
[18:36] highlight that again,
[18:39] for me, MSCI World ex-US, which takes
[18:42] into account the non-tech world in terms
[18:45] of the ex-US, dominated by commodities
[18:48] and things like that. At the same time,
[18:51] when you have this going on, this has
[18:52] got a weaker dollar trend involved with
[18:54] it as well. MSCI World ex-US did
[18:56] outperform last year. To show the
[18:58] correlation between it and Bitcoin, here
[19:00] is the 100-day moving average of both of
[19:02] them.
[19:03] And over time, you can see Bitcoin has
[19:04] diverged since the market making
[19:07] incident and since we went down. So, my
[19:09] belief and the reason that I will not
[19:11] listen to technicians on this and I will
[19:13] still be buying into any dip that occurs
[19:15] if there are any at the beginning of the
[19:17] year is that liquidity is still
[19:19] positive. My outlook in general for this
[19:21] year is positive. There will be time in
[19:24] the future where I'm not as positive on
[19:25] the US side, and I think Bitcoin can
[19:27] have a correction. But right now, it is
[19:30] swimming uh against the liquidity
[19:32] stream, and I think that means the
[19:33] snapback will be violent, just like the
[19:36] snapback here, which also occurred
[19:38] during the same thing. Liquidity was
[19:39] going higher. I believe the same thing
[19:41] is going to happen this year. So, we
[19:44] closed above the 50-day moving average.
[19:47] First time, we'll see what happens. Uh
[19:52] It wasn't the only one. Solana
[19:55] Only time Solana had been below the
[19:56] 50-day moving average again came in
[19:58] during the April period. We closed above
[20:00] it again on the first day of the year.
[20:02] We'll see what happens. What really has
[20:04] me positive from a beta basis and what I
[20:07] like happening related to the PMIs is
[20:09] what's been going on with Ethereum
[20:11] relative to Bitcoin. So we look at a lot
[20:12] of relative positions in stock market,
[20:16] but very seldom will you see people look
[20:17] at Ethereum relative to Bitcoin. Last
[20:20] year during the upswing or when Bitcoin
[20:23] was going higher really beginning from
[20:24] 2022, Ethereum was underperforming. The
[20:28] fact that Ethereum is now going higher
[20:29] at the same time that we're getting the
[20:31] rise of
[20:32] acceptance for both stable coins and
[20:35] tokenization to me means the ecosystem
[20:37] network effects are kicking in. I think
[20:40] you want to be positive on Bitcoin
[20:42] because Ethereum is outperforming. The
[20:44] network effects are in place.
[20:46] Uh John Roark, who I show a lot and I'll
[20:49] be doing a lot more work with. I'm just
[20:51] showing this cuz he's got a very
[20:52] different view and he's saying there's
[20:54] risk to 60,000. He's also still calling
[20:58] for lower levels in
[21:00] MicroStrategy. I'm going to go against
[21:02] John on this one, but on the rest of the
[21:05] ones I'm about to show you we're on the
[21:06] same page. This is some of his big bases
[21:09] that again fit back in with the facts
[21:11] and feelings situation. Citigroup, a
[21:13] bank. As John likes to say and as I've
[21:16] shown on here, you do not get bearish
[21:18] the market when banks are breaking out.
[21:20] But in particular, a 16-year base in
[21:22] Citigroup that takes us back to the
[21:24] great financial crisis. Banks are going
[21:26] higher, deregulation is kicking in.
[21:28] Freeport-McMoRan
[21:30] haven't taken out the highs yet since
[21:33] again the great financial crisis. A big
[21:35] base, but you also have a PMI base going
[21:37] here.
[21:38] Major catalyst. I think copper is going
[21:40] much much much higher on the back of the
[21:42] needs that are associated with
[21:45] AI and the buildout. ExxonMobil, again
[21:49] up at the top end of the range. PMI base
[21:52] here, long-term base. And again, this
[21:55] happened with oil trading down to 55
[21:58] this year.
[21:59] DuPont, there's going to be a major
[22:01] major bull run in chemicals in my
[22:03] opinion because of a lot of things that
[22:04] I'll cover in terms of the Nvidia Grok
[22:07] deal and just a lot of NPU packaging. We
[22:10] are moving away from GPUs and we're
[22:12] moving into advanced packaging.
[22:15] Building out the next phase, which is
[22:16] edge devices. Chemicals will be a big
[22:18] part of this as well. You've got Hecla
[22:20] Mining, so the gold silver side breaking
[22:23] out long base. Newmont Mining, you got
[22:25] to go all the way back to the 80s in
[22:27] terms of breaking out.
[22:29] Delta Airlines, I've highlighted
[22:31] transports, another big PMI theme. All
[22:33] of these are PMI themes.
[22:35] Eli Lilly, I've highlighted the
[22:37] advancement in terms of the shift. So
[22:39] Lilly fits into a different category for
[22:41] me. I think healthcare is going to be a
[22:43] surprising replacement for a lot of tech
[22:45] this year in terms so biotech and pharma
[22:49] as AI drug discovery starts to reprice
[22:51] and re-rate this group with higher
[22:54] multiples. Tesla, a big part of the
[22:56] robotics and PMI theme for me. John
[22:58] likes that one as well. He and I've been
[23:00] on the same page on this back since it
[23:03] was about 330.
[23:05] Both talking about it and as I
[23:07] mentioned, Tesla, no safety monitor in
[23:09] the car.
[23:10] Went on
[23:11] uh uh December 24th with Elon Musk in it
[23:14] they're rolling it out for public rides.
[23:18] And again, AI still. I showed this last
[23:21] week or 2 weeks ago, but in case people
[23:23] didn't see it.
[23:25] This is
[23:26] the biggest risk to the market stability
[23:29] in 2026.
[23:31] We've never seen a single risk score so
[23:33] far ahead of the rest entering a new
[23:35] year, says Deutsche Bank. And again
[23:37] AI bubble, you will be climbing the wall
[23:39] of worry
[23:41] the entire time. Uh I on the market
[23:44] Michael Cembalest did a you know a great
[23:47] report. I like his work. I've
[23:49] highlighted before, but I think he
[23:51] brought up a lot of charts and tables
[23:53] that are worth looking at for you guys
[23:55] in terms of the AI bubble.
[23:58] And again, what we've got here is just
[24:00] total stocks
[24:02] when you include all of the direct AI
[24:04] stocks, the AI utilities, the AI cap
[24:06] equipment, you're dealing with 42
[24:07] stocks. Price return
[24:09] 78%
[24:11] S&P X those names, 22%. So basically the
[24:15] bulk of everything, whether it's
[24:17] earnings growth, whether it's the CapEx,
[24:19] all of this has been related to
[24:22] AI.
[24:24] Not a shock, but I just want to make it.
[24:26] In terms of spending, how big this is.
[24:29] You got everyone kind of spending
[24:32] a wasteful amount of time trying to say
[24:34] how much of this economy is driven by
[24:36] AI. I keep saying it's almost all of it.
[24:39] We have no job creation. If anything,
[24:42] you can see there's no job creation. AI
[24:45] is driving all of these factors at this
[24:47] point. The Fed would not be cutting
[24:48] rates if it wasn't for the job
[24:50] situation. The job situation would be AI
[24:53] is a deflationary force. It is a job
[24:56] replacing force. It has societal
[24:58] impacts. That will be a major risk is
[25:00] going to be the backlash that comes
[25:02] starting with the midterm elections.
[25:05] All of the things of trying to restrict
[25:07] this buildout, you'll have probably a
[25:09] lot of issues that pop up and risk that
[25:12] scare the market this year.
[25:14] But the reality is the tech the tech
[25:16] CapEx is massive right now. If you
[25:19] combine all of these, and again, these
[25:22] are in percent of GDP, so this is
[25:24] normalized. Manhattan Project
[25:26] electricity, Apollo Project, this
[25:28] you're dealing with this. You leave this
[25:30] one out, you're you're dealing about the
[25:31] same size as all of them combined.
[25:33] That's how big the CapEx buildout is.
[25:36] From valuation basis the reason I like
[25:39] showing this stuff is because
[25:41] I wouldn't say he's positive on the on
[25:43] AI, but I would say that he's pushing
[25:45] back against the valuation side. So if
[25:48] you've got a negative view on valuation,
[25:50] you should go spend some time looking
[25:52] the report. If you're worried about the
[25:54] debt situation, he highlights that the
[25:56] debt is big relative
[25:58] to what it was, but then he honestly
[26:00] goes through and highlights
[26:03] the net debt to EBITDA of so zero
[26:06] indicates an excess of cash and
[26:08] marketable securities. The majority of
[26:10] the space has is just loaded with cash.
[26:13] Free cash flow to revenue ratios of of
[26:16] of the AI companies. He goes to and
[26:18] compares everything to the rest of the
[26:20] market in terms of the Russell 3000.
[26:23] So his argument is this these none of
[26:24] these are are are issues for now. If
[26:27] they grow rapidly and we don't see the
[26:28] revenues, but that's the same argument
[26:30] that Jim Chanos make, which I agree
[26:31] with. If we don't see the revenues in a
[26:34] year
[26:35] you're going to have trouble for these
[26:36] names. They're still spending the money
[26:38] and as long as they're spending the
[26:40] money, you're going to see it flow
[26:41] through the S&P 500, you're going to see
[26:42] it flow through GDP. The question is are
[26:44] you going to get the benefits? I'm
[26:46] telling you as a fact, the improvements
[26:49] in AI over the course of the last 3
[26:52] months
[26:53] are impossible to describe. It is moving
[26:56] so rapidly that if you don't use it
[26:59] every day all day
[27:01] the entire day for everything that you
[27:03] do. In your cars, everything. Go buy a
[27:05] new phone, go buy a new car, talk to it
[27:07] all day long in both situations. I took
[27:10] a drive for 3 hours going back out to
[27:13] New Jersey over the course of the
[27:14] holidays. The entire time I was on with
[27:17] Grok. I didn't listen to a podcast. I
[27:19] literally had a conversation. It was
[27:21] beyond amazing for me to sit in the car
[27:24] and do it. Michael Cembalest, his risks
[27:27] for 2026. Power generation constraints.
[27:30] I'm less worried about this, but I do
[27:32] think this can become an issue even if
[27:34] it's just the states pushing back. China
[27:36] scaling the technology mode on its own,
[27:38] this one I think is overblown. China's
[27:40] approach to Taiwan geopolitical risk,
[27:42] this one has the biggest tail if there's
[27:44] any
[27:46] military fears on them invading Taiwan,
[27:48] obviously that is the biggest one. This
[27:51] one to me is the most likely, the
[27:52] metaverse moment for hyperscaler
[27:54] profits. Not that they're not going to
[27:56] get the money, but if it doesn't
[27:58] materialize as expected, it could become
[28:00] an issue next year, particularly with
[28:02] all the issuance that is going to be
[28:04] necessary to get more money. So I took
[28:06] all of his risks, I put them into
[28:11] ChatGPT
[28:13] Pro
[28:15] and I said, all right, go through these
[28:16] and give me your analysis. Give me the
[28:17] outcome probability, worry probability.
[28:20] So again, outcome probability, what's
[28:22] probability it actually happens? Worry
[28:24] probability, what's the probability that
[28:26] you the investors are worried about it?
[28:27] Look how high those numbers are.
[28:29] The nar- dominant narrative probability,
[28:31] market impact and then the So I just
[28:34] went through
[28:35] highest volatility spike not surprising
[28:37] is Taiwan. Most repeatable pain risk is
[28:40] the ROI CapEx, which I agree. You can go
[28:43] look at it on your on your own.
[28:47] Uh okay. So for the final uh
[28:52] slides
[28:53] if you haven't read this, bring it up in
[28:56] ChatGPT. I'm showing you the date. You
[28:59] can go find it in X, but
[29:01] Andrej Karpathy, who's who's arguably
[29:03] the most respected person out there and
[29:06] who's been very very, let's say
[29:08] non-emotional on this whole situation
[29:12] from
[29:13] mainly time at Tesla, but also at OpenAI
[29:16] uh just a phenomenal well-respected
[29:19] person within the space. You can tell,
[29:20] 15.6 million views. He did an interview
[29:22] with Dwarkesh back in October. I
[29:25] received multiple hedge fund reach outs
[29:29] saying that this was a negative for AI.
[29:32] What was my opinion?
[29:33] Where he basically said AGI wasn't
[29:35] coming until for another decade and that
[29:37] AI agents were way behind. Well, this is
[29:40] what he put in. I've never felt this
[29:41] much behind as a programmer. So I want
[29:43] you to remember I've never felt this
[29:46] much behind as a programmer. I have a
[29:47] sense that I could be 10 times more
[29:49] powerful if I
[29:50] just properly string together what has
[29:52] become available over the last year. So,
[29:55] Andre Karpathy is number one doing
[29:57] podcast talking about AI in a negative
[29:59] fashion, but then when he finally sits
[30:02] down and takes the time to use it a lot
[30:06] I've never felt this much behind as a
[30:08] programmer. Clearly some powerful alien
[30:10] tool was handed around
[30:12] except it comes with no manual and
[30:14] everyone has to figure out how to hold
[30:15] it and operate it while the resulting
[30:17] magnitude nine earthquake is rocking the
[30:19] profession. Roll up your sleeves to not
[30:21] fall behind.
[30:22] I'm going to emphasize this point to all
[30:24] of you because this is not just for
[30:26] programmers.
[30:28] This is for you if you're still going to
[30:30] be in the workforce
[30:32] and especially your children who are
[30:34] about to enter the workforce. College is
[30:36] not teaching them this stuff. They need
[30:39] to be using it every single day.
[30:42] This post by Karpathy went massively
[30:44] viral in late December with over 15
[30:47] million views. It captures a pivotal
[30:48] moment in the software engineering AI
[30:50] community serving as a candid wake-up
[30:53] call about how rapidly AI tools are
[30:56] transforming programming. You have
[30:57] personal vulnerability. You're giving up
[31:00] on 10 times productivity boost.
[31:05] The Times of India Business Insider
[31:06] Financial Express framed it as an open
[31:08] letter to software engineers
[31:09] highlighting the urgency of upskilling
[31:11] and AI's rise. In essence, it's a
[31:13] landmark acknowledging that 2025 marked
[31:15] an inflection point where AI began
[31:17] fundamentally changing what it means to
[31:18] be a programmer
[31:20] forcing a rapid evolution in skills. The
[31:22] first-order effects
[31:24] job refactoring, productivity explosion.
[31:26] We get back to the GDP side. We're going
[31:28] to get into the AI agents. Tools like
[31:30] multi-agent systems
[31:33] proliferate.
[31:34] Enterprise adoption surges. This is
[31:37] critical. When you get to this point,
[31:39] now the enterprise adoption is going to
[31:41] surge which means the revenues for the
[31:44] hyperscalers are going to surge. The
[31:46] model producers are going to see it
[31:48] surge because of agents and enterprise
[31:51] adoption will go off.
[31:53] So, you're going to see a big movement
[31:55] in the market in this. The third-order
[31:57] effects, and again, if you guys aren't
[31:58] in the second-order effects, third-order
[32:00] effects, I have a Santa Fe model um
[32:02] system thinkers prompt that I use to
[32:05] basically take any news item or any
[32:07] podcast or anything and get it taken out
[32:09] to the second and third-order effects.
[32:11] This is a critical part that you should
[32:12] be using on anything you have in your
[32:14] life regardless of what it is. I'm going
[32:16] to go through some of the healthcare
[32:17] stuff, some of the investing, but
[32:18] anything where you learned, if I do X,
[32:21] what happens to Y and Z on the other
[32:23] side, you have to start asking questions
[32:26] like that. This is what it's good at uh
[32:28] is the polymath side, the ability of
[32:30] looking at things from a system uh
[32:32] systems thinker perspective.
[32:35] Traditional software motes erosion. This
[32:38] gets back into Salesforce. It gets back
[32:40] into everything in there. The ability of
[32:42] building apps on the fly.
[32:45] Motes move to data pipelines,
[32:47] proprietary data sets like Tesla's
[32:49] driving data or agent ecosystems.
[32:53] We're in the mid-transition.
[32:55] It was hyped as a year of agents, but he
[32:57] noted that they fully don't work yet.
[33:00] So, now we're getting into the point
[33:01] where the path to recursive
[33:03] self-improvement
[33:05] he's still cautious
[33:07] but he didn't mention that in this
[33:09] particular one, but potential for rapid
[33:11] pro progress 2027.
[33:17] Jensen Huang,
[33:19] >> [sighs]
[33:19] >> the most important thing is
[33:21] you're not likely to lose a job to AI
[33:24] and I agree with that. You're going to
[33:26] lose a job to somebody who uses AI. So,
[33:29] I would recommend everybody do what
[33:31] do what we do, which is use artificial
[33:33] intelligence with our work and do our
[33:36] jobs better.
[33:38] So, that brings me into this. So, this
[33:39] year I will be launching this. It'll be
[33:41] coming out in January. We're just
[33:43] getting rid of all of the bugs. You can
[33:45] see my name is spelled wrong here.
[33:46] There's lots of little bugs that need to
[33:48] be
[33:49] need to be finalized, but the goal and
[33:51] part of it's going to be not just this
[33:52] the research you can get, the stocks,
[33:54] the sectors, the themes, but a big part
[33:58] of it is going to be the AI side and
[34:00] just showing you how I do all the things
[34:03] that I do and how you can incorporate AI
[34:06] into everything that you want to do,
[34:07] whether it's cooking, whether it's
[34:08] fantasy football, whether it's health,
[34:10] whether it's whatever. Anything that you
[34:12] find fun that you do on a regular basis
[34:15] where you've historically used Google or
[34:17] books or whatever uh I'm going to
[34:20] basically go in and show you how to do
[34:21] this prompt engineering, everything
[34:23] along those lines. During the two weeks
[34:25] Nvidia, and I just want to bring this up
[34:27] because this is kind of the way that
[34:30] everything goes for me now and what I've
[34:31] set up. So, Gavin Baker goes through the
[34:34] Nvidia versus Grok thing. I go out and
[34:36] write a paper on it in terms of the
[34:38] importance of it, the Nvidia Grok deal,
[34:39] why AI's next frontier requires
[34:41] architectural revolution over Moore's
[34:43] Law. This was a critical critical
[34:46] relationship built between Nvidia and
[34:48] Grok. Um it led to, and this is what
[34:52] I'll be doing for the people who sign up
[34:54] and the institutional clients of 22V.
[34:57] Basically I have about 20 names that are
[35:00] associated with this change. This is a
[35:03] change. Nvidia's not on the list. So,
[35:04] this is a broadening out related to the
[35:07] PMIs of moving into a world where
[35:10] advanced packaging becomes the dominant
[35:12] theme. Now, this has huge implications
[35:15] for a variety. So, the way that I'm
[35:16] going through this and I'm just giving
[35:17] you, how do I go from a thematic idea to
[35:20] an equity framework? Give you the
[35:22] companies. John Roque put together the
[35:25] score. If you don't know John's work,
[35:27] his his work four is
[35:30] is what you want to be involved in in
[35:31] his work. It means the trend is your
[35:33] friend.
[35:34] So, these all have a four technical
[35:36] score. I go through the earnings
[35:37] commentary and what I'm looking for
[35:39] there is I've got an algorithm which
[35:41] gives me the rankings in terms of is
[35:43] this company's progression in their
[35:45] earnings commentary showing the strength
[35:47] where management is getting more
[35:48] optimistic as time is going on related
[35:50] to the AI demand theme. So, you can see
[35:53] these are in there. Lattice
[35:54] Semiconductor continues to grow on this
[35:56] and Qualcomm does. And then revisions.
[35:58] So, revisions are just are the analysts
[36:01] on the street. So, here you have
[36:02] management. Here you have the sell-side
[36:04] research. Here you have technicals and
[36:06] here you have thematic ideas. It's a
[36:08] broad picture. My goal is to be able to
[36:11] provide you guys this uh for asset
[36:13] managers, for separately managed
[36:15] accounts. Anything along those lines, I
[36:16] will be putting all of this stuff
[36:18] together. There's a lot of themes,
[36:19] there's a lot of names, there's a lot of
[36:21] ways to screen. I'm going to be doing
[36:23] this for a variety of people um based on
[36:25] the ask, but also based on the fact that
[36:27] I've set all these up. So, Nvidia just
[36:29] admitted the general-purpose GPU era is
[36:31] ending. I can't think of a more critical
[36:33] period than kind of thinking about this
[36:35] and what it means. It means there's
[36:36] different names that are going to work
[36:37] this year. Now, at the same time why did
[36:40] Meta at the end of the year buy
[36:41] Manifold? What is the signal for your
[36:43] enterprise AI agent strategy?
[36:47] The Wall Street Journal marks said marks
[36:49] one of the clearest signals yet that
[36:50] large tech platforms are no longer just
[36:52] competing on model quality, but on who
[36:54] controls the execution layer of
[36:56] AI-powered work.
[36:57] We're getting into the
[37:00] return on invested
[37:02] capital.
[37:03] AI agents is the critical part. Again, I
[37:07] talk about this in here in terms of
[37:09] this.
[37:10] The with the names that are going to
[37:11] benefit from the enterprises needing to
[37:15] upgrade their systems. This is a major
[37:17] major change. We are going from the
[37:19] cloud.
[37:20] We're going to need to have these
[37:23] on-premise, on-site. Everything is
[37:25] on-site. Bring your own generation. It
[37:27] is very physical, it is very
[37:29] hardware-driven. You're going to have a
[37:30] lot of names that are going to benefit
[37:32] from this.
[37:33] I'm now created another thing which will
[37:35] be on the paywall for my X
[37:38] trending stuff. So, each weekend I will
[37:41] go in and basically look for the
[37:42] analysis of the most trending things out
[37:44] there. The reason I wanted to show this
[37:46] one, so this is from the final week, the
[37:48] agentic and autonomous agents. So, AI
[37:50] agents is the theme you want to be
[37:52] investing in. If you don't have 20, 30,
[37:54] 40 names related to the agentic side
[37:58] you got to start focus on it. This is
[38:00] not about data centers anymore. This is
[38:02] not That's what That's what was up there
[38:03] last year. It's not about the Palantir
[38:05] stuff in terms of the data coming
[38:06] through. You're getting a change in
[38:08] what's there and the charts are showing
[38:10] that the trend is changing. And like I
[38:12] said, reflation is the theme and
[38:14] reflation you want to be less software,
[38:17] you want to be more hardware. I created
[38:19] a market immune system. This is
[38:20] something that took me, as I mentioned
[38:22] at the beginning uh less than
[38:25] less than two hours to build. I sat on
[38:27] my couch early in the morning and I
[38:29] basically said, you know what, I'm
[38:31] finally going to build my turbulence
[38:32] model that I had asked my data scientist
[38:34] to build for years. They were more than
[38:36] capable of doing it, but here's what was
[38:37] not easy. The way that I did this, which
[38:40] you couldn't do back then, is I sat
[38:42] there and I said, I want you to go to
[38:43] GitHub and I want you to go find me the
[38:46] most highly rated
[38:49] code out there for a turbulence model.
[38:52] It comes back. It gives me this one.
[38:56] I go through and now I have to choose
[38:58] the assets. And the way that I wanted to
[39:00] choose the assets, and this is always
[39:02] what I believe when you're dealing with
[39:03] something like a covariance matrix as a
[39:05] warning system for the market. And
[39:07] that's what I wanted. A market immune
[39:08] system for me was basically something
[39:10] that would tell me when correlations and
[39:12] volatility were changing especially when
[39:15] the S&P was not yet changing cuz I
[39:17] believe the market will always show you
[39:19] the risk is moving before the actual
[39:22] index was. I I used models like this
[39:24] back in in Brazil. They were not as
[39:26] sophisticated as this. So, I gave it 100
[39:30] assets, 99 assets. A good percentage of
[39:33] them were related to AI because that's
[39:35] what's been driving the market, but then
[39:36] it also included things across stocks,
[39:38] bonds, commodities, currencies, crypto.
[39:40] When do they all start shaking at the
[39:42] same time? And what I end up getting is
[39:45] a model that basically
[39:47] is measured this way. The red line is
[39:49] where the turbulence is. When the red
[39:51] line is normally up high, these are ends
[39:53] up being bottoms. So, this is after
[39:54] something is already trended. But at the
[39:56] same time, when you get above this line
[39:58] here, this is the warning system,
[40:00] especially when the S&P is trending. So,
[40:02] the last one where we got this where it
[40:04] broke above it while the S&P was
[40:06] trending higher and above its 50-day
[40:07] moving average was here and we went
[40:09] through a correction here. This has
[40:11] typically and going back where the back
[40:12] test is
[40:14] basically gives you a warning signal
[40:15] about 7 to 10 days. So, this will be up
[40:18] on the paywall as well.
[40:20] Um and I'll send it out occasionally uh
[40:22] on the institutional side. And I just
[40:24] want to remind you guys that for AI, you
[40:27] guys have to spend the time on it. I
[40:29] think the two best ways, uh and this is
[40:32] why I've done it this way, are on the
[40:33] investing side, which is what I'm
[40:35] showing you right now, but the other one
[40:37] is HRV. For those of you who have Aura
[40:39] rings, for those of you who monitor your
[40:41] heart rate variability, I'm telling you
[40:43] I've spent more time on this. This has
[40:45] been a long journey for me. This has
[40:47] been about 15 years of focusing on this
[40:50] concept using meditation, focusing on
[40:52] the microbiome, all of these different
[40:54] things. But if you don't know why it's
[40:56] important, if you want to be better as a
[40:59] parent, if you want to be better at your
[41:00] job, there is an element involved in
[41:03] this where the balance between your
[41:06] evolutionary side
[41:08] and your brain. So, your body and your
[41:11] brain, that is what it measures. And so,
[41:13] I've spent an enormous amount of time on
[41:15] this. I'll be rolling this out in
[41:17] Substack.
[41:18] My whole goal for this AI training side
[41:21] in the analyzing things, I approach
[41:23] things in this manner. It's a complex
[41:25] system. So, investing in health, if you
[41:27] move one thing, if you change your
[41:29] protein, it has an impact on your fiber.
[41:31] If you do this, if you extract
[41:32] something, everyone plays around with
[41:34] these. They buy self-help books. They
[41:36] buy things for investing. With AI, it
[41:38] has to be a habit. This is the McRaven
[41:40] side. There's a bet thing. Every
[41:42] decision you're making under
[41:43] uncertainty, this is the Annie Duke
[41:45] side.
[41:46] Think with intelligence, dialogue, not
[41:49] answers. AI is the collaborator on all
[41:51] of this. You do not go for it like
[41:53] Google. Too many people are looking for
[41:55] solutions. It hallucinates. It does
[41:57] this. It is not meant for that. It is
[41:59] meant to help you make better decisions.
[42:00] But to do that, you have to be thinking
[42:02] with it and you have to be asking it the
[42:04] right questions. As you do it every
[42:06] single day,
[42:08] that's what I was saying. This has
[42:09] become so powerful for me, and this is
[42:12] what I'm going to focus on. Habit comes
[42:13] before outcomes. Every decision is a
[42:16] probabilistic bet. AI works as the best
[42:19] collaborator, not an answer, not an
[42:21] oracle.
[42:23] I'm going to
[42:24] basically get you to get it into a
[42:25] habit.
[42:27] The goal is not being right. The goal is
[42:29] making better bets repeatedly. That's
[42:30] the Annie Duke philosophy, the
[42:32] collaboration side.
[42:34] I can't I really can't emphasize to you
[42:37] how much it goes through. On the prompt
[42:38] engineering side, what you're going to
[42:40] see is to give you an idea of my
[42:42] prompts, if you just type in and you go,
[42:44] "Hey, I want an answer for this." Ultra
[42:47] think reasoning optimized operating
[42:48] contract. All I do when I ask it a
[42:50] question of something that I want is
[42:52] I'll put in what I'm trying to solve for
[42:54] and then I'll type it
[42:56] I'll I'll paste in this prompt,
[42:58] which
[43:00] goes all the way here. It's two pages.
[43:03] If I want to do Socrates mode, your
[43:05] child might be someone who doesn't raise
[43:07] their hand in class, but if someone asks
[43:10] them a question, they're fine.
[43:12] This is the way. You're not here you you
[43:14] literally put it in Socrates mode, and
[43:17] that way it asks you the questions.
[43:18] Think about what a doctor does when it
[43:20] says, "What did you have for lunch? What
[43:21] did you have this?" That's what Socrates
[43:22] mode is.
[43:24] The process in one sentence show up
[43:26] daily, think in probabilities,
[43:27] collaborate with intelligence systems in
[43:29] systems that never stop changing. You
[43:31] have to do this. Investing in health are
[43:33] the easiest way in my mind. But if
[43:35] that's not what you're interested in, I
[43:37] guarantee you by reading the Substack
[43:40] and by going in and and watching the
[43:42] videos, whether it's cooking, whether
[43:44] it's fantasy football, travel planning,
[43:46] learning a new skill, anything, chess,
[43:48] painting, whatever, career development,
[43:51] any of those things, you will get the
[43:52] answer for. You'll be better in terms of
[43:54] thinking. If you're not using it now,
[43:57] this is where you're going to be at the
[43:58] end of anything involved with me. That's
[44:00] how I got to where I am with doing these
[44:02] videos and kind of showing you guys how
[44:04] to go from podcast, thematic ideas, down
[44:07] to stocks, incorporating all the
[44:09] information. Everything that I do there,
[44:11] everything that I built, the whole
[44:13] immune system thing,
[44:14] it was all done basically verbally. And
[44:16] like I said, driving out now anywhere,
[44:18] New Jersey with my kids, going through
[44:20] this, I'm just on Grok verbally. If you
[44:22] guys haven't started to do it, you're
[44:24] missing out on something big. The goal
[44:26] for this year for all of you is to get
[44:28] you to the point where AI is something
[44:30] you're collaborating with. It'll bring
[44:32] more uh power to you. And if you are
[44:34] focused on the health side, I'm telling
[44:36] you, if you take a journey with me,
[44:39] there will be hundreds of things that
[44:42] you will learn, even if you're not
[44:43] focused on HRV, but you're just focused
[44:46] on being happier and getting a better
[44:48] balance in your life between what your
[44:49] body is saying and what your mind is
[44:51] saying.
[44:52] Trust me, that's what the whole thing is
[44:53] about.
[44:54] Happy New Year to you all. Again, thanks
[44:56] for uh being here. Share the video. Do
[44:59] whatever you need. Uh we're going to
[45:01] reset for the year. We're going to make
[45:02] some money together. We're going to be
[45:04] up on the news together. And for those
[45:06] of you who join me on an AI journey, I
[45:08] promise you I will make you much more
[45:10] powerful as an individual. And for the
[45:12] younger people out there, uh our job is
[45:14] to find a way to make sure that you're
[45:15] prepared for what the world's going to
[45:16] look like. I'll see you guys again next
[45:18] week.