Alternatives Access: The Reimagined S&P 500

00:00:00:16 - 00:00:29:22

VO

This session is closed to the press. Welcome to the J.P. Morgan webcast. This is intended

for informational purposes only. The opinions expressed herein are those of the speakers

and may differ from those of other J.P. Morgan employees and affiliates. Historical

information and outlooks are not guarantees of future results. Any views and strategies

described may not be appropriate for all participants and should not be intended as

personal, investment, financial or other advice.

00:00:29:24 - 00:00:53:27

VO

As a reminder, investment products are not FDIC insured, do not have bank guarantee, and

they may lose value. The webcast may now begin.

00:00:54:00 - 00:01:14:01

Jasmine

Hello everyone, and good morning. I'm so glad that you're going to spend the next 15

minutes with us. We're back for another alt access, and I think today is a really special

episode. Long term friend and partner to the JP Morgan private Bank, but also founder and

portfolio manager of Co two management, fully BlockFi is with us. Thanks for being here

fully.

00:01:14:03 - 00:01:15:05

Philippe

Hey jazz. Thank you.

00:01:15:06 - 00:01:41:16

Jasmine

It's going to be a lot of fun. Well I say that every episode and everyone always delivers, so

I'm sure you will as well. But really, the goal for today is for you to just make us all smarter

on all things AI. You've been doing this since 1999 at Co two, and really have always been

sort of a four front thinker and investor, until we want to pull some of that out of your brain

today and make everybody smarter.

00:01:41:19 - 00:02:04:12

Jasmine

Let's just dive in. My goal is to get through five ideas in the next 15 minutes for everyone

who's joined us. And we're just going to start with the breath of the AI opportunity. So many

people are asking the question, has it run its course? Are public markets the only way to

access to this trend? What are your thoughts on sort of the wave of this technology and

where we are today?

00:02:04:19 - 00:02:35:07

Philippe

Yes. Good morning. So when we think of technology, we think a lot about these waves. And,

I've been a sort of a historian and student of technology, literally going back to 1718

hundred, the very earliest stock markets around the world. And some, principles repeat

themselves. I mean, you've got these big waves. And if you think even in the modern era,

you started with a mainframe, then you had the PC, then you connected the PC.

00:02:35:07 - 00:02:57:00

Philippe

That was the networking that gave the internet. Then we said, we want the internet in our

pocket, and then after that we said, well, let's have some applications for the phone. And

that was cloud. And now we have AI. These waves tend to last ten plus years. And I feel for

this AI. First we're probably in year three or so.

00:02:57:05 - 00:03:22:11

Philippe

So maybe we're 30% of the way there. And on top of that, this concept of sort of

encapsulating human intelligence first through software, but then in machines feels like so

big, so powerful. And the accumulation of all these previous waves to get to where we are,

that, we have a very long way to go. And this is really just the beginning.

00:03:22:12 - 00:03:43:19

Philippe

I mean, one has to imagine, like, the transformer paper was probably written ten years ago,

but Nvidia really took off at the end of 22, in the beginning of 23. And if you think of Nvidia

as being sort of the first really big sign of this AI wave, we're about three years, into it.

00:03:43:22 - 00:04:04:06

Jasmine

And it's interesting that you bring up Nvidia, I think, for call it the last 2 or 3 years, we've

been talking to our clients about accessing this trend in public markets. I think what's been

even more interesting to watch is this trend of companies staying private longer, and they're

being more access to this trend. AI in particular, also in private markets.

00:04:04:06 - 00:04:22:25

Jasmine

But these companies are bigger than ever. And so I'm curious what you think is sort of our

next idea or point to sort of think through is why are companies staying private longer? And

is this trend only accessible in either or? How do you think about sort of the transferability

between private and public?

00:04:22:27 - 00:04:51:14

Philippe

So I think at the beginning, you can access this trend in the public markets in the sense that

we had to build an initial, infrastructure layer that infrastructure layer requires a lot of

semiconductors, semiconductors, this sector that historically was very volatile still is and

sort of a little bit abandoned by the venture community. And it was these old companies

that are public that were created in the 50s, 60s and 70s.

00:04:51:17 - 00:05:21:25

Philippe

So that's how, the AI wave sort of started. Right now it's a little bit more great. Now that we

have this computing power, what do we do with it? That type of company is still private. And

so I think that an investor that doesn't have to that doesn't have exposure to okay, once I

have the AI compute, what do I do with it is at risk of basically being involved in the index of

the past instead of the index of the future.

00:05:22:02 - 00:05:44:21

Philippe

So, for instance, not being involved in entropic or open AI, I think these companies have a

chance to become very big companies in the future. If you think of the stock market today

in the US that say is a 60 or so trillion and maybe around the world it's 120. Let's imagine

that in ten years or so, the stock market could be 200 trillion.

00:05:44:23 - 00:06:09:00

Philippe

And a large company today like Apple, Apple is like 5 trillion out of the 60 of the US, or at

least 5 trillion out of 120. So Apple could represent 4% of the stock market of the world

could in the future. A couple companies that are providing incredible intelligence,

incredible application represent 4% of the stock market of the future.

00:06:09:00 - 00:06:34:28

Philippe

And the answer is like possibly right, 4% of 200 would be a trillion. Now the question is, do

you want to let a lot of that 8 trillion value accrue only to private investors? Or do you want it

to also, accrue to yourself? And that's why I think clients need to have strategies with,

managers that can sort of a little bit go to private, go to public.

00:06:35:01 - 00:06:44:01

Philippe

And at the end they just solve the question, is AI a massive trend. And do we need to have

special exposure to AI?

00:06:44:03 - 00:07:03:14

Jasmine

Well, I think it's a great point because so often our clients are deciding one or the other

when in reality, in today's world, particularly on this theme, it's probably a question of of

both. Right. And so being with a manager who has the ability to sort of invest through the

lifecycle will be important just to have sort of durability and a portfolio.

00:07:03:18 - 00:07:29:06

Jasmine

Make sure you're accessing sort of the trend holistically. You alluded to this as well, just

this idea around what could be the future of sizing or concentration in public markets. Talk

a little bit about your view on public markets today. Obviously, you've been an investor in

the public markets for almost three decades. But I think what's interesting about public

markets today is the number one thing I hear clients say it's it's incredibly concentrated.

00:07:29:08 - 00:07:44:07

Jasmine

Talk about one. Sort of just your view on on public markets in general today, whether it be

concentration or even valuations. But then how you think about readiness for a company,

you see on the private side to enter this part of the market. Yes.

00:07:44:10 - 00:08:21:22

Philippe

So I think the, the public markets, in the US, when we just sort of look at the valuation,

valuation to us appear within historical ranges. It's a great time to invest in the US. We've

got all the AI, we've got all of the imagination of the West Coast. And on top of this, we've

got all the power of the East Coast, of the financial markets that can basically draw and put

the capital together, that these dreams necess, need.

00:08:21:25 - 00:08:46:16

Philippe

If you look at Europe, Europe doesn't have the depth of, I think, financial markets and

certainly doesn't have all the AI imagination. If you look at Asia, China is sort of an island of

its own, and then the rest of Asia is obviously very strong and all the semiconductor, but

again, doesn't have necessarily all the AI software, that we have.

00:08:46:16 - 00:09:19:13

Philippe

So I think the US is in a great position. The US market is a great position. The issue with the

public markets is the public markets have become a little bit dominated by the BlackRock's

and Vanguard's and all the funds that claim they're active, but really are mostly passive and

only a tiny bit active. And as a result of that, most of the market basically follows the same

rules, follows the same weights, follows whatever the MSCI bureaucrats, tell them to do, or

the S&P on the weights.

00:09:19:13 - 00:09:44:21

Philippe

And so it's not a market that's very forward thinking. We're sort of all in a room. And

everybody thinks the same way. That's why I think there's not that many IPOs. There's not

that many IPOs because the public markets oh this company is losing money too difficult

to understand. And as a result of that, just the same way that investment grade credit was

so rigid and it basically created this private credit.

00:09:44:24 - 00:09:45:09

Jasmine

Exactly.

00:09:45:09 - 00:10:18:01

Philippe

In essence, the public markets have become so rigid that they've created this sort of private

equity market that's now becoming very big. And I think it's a mistake of the public markets

to have sort of, abandoned a lot of their sort of thinking to index based, funds. So I think

that to me, at the end of the day, you need to a little bit, if you want to participate in I be

willing to be different.

00:10:18:03 - 00:10:39:22

Philippe

And that's hard because there'll be some ups, there'll be some downs, there'll be some

volatility. But the ability to be in private, the ability to be in cash, the ability to manage risk,

the ability to be in publics that at the end of the day, I think, is this the new skill and what a

modern day investor needs to be like, or.

00:10:39:25 - 00:11:10:12

Jasmine

One of the things that I think you brought to life in that question really is this idea around

the acceptance into public markets. We all know we sort of hear that stat often 20% less

publicly traded companies today relative to 30 years ago. One of the things that's

interesting, and we saw it play out earlier this year with software, we continue to see it play

out with a number of other large cap tech names is if there's any part of your earnings call

where you talk about losing money or something just didn't meet growth expectations.

00:11:10:15 - 00:11:30:29

Jasmine

You sort of have that negative sentiment that then follows. How do you think about in this

era of intelligence and needing to continue to spend, to keep up, to be a leader, and then

even the economy of tokens in the usage there? How do you think about cost and spending

and what a company may be losing in order to be a winner later on?

00:11:31:06 - 00:11:53:11

Philippe

Yes. So I think it goes two ways for the people that are supplying AI. You have to buy the

infrastructure before you get the revenue. So if you're in hyperscaler like Amazon or

Microsoft, or even if you're OpenAI or anthropic, you're buying equipment, then you have to

train your model. When you train your model, you're not getting revenue.

00:11:53:11 - 00:12:12:11

Philippe

So this infrastructure that you have, you need to balance how much you use it for training,

how much you use it just to make revenue, making revenues the inference side versus the

training side. So these companies are taking a lot of risk trying to predict what the future

computing needs are going to be, how much should I spend on training on inference?

00:12:12:11 - 00:12:35:02

Philippe

How much as a result of that? There is a reason why some of these companies are in the

private space, like anthropic, OpenAI and the market. Certainly, it's not like the market is

nervous about Amazon and Facebook and the hyperscalers, but the market is just sort of

wondering how much are you going to spend? Is where is the ROI?

00:12:35:04 - 00:13:01:17

Philippe

Then let's say you're a corporation and you've heard a lot about Max token, and, we're

spending too much on token. I think it's okay at the beginning to experiment, and maybe

one month or another you're spending too much. But if you go back much more, simpler,

and you say, look, the US economy is probably the GDP is probably in the 30s, mid 30s.

00:13:01:24 - 00:13:14:07

Philippe

I think labor spending in the US is 20 trillion. I think the average person in the US makes, I

don't know, 60 or 70,000. Let's even say that with some extra thing it's 100,000.

00:13:14:10 - 00:13:14:21

Jasmine

Sure.

00:13:14:23 - 00:13:23:25

Philippe

Why would you not spend 5 or 10,000 a month to make your $100,000 employee much

smarter?

00:13:24:01 - 00:13:26:23

Jasmine

Yep. Much active, much more productive.

00:13:26:29 - 00:13:51:03

Philippe

And on top of that, if you don't, but then your competitor does, what do you do? And so I

think we're going to be in a world where all employees are going to be enhanced with AI.

And so if we mentioned just that labor costs in the US are 20 trillion, with 10 or 20% be a

possible number, that would be 2 or 4 trillion.

00:13:51:06 - 00:14:14:11

Philippe

The US spends about $6 trillion a year on it. Part of that 4 trillion in AI might come at the

expense of some old kind of software, some old kind of hardware. There is going to be a

little bit of reorganizing there. And so I just think that, like at the enterprise level, people are

going to spend now, let's go at the consumer level.

00:14:14:13 - 00:14:40:11

Philippe

You spend, $20 a month on Netflix. You might spend $20 a month on a second video

subscription. You might have 2 or 3 other subscription. You're telling me that you're not

going to spend 20 to 50 to maybe $100 a month to, like, be more of an intelligent person

than organize your personal life and stuff? I think that's going to happen to now that $100

might not even be cash out of your pocket.

00:14:40:11 - 00:15:07:09

Philippe

It might be generated through advertising or things like that. So I look at it and I'm like, oh

my God, there's so much potential for AI revenues. And then on top of that, we have even

spoken about intelligent machines. Cars would be the first one. Robots might be second.

What if, I, becomes big in biology and we start inventing some new drugs, new discoveries.

00:15:07:14 - 00:15:23:25

Philippe

So much can go on. And so I have the confidence that there's definitely going to be some

ups and downs. Markets can be very volatile over short periods of time, but that the long

term path for the US and for AI is a very strong path.

00:15:23:28 - 00:15:39:20

Jasmine

Well, I was going to ask you, what are you most excited about in what's to come? You gave

me a little bit of that just there. I think what's always so funny, right, is when I look at some

of these AI headlines and you go back 30 years, you saw them at the inception of the

computer or at the consumer level.

00:15:39:20 - 00:15:57:12

Jasmine

If you go back 100 years, you saw that the inception of the calculator and the telephone.

And so time sort of repeats itself when it comes to selling newspapers. But I think you gave

us a lot to think about. Any final things that you want someone to walk away? This

conversation with about what's most exciting and what's to come?

00:15:57:15 - 00:16:22:00

Philippe

I think on the AI side, if you look at it, first of all, we had the ChatGPT. You ask something

and then, oh my God, what a clever answer. Then after that we use AI to basically replace or

enhance coders. Oh my god. I used to be able to write x number of lines of code, but now

maybe it's a ten x that number.

00:16:22:03 - 00:16:43:11

Philippe

Now instead of just being a great coder, we're using and then a bit we're in the agent tech

world. Hey, let me just create some virtual human beings. We'll call them agents, and I'll

pass them with doing some work. They'll go do some work in the background. So I think

we've gone from ChatGPT to coding to the agent world.

00:16:43:14 - 00:17:01:16

Philippe

To me, the next big step is going to be, persistent AI. And what I mean by that is that even in

the world of agent, it's like you're asking the agents to do thing they do think and they report

back to you. But if you think about the AI in the way it should be done is the AI should be.

00:17:01:16 - 00:17:30:03

Philippe

Maybe I have a little mike here and the mic turns on so that people around me can see

when I'm recording or not. In time, I'll be comfortable that things are recorded and we'll just

ask for things not to be recorded or your AI will be smart enough to sort of turn itself on and

off. And the AI should be able to say and listen to me and listen to you and at all times

maybe say, oh, Philip, I noticed that you're going to see jazz this morning.

00:17:30:03 - 00:17:51:26

Philippe

You're going to record something at JPMorgan. Look, based on all the conversations I've

heard you in the last three months, here are the five things that I would highlight that might

be of interest to the clients. So now the AI is, like, persistent. I mean, it's always around me

and the AI is offering me the solution before I need to ask for it.

00:17:51:28 - 00:18:17:19

Philippe

The AI could be in my ear and said, hey, Philip, by the way, you mentioned this point once,

three weeks ago. Declined that, really liked it. Why don't you close with this? Super

interesting. And so I just think, and if you think about it from a computing need, when you

ask something to the AI, it only works when you ask, but when it listens, then it means it's

always using computing in order to come up with the right answer.

00:18:17:22 - 00:18:47:07

Philippe

So almost like you need the AI to be computing 24 over seven now. So I think the computing

needs will explode. And so going forward, I see a world where, the cost of computing goes

down dramatically. One where models get so much smarter every year that if a model

required ten units of compute to get to an answer, maybe next year it's five, next year it's

two and a half, next year it's one.

00:18:47:10 - 00:19:17:12

Philippe

And so models are going to get so much smarter. So in general, the amount of intelligence

divided by the amount of hardware is going to explode. We're going to need less and less

hardware to make better and better decisions. We're living in a really sort of interesting

world. I've been doing this probably investing in tech since the early 90s, and I still think

that this is probably the biggest wave that I will invest in, you know, in my lifetime.

00:19:17:12 - 00:19:32:03

Jasmine

It's amazing. Well, before we sat down, I tasked you with, saying something that at least I'd

never heard before. I'm sure you gave me at least ten. And so hopefully, for everyone who

joined us this morning, this was incredibly interesting. Felipe, it's always such a pleasure to

spend time with you. Thanks so much.

00:19:32:03 - 00:19:33:29

Philippe

Thanks for having me, jazz. Cheers.

00:19:34:01 - 00:19:46:28

Jasmine

As always, I want to thank you all for hopping on and spending your morning with us. It was

a pleasure to get to spend the time chatting with Felipe this morning. I hope that we see you

all for the next one. Thanks for joining All Access.

00:19:47:01 - 00:19:58:15

VO

Thank you for joining us. Prior to making financial or investment decisions, you should

speak with a qualified professional in your JP Morgan team. This concludes today's

webcast. You may now disconnect.