May 16, 2023 • 42 mins
Artificial intelligence has arrived courtesy of ChatGPT, the large language model software that already has more than 100 million users. ChatGPT’s debut signals that any number of jobs could be disrupted (and replaced) by bots, including money management – the science and art of successful investing for institutions and individuals. Investing has already been transformed over the last several decades by computers; an avalanche of ubiquitous global market, corporate, and financial data; oceans of liquidity; stronger risk management tools; and evermore probing and state-of-the-art quantitative analysis. AI promises to up the ante further. Aaron Brown and Nir Kaissar are both contributing columnists for Bloomberg Opinion and successful investors – Aaron is the former chief risk officer for one of the world’s largest hedge funds, AQR Capital Management, while Nir is the founder of Unison Advisors, an investment firm specializing in multi-asset portfolios.
See omnystudio.com/listener for privacy information.
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Speaker 1 (00:01):
Welcome to Crash Course, a podcast about business, political, and
social disruption and what we can learn from it. I'm
Tim O'Brien. Today's Crash Course AI versus Money Managers, as
we've all heard by now, AI or artificial intelligence has
arrived courtesy of chat GPT, the large language model software
(00:24):
that already has more than one hundred million users. It
processes questions and turns out brilliant answers in the blink
of an eye. It isn't sentiented at software after all,
but still it gives every appearance of having the ability
to learn and adapt to refine its output in increasingly
sophisticated ways. We're also used to economic and technological revolutions
(00:49):
battering blue collar workers the hardest, but chat GPT's debut
signals that any number of white collar jobs could be
disrupted and replaced by bots. Writing, tea, teaching, programming, client services,
analysis of many stripes, consulting, healthcare, and many other professions
may all be up for grabs in this brave new world.
(01:11):
Money management, the science and art of successfully investing for
institutions and individuals could also be in ais crosshairs. Investing
has already been transformed over the last several decades by computers,
an avalanche of ubiquitous global market, corporate and financial data,
oceans of liquidity, stronger risk management tools, and evermore probing
(01:34):
and state of the art quantitative analysis. AI promises to
up the ante even further, joining me to talk about
whether or not bots are going to devour money. Manager's
business and livelihoods are Aaron Brown and Near Kasar. Both
of these gentlemen are contributing columnists for Boomberg Opinion, and
both of them are successful investors. Aaron, you're the former
(01:57):
chief risk officer for one of the world's largest funds,
AQR Capital Management, and a prolific author. Welcome to the show.
Speaker 2 (02:06):
Thank you very much.
Speaker 1 (02:06):
Tim, and Niir you're the founder of Unison Advisors, an
investment firm specializing in multi asset portfolios. Thanks for joining us.
Thanks Tim okay. So, this is the first time I've
had two guests on simultaneously, So let's see how this goes.
I'd like you both to jump in when you feel inspired,
but I'll also try to steer things along so our
listeners can keep track. And a warning, guys, if you
(02:29):
start throwing around alpha's beta's VARs and other fancy lingo.
I'm going to stop you and ask you to explain yourselves.
All right, Eron, let's start with you, and let's move
through some of the basics. What is money management? To
find it for me? And I'm looking for an umbrella
term here, which is why I chose money management. Maybe
(02:51):
I should have stuck with plain old investing. Anyway, have
at it. What is money management?
Speaker 2 (02:56):
Well, from a practical business standpoint, there are two main
things you have to do. One of them is to
get people to trust you with their money, and the
second is to either convince them that you're doing this
or to actually do this is to provide them any
better return they can get from purely passive index funds
or other inexpensive simple techniques. And you know it can
(03:20):
range from while even running index funds on one hand,
just getting them more cheaply than other people, to some
of the most sophisticated strategies where you demand fifty percent
of the profits in return for delivering or promising really
exceptional returns.
Speaker 1 (03:38):
Do you agree with that definition?
Speaker 3 (03:40):
I agree with what Erin has said. I would just
take one step back and say we all have a
practical problem that we face, and the practical problem that
we have is if we are fortunate enough to have
savings in our life, then we have to do something
with those savings. And you can put that in the
bank or there's various places that you can put in,
but you have to do something with it, and primarily
you have to put it. It's great if you can
(04:01):
grow it, but you don't want to lose the money
that you save, and so you have an issue, and
the question is what is the best way to handle
that money. There are professionals money managers, let's say, who
will come along and we'll manage it for you, just
like there are people who will do any number of
services in your life, and we'll charge your fee to
do it. Increasingly, it's getting easier to do that yourself.
But at the end of the day, money management seeks
(04:22):
to solve a very simple problem, which is I have savings.
I want to preserve and grow those savings.
Speaker 1 (04:27):
So now that we know what money management is, let's
take that up a step further than here. What is
successful money management?
Speaker 3 (04:34):
Well, that's a trickier question, and you're going to get
a far broader range of opinions on that. I think
what I would say is, ultimately, this also is a
practical question, which is, you have savings that we have
said you want to preserve and grow, and at the
same time you have inflation nipping at your heels. In
other words, every single day that you hold a dollar,
the value of that dollar goes down because inflation is
(04:56):
catching up to it. So at the very least, if
you want to preserve the value of your savings, you
have to make it grow faster than the rate of inflation.
And there are various ways that you can do that.
You can invest in bonds, you can invest in stocks,
all things you can get into if you want. But
the important thing to know is you can't just take
and stuff it in your mattress, because if you do,
the value of that will go down over time.
Speaker 1 (05:17):
And so expertise matters in this realm.
Speaker 3 (05:21):
Expertise is important, I would say, to avoid many of
the pitfalls. However, one of the things that's interesting about
technology and the way that markets have evolved is expertise
is becoming less and less important if your goal is
merely to try to grow your money faster than the
rate of inflation.
Speaker 1 (05:38):
So Aaron jump in, do you agree with Near's notion
of what successful money management.
Speaker 2 (05:44):
Is, I don't disagree with him, but most of successful
money management is in the head of the investor or
the beneficial owner. Yes, it's nice if your money grows
faster than inflation, but it's more important that you sleep
at night, that you feel financially secure, that you trust
what's going on. And so a lot of money management
(06:07):
is about either reassuring individuals or explaining to them precisely
what you are going to do when you are going
to lose or make money, or, in the case of institutions,
dealing with their institutional constraints peer risk. You know, chief
investment officers of pension funds seem to spend a lot
more time worrying about how they're outperforming their peers than
(06:28):
about any kind of absolute goal. So I think money
management goes well beyond getting an inflation beating return or
reducing risk. The investor has to understand and appreciate it.
Speaker 1 (06:41):
So you think it actually goes into being a nanny
or a therapist.
Speaker 2 (06:46):
Well, there is quite a lot of that. You spend
an awful lot of time as a money manager on
the phone with clients, prepairing presentations, that make people feel
good about what you're doing.
Speaker 1 (06:55):
I guess you know, boy, we could really just go
off on that one, because obviously comes out of the
fact that people are ignorant, as we all are, about
any number of things, including money, and if they're not
doing the hard work to get up to speed on
how they're investing their money, it opens up an ability
for a money manager to give them calming advice, soothing advice,
(07:18):
stuff that'll let them sleep well at night. That's a
different kind of intelligence, and I'm not saying it's not worthy,
but that's eq to a certain extent as opposed to IQ.
Speaker 2 (07:28):
Right, I would agree with that, Yes, it is, or
in the case of institutions, it's I don't know, there's
a thing for that, you know, institutional queue.
Speaker 1 (07:37):
And I guess it can open the door to hucksterism too, Right,
If you're able to soothe people's nerves who actually aren't
up to speed fully on how the markets work or
what they should do with their money, they can also
become marks, can't they.
Speaker 2 (07:50):
Yeah, that's absolutely true, but I think people tend to
veer too quickly in that direction. Traditional financial advisors, the
kind of people we used to used you used to
work with individuals. Some of them were hucksters, some of
them were just churning for commissions and putting people in
all kinds of bad stuff, But many of them really
gave their clients a secure, happy financial life. They understood
(08:14):
insurance and taxes and wills and other matters. They took
care of a million details that keep people up at night.
A lot of people today are unhappy with all the
choices and information and things they have to worry about.
And the old fashioned financial advisor, even charging fairly high fees,
really gave many people a great service.
Speaker 1 (08:35):
There's also a lot more information because of technological developments
available to investors now. There's a certain amount of transparency
and intellectual dexterity that's available to average investors that wasn't present, say,
even three decades ago. Isn't that right, Neir?
Speaker 3 (08:51):
Certainly, there are a lot of people doing a lot
of great things within money management, and there are some
people I would say the bias here here it tends
to be more sort of in the direction of Wall Street.
There's still a lot of shenanigans that go on. I
would say the industry taking advantage of people who aren't
as well versed, but certainly there are more tools available
to ordinary investors today than there have ever been, and
(09:15):
it's easier for them to manage their money if they
want to manage their money, and cheaper for them to
do it than it has ever been. And it's so
cheap and easy actually that it's hard to imagine it
moving further in that direction. The question is whether we
can get the information we as an industry can get
that information to ordinary investors, and maybe more importantly, whether
we have the financial incentive to get that information to investors.
(09:39):
And that's something that the industry has been grappling with
for probably ten or twenty years, as this information and
the ability to invest cheaply and yourself has become available.
Speaker 1 (09:49):
So in that context, then what goal ideally does successful
investing serve from.
Speaker 3 (09:56):
The perspective of the investor? Yes, I mean I think
I will marry sort of the two perspectives that are complimentary,
not opposing that Aaron and I have both offered. I mean, ultimately,
what you want is you want to preserve and grow
your money, and you also want to be at peace
with your money. I mean, there are a lot of
people and everybody knows people like this who are just
(10:19):
constantly anxious about how much money they have and do
they have enough money? And are they going to lose
their money? And you know, money is a big deal.
And ultimately, yes, you want your money to be in
a good place and you want it to grow, but
you also want to be in harmony with your money.
You don't want to have to be anxious about it
all the time. And so that is the picture in
my view of like a full life in relation to
(10:42):
your money.
Speaker 1 (10:43):
Is that sort of where you come down erin?
Speaker 2 (10:45):
Yeah, personally, I kind of I'm perfectly happy to put
a lot of my money in index funds and just
take the average and you know, if it goes up,
goes up. If it goes down, it goes down. But
many people are not you know, many people want to
know that someone's looking at their money. That person may
not be able to prove they actually do any good
that you know, when they pull out of companies, those
(11:06):
companies go down. When they overinvest overweight a company, it
goes up, but it's very reassuring to people and know
someone is watching it.
Speaker 3 (11:14):
The question, though, I would ask Tim and Aaron, is
the people who don't prefer to invest in index funds?
Are they doing that from a knowing place? In other words,
Aaron I would argue, the reason you prefer index funds
is because you're highly sophisticated about financial markets, and you
have all of the evidence in a peer of view literature,
et cetera at your fingertips, and you know what it says,
(11:35):
and you probably know that you're best off in index funds.
The question is, are most people that's sophisticated when they
say they prefer something other than an index fund? Are
they making a knowing decision? Are they making a decision
out of ignorance? Or have they been directed by an
industry to more expensive products? And index funds is an
(11:55):
alternative fully sort of on a fully informed basis.
Speaker 2 (11:59):
Yeah, yeah, my argument, And of course maybe I'm the
last person to tell how good my thinking is, but yeah,
I mean I like the index funds, and I also
likes I pay very high fees or use some very
sophisticated strategies where I think I have an edge. There
is a middle ground of sort of your traditional actively
managed mutual fund that has no strong track record of
(12:21):
success that I think are pretty clearly not good that
pretty clearly lose money to inflation and taxes in the
long run after fees, and especially do not outperform index funds.
And that is clearly a marketing effort that this product
has been sold more or less for the same reason
that you know, people pick a name brand Coca Cola
(12:41):
over a generic store cola with identical ingredients. You know,
it's the branding.
Speaker 3 (12:46):
But Aaron, this is what I mean by making a
knowing decision. I think when we say there are some
good managers some bad managers, I think we underestimate the problem.
I mean, the evidence is that the vast majority of
people who are going to try to beat the index
are going to use to the index, and that includes
the most sophisticated managers charging the highest fees. So it's
not just that, yes, you could find a manager who's
(13:07):
going to beat the index, but the probability of your
doing that, and you're doing that in advance, is very,
very very low. Would you agree with that?
Speaker 2 (13:15):
No, No, I wouldn't. I believe that if you apply
sensible filters, so you kind of filter out the people
who aren't doing anything than people with actually bad track
records and so on, I think that certainly an institution
can do a much better job combining some low fee
index products with some very sophisticated high fee products. Not
(13:36):
necessarily that we'll beat the index, but that will combine
with the index to give a better risk adjusted return.
The average retail investor who doesn't have access to the
good hedge funds, who doesn't have the time or expertise
to pick solid mutual funds from just branded ones that
have no advantage, I think that person is much better
(13:57):
off sticking with the cheapest, most tax efficient, well DIVERSI
fied index funds.
Speaker 1 (14:02):
Now, we could stay on this particular topic for the
rest of the show, but we're not doing index funds.
We're going to get into AI. The only reason I'm
calling a halt to this avenue of discussion is because
we'll never keep going on to the other ones, which
means you guys will have to come back again so
we can talk about other things. One other foundational idea
I want to lay down now as we continue to
(14:23):
talk about this is the idea of is investing different
from trading? And I want to make that distinction now
because I want to come back to it later when
we talk about AI's role in all of this. I
think of trading as a time constrained tool that might
involve strategy, but I think of investing as a less
time constrained philosophy that has to involve strategy. But go ahead,
(14:48):
erin school me if you disagree with how I've partsed
the differences between these two things.
Speaker 2 (14:53):
Yeah, that's a good way to do it. And in
the context of AI and machine learning, they've been enormously
successful than trading, and in fact, really for about thirty
years now, computers have done all the serious trading, not
necessarily with artificial intelligence, the original algorithms and systematic things.
We're very simple. Now you think, you know, trading is machines.
(15:16):
It's like, you know, the autopilot and an airplane. It's
just much much better than humans investing. So far, we
have not seen much real success from artificial intelligence.
Speaker 1 (15:27):
Near Where do you reside on that?
Speaker 3 (15:29):
I agree with all that, I would just add for me,
there's an element of duration, So I tend to think
of trading as just a short term bet, whereas investing
is a longer term bet, and it's not necessarily binary.
You know, one can sort of define those in different ways,
but certainly the longer term your horizon, the more you're
investing in, the less you're trading.
Speaker 1 (15:46):
Okay, now that we've cleared up all of that, I
want to take a brief break to hear from one
of our sponsors, and then we'll be right back. We're
back with two smarties, Aaron Brown in their case, are
and we're talking about whether AI is going to up
end Wall Street and the legions of people who work
as money managers in one form or another. So Aaron,
(16:09):
let's continue. We've tried to define what investing is. Now
let's talk about how AI might tip the entire Apple card.
Lead us off with that thought, what do you think
the arrival of AI means for the investing world?
Speaker 2 (16:23):
Well, okay, so we're talking about investment here, not trading.
I spent an awful out of the last fifteen years
trying to use advances and machine learning and artificial intelligence
and data science to improve investing decisions. I mean, in theory,
you can turn your algorithm loose unlimited amounts of data
in any language, reacting instantly to news announcements before a
(16:45):
human has read the first word. It can digest whether
it's positive or negative and put in trades. And this
huge broad perspective, able to instantly correlate information of totally
different accounting data, research reports of government statistics and so on,
should allow you to make much better investment decisions. Should
be able to tell you you know which companies are
(17:06):
going to succeed and grow and which ones are going
to stumble and fall. It just doesn't yet seem to work,
and I don't know why.
Speaker 1 (17:13):
Last year you wrote, and I'll quote you, that a
tireless machine, able to digest all information and immune to
biases should be clearly superior to humans when it comes
to investing. Except it's not close.
Speaker 2 (17:27):
Quote.
Speaker 1 (17:28):
You've also said that humans are essential in the investment process.
So how so why does warm blood matter in this argument?
Speaker 2 (17:36):
Well, that's a dated quote. The humans being essential in
the investment process. I said was that machine learning tends
to be a black box and it can't explain itself.
You can't give it new information and come up with
a better joint decision. But chat GPT kind of blew
that out of the water. Now computers are actually better
(17:57):
at explaining their decisions and conversing with you. You can
give it new information, you can ask it questions and
you can change it OWT comes. Unfortunately, what we know
about chat GPT is that it often gives very, very convincing,
persuasive answers that sound like the world's greatest expert that
are completely wrong. So we need to take that next step.
(18:18):
But if we can take that next step and chat
chief TP confine itself to actual things, it knows. Yeah,
I'm not sure we do need humans anymore.
Speaker 1 (18:26):
You know, you cited to research papers in a more
recent column you wrote, and you noted that the interpretation
of text and financial price movements may forecast a different
outcome with AI. You said, it's a long way from
taking over Wall Street, but there's no reason to think
it can't. Is that how you see it? Still?
Speaker 2 (18:47):
Yeah, what we've seen, the successes we have seen is
you know what I call the wrong answer Faster machines
are getting pretty good at guessing how humans will interpret something,
and they can do it in microseconds instead of half
an hour, So you get a huge trading advantage from that,
But you're still only predicting how humans will view something.
(19:09):
You're not actually getting at fundamental economic reality.
Speaker 1 (19:12):
So near is AI going to eat money? Manager's lunch.
Speaker 3 (19:17):
I think it's entirely plausible that AI will displace what
humans do now. But I think to fully understand that
we have to set the stage of what is happening
today before you inject AI into it. And that is
to say, you have the market at large, the stock
market A lot of people look at, say the SP
five hundred or whatever it is. You have the market
(19:38):
on the stock side. You have the market on the
bond side, and that is already automated. If you want
to buy the market, the bond market or the stock market,
you can do that now with very little interference from
human beings, and so technology has already taken that. Over
on the other side, you have a bunch of humans
running around and there has been for a couple of
decades trying to do better than that. And this is
(19:59):
the part that it's entirely plausible that AI will come
in and replace those people. So now AI is going
to try to do better than the market. And the
thing that we have to understand about that is that
the vast majority of these humans running around trying to
do this are wasting their time because the vast majority
of them will fail at it. Over any reasonable period,
more than ninety percent of them will not be able
(20:19):
to do it, And so you can stop and ask
yourself philosophically, why are they doing this if they can't win.
Let's leave that aside for one second, although we can
come back to it if you want. Now inject AI
into those people's chairs and ask yourself a couple of questions.
One is, Okay, if the AI replaces the humans and
the humans can't do it, is there a reason to
believe that AI is going to do it any better
(20:41):
than them? And the truth is, we don't know the
answer to that question. But let's give AI the benefit
of the doubt, and let's say that AI is so
much smarter than what humans can do that it will
now be able to beat the market to do what
humans have always wanted to do. The problem with that
is the first person in the AI is going to
have an advantage, clearly, but as AI becomes more adopted,
(21:04):
because ultimately profits are going to attract competition, everybody has
an AI, then no one can beat the market.
Speaker 1 (21:11):
In other words, you lose the first mover advantage over time.
Speaker 3 (21:14):
Exactly if it turns out that AI will be as
good as an investor. As we think, then what will
happen is markets will become more efficient. Whatever advantages AI
finds in the markets will disappear, and ultimately no one
will be able to win.
Speaker 1 (21:30):
You know that ties in. I guess to the argument
you made earlier that people are better off investing in
index funds because anything that tracks a broad market will
generally outperform what individual stock pickers or people trying to
beat averages are going to be able to do over time.
Based on the performance data. We already know that's out there,
(21:51):
and you would say that that's going to occrue to
AI as well.
Speaker 3 (21:56):
Well, Yes, one of two things is going to happen.
Either AI is as good as the human but no better,
in which case that's a fail. Or AI is a
lot better than the humans, in which case the competition
among all the AI bots is going to basically the
profits from that activity will disappear, which, by the way,
it's worth just saying we have a history here active managers'
(22:16):
humans have gotten better over time. If you went and
you looked at the average stock picker in the nineteen
sixties and nineteen seventies, that person was a lot less
skilled than the active managers today. And so we have
a history of humans becoming better at their craft, and
we also have a track record. And what do we
know about that track record is as the humans became
better at their craft, counterintuitively, their ability to generate excess returns,
(22:39):
in other words, to beat the market went down.
Speaker 1 (22:42):
Because so many other people were developing the same expertise contemporaneously.
Speaker 3 (22:46):
Exactly. Competition makes it harder to make money, not the reverse.
I think what that history tells you is that AI
is likely to do the same thing. If it turns
out that AI is better than the humans.
Speaker 1 (22:57):
I was going to say, we could put Warren Buffett
off to the side in this equation, although even more
recently he hasn't had the same kind of fantastic returns
he had early in his career.
Speaker 3 (23:05):
That's right, and that might explain part of it.
Speaker 1 (23:07):
So, Aaron, you work for a large hedge fund, come
to the defense of hedges and active stock picking. Here.
If no active stock pickers can reliably beat the market,
then maybe AI can either Or am I wrong headed
in that perspective?
Speaker 2 (23:22):
Yes, you and near are wrong headed. And this is
a fundamental issue. And I think the view expressed byoneer
is deeply pessimistic and ignores the connections between finance and
the real economy. Here's how I see it. The reason
the index funds are both so cheap and efficient and
(23:43):
produce such a great return, it's not magic. It is
ordained by God. When the world began, it's because of
work by more aggressive investors. That's individuals who are aggressive
and talented, and also hedge funds. And the way I
think about it is, these investors go out, they look
for exotic assets, liquid assets, They trade things in new ways,
(24:08):
and they generate alpha. They beat the market. Of course,
as they do this, you get more liquidity in those assets,
you get more familiarity and information, and so what I
call hedge fund beta comes up. These are people who
perform sort of well known hedge fund strategies like merger
arbitrage or convertible arbitrage or something like that. But they
(24:28):
learned how to do it, you know, from a paper
or from working at some other place, and they kind
of do it mechanically, and that's worth some fees to
investors more than index funds. But it's nothing special, and
eventually it turns into beta. It turns into what anybody
can get in an index fund, but it's a continuous process.
Speaker 1 (24:45):
I told you at some point i'd stop you. I
knew you'd say beta. So let's just tell our listeners
right now what is beta and what is alpha.
Speaker 2 (24:52):
Well, for the purposes of this discussion, beta is something
anybody can get cheap or free. So beta is what
you get from an index fun. Alpha is what it
takes work and money to get. So it's an additional
return you can get, but only by investing more in
systems and people, in information whatever. But the point is
this is a continuous process, and this does two things.
(25:15):
It makes the economy more efficient, and it gives index
funds eventually, not immediately, but eventually the index funds improve
their risk adjusted returns because they're embracing more and more
strategies that used to be alpha and now have migrated
all the way to beta. The promise of AI and
investing is in making the economy better, in directing capital
(25:39):
to the firms that are going to use it to
grow and succeed, and directing capital away from those people
who are going to waste it. And you know, if
it can do this, then a long term expected return
on the market above inflation could move up from you know,
the six percent it was in a twentieth century to eight, ten,
twelve percent. And what this does is it means more
(25:59):
and more people, ordinary people with median or below median
incomes can save money from their own work and have
financial security and retire and comfort. The higher the return
you can get from real economy, the more financially secure
people could be. So this is how we will measure
the return of AI. It may be absolutely the case
(26:21):
that in fifty years AI is just beating the index funds,
but only because AI made the index funds so much better.
Speaker 1 (26:28):
Near you, you had a broad smile on your face
at a couple different moments here, tell me what's on
your mind.
Speaker 3 (26:34):
Well, you know, I have a column for Bloomer Opinion
about A and investing, and one of the things that
I mentioned in there half jokingly is that one of
the things that indexers, people who are fans of index
funds often here from active managers, is that active managers
are very important because it is their function. It is
what they do that allows index funds to work. Along
(26:55):
the lines of what Aaron laid out. And while all
of that is true, what amuses me, is this idea
that AI could come in and do that function along
the lines of what Aaron is describing. And the reason
that's humorous to me is because indexers can finally stop
listening to the winding from active managers that what they
do is so important that they didn't go away, and
if they went away, that it would be difficult for
(27:15):
indexers to do what they do. But that's where I
am optimistic. I think that to the except that AI
can come along and do the job of active managers
and can take over from Wall Street, then perhaps that
function can be done more cheaply, perhaps markets can become
even more efficient than they are, and perhaps we can
stop having these discussions about the fact that if there
weren't any active managers, then indexers would ruin the world.
(27:38):
So from that perspective, I'm relatively optimistic.
Speaker 1 (27:41):
Okay, on that note of optimism, let's take another break.
I want to hear from our sponsor, and we will
come right back. We're back with Aaron Brown in their
CaSR and we're talking about whether AI is going to
upend professional investing near You've gone so far as to
say that large language models like chat, GPT looking at
(28:06):
FED statements on inflation or looking at financial news aren't
really going to make a difference because both of those
things FED statements and financial news have very little or
perhaps no predictive value. So it doesn't matter who's analyzing
them that there's no inherent predictive value in them. Am
(28:26):
I interpreting your thoughts on that correctly? Yeah, that's right.
Speaker 3 (28:29):
I mean, I think the point of departure question that
we should ask ourselves about anything that is being automated.
Is that process worth automating to begin with? And you have,
and you have long had humans on Wall Street and
elsewhere looking for advantages, looking for ways to outsmart the market,
(28:50):
reading headlines, mining headlines for information, looking at the pronouncements
of central bankers and other policy makers to see what
that would portend markets. And you know, the best evidence
that we have is that none of this is useful,
because at the end of the day, most of those
people don't do any better than the market. And so
you have to stop and ask yourself, why are we
(29:11):
automating something that, by all available evidence, has no use
The only answer that you can really come up with,
I think, is to say, well, there's information there humans
may just not be smart enough to glean it, whereas
AI may come along and be able to do a
better job of that.
Speaker 1 (29:28):
In fact, it's quite possible that AI could be better
at pattern recognition than humans are, and all of that
documentation and data may have a predictive value in it
that we haven't discovered yet for sure.
Speaker 3 (29:40):
But I guess the point is we should be skeptical
about that. We should be open minded about that, but
we should be skeptical, and we should be mindful that
ultimately what we're asking AI to do is something that
we have really very little evidence that that activity has
any value today.
Speaker 1 (29:56):
Erin, I want to jump in here with a counter
argument to near that. I will ask you to voice
for me, because you're smarter than I am. But I
think of a firm like Renaissance. They've had one fund
that specializes in quantitative analysis. They've beat the market for
decades with that fund. It's a boutique fund. It is
not a large fund by many hedge fund firms, sizes
(30:17):
maybe ten billion dollars in assets. Why couldn't AI eventually
just be a version of that Renaissance fund on steroids?
Speaker 2 (30:26):
I think not only could it it might already be.
We don't. Renaissance, of course, is one of the most
intensely secretive firms, and we don't really know exactly what
they do. But that's almost just a change of degree.
Renaissance has always done the kinds of models that are
similar to artificial intelligence, and in fact, what I would
say is that the key difference or the key you know,
(30:47):
the rubicon you cross, is when you let the model
start updating itself. So Renaissance, what it seems to be
is that people are using very similar kind of pattern recognition.
Many of the original algorithms they used were based on
speech recognition, which is an area of artificial intelligence, but
they did not allow the models to update themselves. They
(31:09):
did not use machine learning. They used human learning to
monitor these models. They may already have switched over, and
I would be surprised if they haven't at least looked
very hard at it to let the models update themselves.
But I think Renaissance moving to artificial intelligence would be
sort of a minor tweak to their operation.
Speaker 1 (31:28):
I guess you've raised an interesting point in that Renaissance's
outperformance existed in a black box and their own internal
model that no one else can see, and we're lying
on their own reporting that they've outperformed. No one's thought
that they've been lying about that, including their very happy
investors for decades. But if that black box essentially comes
(31:50):
out into the open through AI and becomes a tool
for the masses, you no longer have to have vast
computing power, You no longer have to have spoke inputs
to have a quantitative model that produces superlative returns. Doesn't
that mean then that over time everything will revert to
(32:10):
the mean, and that your advantage in data processing and
data analytics gets shaved because everyone has the same tool
you do.
Speaker 2 (32:20):
Well. Yes, that's true. So Renaissance may be out of
business in ten years because there's a Renaissance GPT out
there that anybody can get free on the Internet. But
if that's true, I think that will mean the economy
is growing faster than it otherwise would have. I think
that means there will be generally smarter investment decisions.
Speaker 1 (32:42):
And I suppose if an important role in your worldview
of money managers is they allow people to sleep soundly
at night, because it nurses you to sleep with the
thought that your money will be all well, and good
bots probably can't do a great job of that part
of investment advice. Right maybe two hundred years from now,
some will to listen to this podcast and laugh at
(33:03):
us because there's a loving robot sitting next to someone's
bed patting their wallet. But at least in the near
term right now, that's really not on the horizon, the
sort of EQ side of this.
Speaker 2 (33:13):
Equation, right, No, I would disagree with that. I mean,
chat GPT is very reassuring. The problem is it's not
right often enough, and so we have to solve that
problem eron.
Speaker 1 (33:25):
I think chat GPT might be reassuring to you because
you're an expert and you're in the markets every day.
But I don't know that someone at home who's worried
about their mortgage or their IRA or their kids' college
savings would feel an answer coming back out of a
machine is the same as a human being saying you
might want to do this with your money, you might
want to do that with your money.
Speaker 2 (33:44):
I don't know. I mean, it produces much more reassuring
answers than a human If you called up your stockbroker
and say, how come the stock you recommended last week
went down. You know, he's going to scramble around. He's
going to give you some platitudes. Chat Gep is going
to say, well, you know, there was dis announcement this day,
but I'm not really worried about it because of this
and that, And it will give you a very You
(34:05):
will feel that you are speaking to an expert who
really knows and is honest and never admits any kind
of doubt about things. Whether or not people are going
to trust that. I mean, granted, a lot of people
are going to say, well, that comes from a computer,
I can't trust it, just like many people would be
nervous if they knew that a computer was flying their
airplane or a computer was running their electrical grid or
things like that. But that's been true for decades, and
(34:28):
many essential functions that your life depends on are decisions
made by computers.
Speaker 1 (34:33):
So then that aspect of this that you raised being
in the show that an important part of money management
is padding people on the hand, is actually completely dispensable.
Speaker 2 (34:42):
Well except that some of those very convincing answers are true,
and until you solve that problem that it's equally convincing
when it's right and when it's wrong, people are going
to have trouble trusting it.
Speaker 1 (34:57):
Neier. Would you turn your money over to a bot?
Speaker 3 (35:00):
Well, I mean I would argue that I mostly do,
because you know, like I said, most index funds, I
would argue, are bots. And one of the things that
we haven't touched on yet, but it's I think worth mentioning,
is that a lot of the methods that historically humans
used to try to beat the market to generate what
(35:23):
Aaron called alpha, and alpha just define as returns above
the market. It's a little bit more complicated than that,
but I think it's a good, useful, simple definition. A
lot of those methods have already been automated and now
are indexes. So a lot of the ways that we
traditionally thought of as alpha are now just wrapped in
an index fund. You can buy it and a bot
(35:43):
will do it for you, and that's great too. And
so one of the things just thinking about this sort
of from a financial evolution perspective, what's going to happen
is AI is going to come along and is going
to say, hey, I found alpha, and that alpha will
be automated delivered on mass, and all of a sudden,
it will no longer be alpha, and you could imagine
(36:03):
a scenario where that just keeps going endlessly. But the
thing is, the more likely outcome is that and we're
seeing this already. When you look at traditional sources of alpha,
what you see is that the excess return that they
generated historically is higher than the alpha that they produce today.
And that's just because of adoption. You could imagine a
(36:24):
world where both the number of sources of alpha diminishes
and their impact also diminishes. And so you have these
bots phonetically going out there trying to find new sources,
new sources, new sources, and their efficacy diminishes over time.
Speaker 1 (36:39):
Because ALPA is always getting absorbed and then replicated.
Speaker 3 (36:42):
Right, and ask yourself, I mean, is there an infinite
amount of alpha in the universe? I don't know the
answer to that. My guess is probably not. But what
does that do to the pricing power of the industry?
In other words, thirty forty years ago, what was alpha?
People charged one to two percent of your money a year.
For now they're charging zero point one percent four Right, Well,
(37:02):
there's two reasons. One is it's become automated, but also
because it just doesn't generate as much alph as it
used to, right, And so you could imagine a world
where AI becomes ubiquitous, where AI is awesome in terms
of investing, and it just brings down the cost of everything.
So that this world that we inhabit now that Aaron
is describing, where you have hedge funds and other people
(37:25):
at the top of the mountain who are deemed to
be worth paying two and twenty two percent of your
money a year plus twenty percent of the profits, where
they just can't demand those kinds of fees anymore. And
that's quite an optimistic picture, indeed, from my perspective.
Speaker 1 (37:40):
So infinite alpha would be a good name for a
hedge fund since you brought it up, or an AI
or an AI.
Speaker 2 (37:46):
But I'd like to say we can measure the amount
of potential alpha. It's not infinite. It's roughly three times
the current value of everything in the world. And the
way I get to that figure is if we had
infinite alpha, if we had you know, perfect AI systems,
then every investment would return the risk free rate because
there would be no risk. And in that case, the
(38:07):
world is worth about three times what it's worth today,
at least I'm talking about the financial assets of the world,
so that makes a lot of people rich. I mean,
that would be a really really good world if we
could do it, and we've got lots and lots of
work to make that happen. And if that's the case,
and if computers are running everything and everybody is three
times as rich as they are now, well maybe that's
an okay world and we can stop worrying about investing.
Speaker 1 (38:30):
Arin. We don't let anybody escape the show without telling
us what they've learned from some of the collisions we're
observing and the disruptions we've observed. What have you learned
watching the advent of chat GPT and AI's future coming
a little closer to the present as it pertains to
(38:51):
money management.
Speaker 2 (38:52):
Well before last fall, before the chat GPT and the
reaction to it, I would have said humans are going
to be essential to the money management process for decades
at least, because they can explain what they do, because
they can have conversations with their investors. Chat GPT convinces
me that's no longer a barrier, both the quality of
(39:14):
the program itself and how people react to it. We
still have the problem of getting artificial intelligence and machine
learning to make good investment decisions, or better investment decisions
than humans. That has not been solved that I haven't
learned much about in the last six months, but what
I previously saw as the main barrier to adoption and
has gone away.
Speaker 1 (39:34):
No, how about you, what have you learned?
Speaker 3 (39:36):
I mean, I would echo largely what Aaron has said,
which is, it's shocking how human like chat GPT can be,
and it's shocking how early in this whole process we are,
and it's as good as it is, and so, you know,
having no other information, I'm inclined to extrapolate into the
future and say I don't see why it couldn't do
(39:59):
everything that that everyone in the money management business does,
from back office people on Wall Street to client facing advisors.
But this is my real concern about it from an
investing perspective. If I can bring it up here, which
is I think all new technology eventually, and I think
we're probably closer to that day than even I think,
(40:22):
takes on an air of a can't miss investing opportunity,
and I worry as I listen to these kinds of conversations,
as I read a lot of the coverage about AI
and chat, GPT, et cetera, that that hype is quickly
revving up in the AI space. What that means is
that a lot of people, and those people are already
invested in AI. The early investors, the people with their
(40:43):
chips already on the table, are going to make a
fortune on it. But what's going to happen is this
going to attract a lot more investment later on, and
those people, inevitably a lot of those people are going
to get in late in the game, just before sort
of all the hype pops and the prices for AI
as it's rest in financial markets come down, and people
will lose a lot of money. I mean, we've seen
(41:03):
this just in the last couple of decades. We've seen
this with Internet, with crypto, with almost everything. And I
worry that from that perspective, AI is going to cost
people more money that's going to make.
Speaker 1 (41:13):
Them Aaron and Nere, we're out of time. Thanks for
helping us sort out a complex topic, the collision of
bots with humans.
Speaker 2 (41:21):
Thank you very much, Miir Tiem.
Speaker 1 (41:23):
Thank you.
Speaker 3 (41:23):
Erin.
Speaker 1 (41:24):
You can find Aaron brown in near Kasar's columns on
the Bloomberg Opinion website, and you can follow Near on
Twitter at near Kasar and if you want to learn
more about AI chatbots, we did another episode with fellow
columnists parme Elsen and Tyler Cowen. You can find that
in the crash Course show feed here at crash Course.
We believe that collisions can be messy, impressive, challenging, surprising,
(41:49):
and always instructive. In today's crash Course, I learned that AI,
while it seems to have an advantage that's unbeatable, may
just like humans, loses it over time. What did you learn?
We'd love to hear from you. You can tweet at
the Bloomberg Opinion handle at Opinion or me at Tim
(42:09):
O'Brien using the hashtag Bloomberg crash Course. You can also
subscribe to our show wherever you're listening right now and
leave us a review. It helps people find the show.
This episode was produced by the Indispensable Animasarakus, moses On,
Dam and Me. Our supervising producer is Magnus Hendrickson, and
we had editing help from Stage Bauman, Katie Boyce, Jeff Grocott,
(42:33):
Mike Nizza, and Christine Banden Bilart. Blake Maples says our
sound engineering and our original theme song was composed by
Luis Gara. I'm Tim O'Brien. We'll be back next week
with another crash course.
Welcome to Crash Course, a podcast about business, political, and
social disruption and what we can learn from it. I'm
Tim O'Brien. Today's Crash Course AI versus Money Managers, as
we've all heard by now, AI or artificial intelligence has
arrived courtesy of chat GPT, the large language model software
(00:24):
that already has more than one hundred million users. It
processes questions and turns out brilliant answers in the blink
of an eye. It isn't sentiented at software after all,
but still it gives every appearance of having the ability
to learn and adapt to refine its output in increasingly
sophisticated ways. We're also used to economic and technological revolutions
(00:49):
battering blue collar workers the hardest, but chat GPT's debut
signals that any number of white collar jobs could be
disrupted and replaced by bots. Writing, tea, teaching, programming, client services,
analysis of many stripes, consulting, healthcare, and many other professions
may all be up for grabs in this brave new world.
(01:11):
Money management, the science and art of successfully investing for
institutions and individuals could also be in ais crosshairs. Investing
has already been transformed over the last several decades by computers,
an avalanche of ubiquitous global market, corporate and financial data,
oceans of liquidity, stronger risk management tools, and evermore probing
(01:34):
and state of the art quantitative analysis. AI promises to
up the ante even further, joining me to talk about
whether or not bots are going to devour money. Manager's
business and livelihoods are Aaron Brown and Near Kasar. Both
of these gentlemen are contributing columnists for Boomberg Opinion, and
both of them are successful investors. Aaron, you're the former
(01:57):
chief risk officer for one of the world's largest funds,
AQR Capital Management, and a prolific author. Welcome to the show.
Speaker 2 (02:06):
Thank you very much.
Speaker 1 (02:06):
Tim, and Niir you're the founder of Unison Advisors, an
investment firm specializing in multi asset portfolios. Thanks for joining us.
Thanks Tim okay. So, this is the first time I've
had two guests on simultaneously, So let's see how this goes.
I'd like you both to jump in when you feel inspired,
but I'll also try to steer things along so our
listeners can keep track. And a warning, guys, if you
(02:29):
start throwing around alpha's beta's VARs and other fancy lingo.
I'm going to stop you and ask you to explain yourselves.
All right, Eron, let's start with you, and let's move
through some of the basics. What is money management? To
find it for me? And I'm looking for an umbrella
term here, which is why I chose money management. Maybe
(02:51):
I should have stuck with plain old investing. Anyway, have
at it. What is money management?
Speaker 2 (02:56):
Well, from a practical business standpoint, there are two main
things you have to do. One of them is to
get people to trust you with their money, and the
second is to either convince them that you're doing this
or to actually do this is to provide them any
better return they can get from purely passive index funds
or other inexpensive simple techniques. And you know it can
(03:20):
range from while even running index funds on one hand,
just getting them more cheaply than other people, to some
of the most sophisticated strategies where you demand fifty percent
of the profits in return for delivering or promising really
exceptional returns.
Speaker 1 (03:38):
Do you agree with that definition?
Speaker 3 (03:40):
I agree with what Erin has said. I would just
take one step back and say we all have a
practical problem that we face, and the practical problem that
we have is if we are fortunate enough to have
savings in our life, then we have to do something
with those savings. And you can put that in the
bank or there's various places that you can put in,
but you have to do something with it, and primarily
you have to put it. It's great if you can
(04:01):
grow it, but you don't want to lose the money
that you save, and so you have an issue, and
the question is what is the best way to handle
that money. There are professionals money managers, let's say, who
will come along and we'll manage it for you, just
like there are people who will do any number of
services in your life, and we'll charge your fee to
do it. Increasingly, it's getting easier to do that yourself.
But at the end of the day, money management seeks
(04:22):
to solve a very simple problem, which is I have savings.
I want to preserve and grow those savings.
Speaker 1 (04:27):
So now that we know what money management is, let's
take that up a step further than here. What is
successful money management?
Speaker 3 (04:34):
Well, that's a trickier question, and you're going to get
a far broader range of opinions on that. I think
what I would say is, ultimately, this also is a
practical question, which is, you have savings that we have
said you want to preserve and grow, and at the
same time you have inflation nipping at your heels. In
other words, every single day that you hold a dollar,
the value of that dollar goes down because inflation is
(04:56):
catching up to it. So at the very least, if
you want to preserve the value of your savings, you
have to make it grow faster than the rate of inflation.
And there are various ways that you can do that.
You can invest in bonds, you can invest in stocks,
all things you can get into if you want. But
the important thing to know is you can't just take
and stuff it in your mattress, because if you do,
the value of that will go down over time.
Speaker 1 (05:17):
And so expertise matters in this realm.
Speaker 3 (05:21):
Expertise is important, I would say, to avoid many of
the pitfalls. However, one of the things that's interesting about
technology and the way that markets have evolved is expertise
is becoming less and less important if your goal is
merely to try to grow your money faster than the
rate of inflation.
Speaker 1 (05:38):
So Aaron jump in, do you agree with Near's notion
of what successful money management.
Speaker 2 (05:44):
Is, I don't disagree with him, but most of successful
money management is in the head of the investor or
the beneficial owner. Yes, it's nice if your money grows
faster than inflation, but it's more important that you sleep
at night, that you feel financially secure, that you trust
what's going on. And so a lot of money management
(06:07):
is about either reassuring individuals or explaining to them precisely
what you are going to do when you are going
to lose or make money, or, in the case of institutions,
dealing with their institutional constraints peer risk. You know, chief
investment officers of pension funds seem to spend a lot
more time worrying about how they're outperforming their peers than
(06:28):
about any kind of absolute goal. So I think money
management goes well beyond getting an inflation beating return or
reducing risk. The investor has to understand and appreciate it.
Speaker 1 (06:41):
So you think it actually goes into being a nanny
or a therapist.
Speaker 2 (06:46):
Well, there is quite a lot of that. You spend
an awful lot of time as a money manager on
the phone with clients, prepairing presentations, that make people feel
good about what you're doing.
Speaker 1 (06:55):
I guess you know, boy, we could really just go
off on that one, because obviously comes out of the
fact that people are ignorant, as we all are, about
any number of things, including money, and if they're not
doing the hard work to get up to speed on
how they're investing their money, it opens up an ability
for a money manager to give them calming advice, soothing advice,
(07:18):
stuff that'll let them sleep well at night. That's a
different kind of intelligence, and I'm not saying it's not worthy,
but that's eq to a certain extent as opposed to IQ.
Speaker 2 (07:28):
Right, I would agree with that, Yes, it is, or
in the case of institutions, it's I don't know, there's
a thing for that, you know, institutional queue.
Speaker 1 (07:37):
And I guess it can open the door to hucksterism too, Right,
If you're able to soothe people's nerves who actually aren't
up to speed fully on how the markets work or
what they should do with their money, they can also
become marks, can't they.
Speaker 2 (07:50):
Yeah, that's absolutely true, but I think people tend to
veer too quickly in that direction. Traditional financial advisors, the
kind of people we used to used you used to
work with individuals. Some of them were hucksters, some of
them were just churning for commissions and putting people in
all kinds of bad stuff, But many of them really
gave their clients a secure, happy financial life. They understood
(08:14):
insurance and taxes and wills and other matters. They took
care of a million details that keep people up at night.
A lot of people today are unhappy with all the
choices and information and things they have to worry about.
And the old fashioned financial advisor, even charging fairly high fees,
really gave many people a great service.
Speaker 1 (08:35):
There's also a lot more information because of technological developments
available to investors now. There's a certain amount of transparency
and intellectual dexterity that's available to average investors that wasn't present, say,
even three decades ago. Isn't that right, Neir?
Speaker 3 (08:51):
Certainly, there are a lot of people doing a lot
of great things within money management, and there are some
people I would say the bias here here it tends
to be more sort of in the direction of Wall Street.
There's still a lot of shenanigans that go on. I
would say the industry taking advantage of people who aren't
as well versed, but certainly there are more tools available
to ordinary investors today than there have ever been, and
(09:15):
it's easier for them to manage their money if they
want to manage their money, and cheaper for them to
do it than it has ever been. And it's so
cheap and easy actually that it's hard to imagine it
moving further in that direction. The question is whether we
can get the information we as an industry can get
that information to ordinary investors, and maybe more importantly, whether
we have the financial incentive to get that information to investors.
(09:39):
And that's something that the industry has been grappling with
for probably ten or twenty years, as this information and
the ability to invest cheaply and yourself has become available.
Speaker 1 (09:49):
So in that context, then what goal ideally does successful
investing serve from.
Speaker 3 (09:56):
The perspective of the investor? Yes, I mean I think
I will marry sort of the two perspectives that are complimentary,
not opposing that Aaron and I have both offered. I mean, ultimately,
what you want is you want to preserve and grow
your money, and you also want to be at peace
with your money. I mean, there are a lot of
people and everybody knows people like this who are just
(10:19):
constantly anxious about how much money they have and do
they have enough money? And are they going to lose
their money? And you know, money is a big deal.
And ultimately, yes, you want your money to be in
a good place and you want it to grow, but
you also want to be in harmony with your money.
You don't want to have to be anxious about it
all the time. And so that is the picture in
my view of like a full life in relation to
(10:42):
your money.
Speaker 1 (10:43):
Is that sort of where you come down erin?
Speaker 2 (10:45):
Yeah, personally, I kind of I'm perfectly happy to put
a lot of my money in index funds and just
take the average and you know, if it goes up,
goes up. If it goes down, it goes down. But
many people are not you know, many people want to
know that someone's looking at their money. That person may
not be able to prove they actually do any good
that you know, when they pull out of companies, those
(11:06):
companies go down. When they overinvest overweight a company, it
goes up, but it's very reassuring to people and know
someone is watching it.
Speaker 3 (11:14):
The question, though, I would ask Tim and Aaron, is
the people who don't prefer to invest in index funds?
Are they doing that from a knowing place? In other words,
Aaron I would argue, the reason you prefer index funds
is because you're highly sophisticated about financial markets, and you
have all of the evidence in a peer of view literature,
et cetera at your fingertips, and you know what it says,
(11:35):
and you probably know that you're best off in index funds.
The question is, are most people that's sophisticated when they
say they prefer something other than an index fund? Are
they making a knowing decision? Are they making a decision
out of ignorance? Or have they been directed by an
industry to more expensive products? And index funds is an
(11:55):
alternative fully sort of on a fully informed basis.
Speaker 2 (11:59):
Yeah, yeah, my argument, And of course maybe I'm the
last person to tell how good my thinking is, but yeah,
I mean I like the index funds, and I also
likes I pay very high fees or use some very
sophisticated strategies where I think I have an edge. There
is a middle ground of sort of your traditional actively
managed mutual fund that has no strong track record of
(12:21):
success that I think are pretty clearly not good that
pretty clearly lose money to inflation and taxes in the
long run after fees, and especially do not outperform index funds.
And that is clearly a marketing effort that this product
has been sold more or less for the same reason
that you know, people pick a name brand Coca Cola
(12:41):
over a generic store cola with identical ingredients. You know,
it's the branding.
Speaker 3 (12:46):
But Aaron, this is what I mean by making a
knowing decision. I think when we say there are some
good managers some bad managers, I think we underestimate the problem.
I mean, the evidence is that the vast majority of
people who are going to try to beat the index
are going to use to the index, and that includes
the most sophisticated managers charging the highest fees. So it's
not just that, yes, you could find a manager who's
(13:07):
going to beat the index, but the probability of your
doing that, and you're doing that in advance, is very,
very very low. Would you agree with that?
Speaker 2 (13:15):
No, No, I wouldn't. I believe that if you apply
sensible filters, so you kind of filter out the people
who aren't doing anything than people with actually bad track
records and so on, I think that certainly an institution
can do a much better job combining some low fee
index products with some very sophisticated high fee products. Not
(13:36):
necessarily that we'll beat the index, but that will combine
with the index to give a better risk adjusted return.
The average retail investor who doesn't have access to the
good hedge funds, who doesn't have the time or expertise
to pick solid mutual funds from just branded ones that
have no advantage, I think that person is much better
(13:57):
off sticking with the cheapest, most tax efficient, well DIVERSI
fied index funds.
Speaker 1 (14:02):
Now, we could stay on this particular topic for the
rest of the show, but we're not doing index funds.
We're going to get into AI. The only reason I'm
calling a halt to this avenue of discussion is because
we'll never keep going on to the other ones, which
means you guys will have to come back again so
we can talk about other things. One other foundational idea
I want to lay down now as we continue to
(14:23):
talk about this is the idea of is investing different
from trading? And I want to make that distinction now
because I want to come back to it later when
we talk about AI's role in all of this. I
think of trading as a time constrained tool that might
involve strategy, but I think of investing as a less
time constrained philosophy that has to involve strategy. But go ahead,
(14:48):
erin school me if you disagree with how I've partsed
the differences between these two things.
Speaker 2 (14:53):
Yeah, that's a good way to do it. And in
the context of AI and machine learning, they've been enormously
successful than trading, and in fact, really for about thirty
years now, computers have done all the serious trading, not
necessarily with artificial intelligence, the original algorithms and systematic things.
We're very simple. Now you think, you know, trading is machines.
(15:16):
It's like, you know, the autopilot and an airplane. It's
just much much better than humans investing. So far, we
have not seen much real success from artificial intelligence.
Speaker 1 (15:27):
Near Where do you reside on that?
Speaker 3 (15:29):
I agree with all that, I would just add for me,
there's an element of duration, So I tend to think
of trading as just a short term bet, whereas investing
is a longer term bet, and it's not necessarily binary.
You know, one can sort of define those in different ways,
but certainly the longer term your horizon, the more you're
investing in, the less you're trading.
Speaker 1 (15:46):
Okay, now that we've cleared up all of that, I
want to take a brief break to hear from one
of our sponsors, and then we'll be right back. We're
back with two smarties, Aaron Brown in their case, are
and we're talking about whether AI is going to up
end Wall Street and the legions of people who work
as money managers in one form or another. So Aaron,
(16:09):
let's continue. We've tried to define what investing is. Now
let's talk about how AI might tip the entire Apple card.
Lead us off with that thought, what do you think
the arrival of AI means for the investing world?
Speaker 2 (16:23):
Well, okay, so we're talking about investment here, not trading.
I spent an awful out of the last fifteen years
trying to use advances and machine learning and artificial intelligence
and data science to improve investing decisions. I mean, in theory,
you can turn your algorithm loose unlimited amounts of data
in any language, reacting instantly to news announcements before a
(16:45):
human has read the first word. It can digest whether
it's positive or negative and put in trades. And this
huge broad perspective, able to instantly correlate information of totally
different accounting data, research reports of government statistics and so on,
should allow you to make much better investment decisions. Should
be able to tell you you know which companies are
(17:06):
going to succeed and grow and which ones are going
to stumble and fall. It just doesn't yet seem to work,
and I don't know why.
Speaker 1 (17:13):
Last year you wrote, and I'll quote you, that a
tireless machine, able to digest all information and immune to
biases should be clearly superior to humans when it comes
to investing. Except it's not close.
Speaker 2 (17:27):
Quote.
Speaker 1 (17:28):
You've also said that humans are essential in the investment process.
So how so why does warm blood matter in this argument?
Speaker 2 (17:36):
Well, that's a dated quote. The humans being essential in
the investment process. I said was that machine learning tends
to be a black box and it can't explain itself.
You can't give it new information and come up with
a better joint decision. But chat GPT kind of blew
that out of the water. Now computers are actually better
(17:57):
at explaining their decisions and conversing with you. You can
give it new information, you can ask it questions and
you can change it OWT comes. Unfortunately, what we know
about chat GPT is that it often gives very, very convincing,
persuasive answers that sound like the world's greatest expert that
are completely wrong. So we need to take that next step.
(18:18):
But if we can take that next step and chat
chief TP confine itself to actual things, it knows. Yeah,
I'm not sure we do need humans anymore.
Speaker 1 (18:26):
You know, you cited to research papers in a more
recent column you wrote, and you noted that the interpretation
of text and financial price movements may forecast a different
outcome with AI. You said, it's a long way from
taking over Wall Street, but there's no reason to think
it can't. Is that how you see it? Still?
Speaker 2 (18:47):
Yeah, what we've seen, the successes we have seen is
you know what I call the wrong answer Faster machines
are getting pretty good at guessing how humans will interpret something,
and they can do it in microseconds instead of half
an hour, So you get a huge trading advantage from that,
But you're still only predicting how humans will view something.
(19:09):
You're not actually getting at fundamental economic reality.
Speaker 1 (19:12):
So near is AI going to eat money? Manager's lunch.
Speaker 3 (19:17):
I think it's entirely plausible that AI will displace what
humans do now. But I think to fully understand that
we have to set the stage of what is happening
today before you inject AI into it. And that is
to say, you have the market at large, the stock
market A lot of people look at, say the SP
five hundred or whatever it is. You have the market
(19:38):
on the stock side. You have the market on the
bond side, and that is already automated. If you want
to buy the market, the bond market or the stock market,
you can do that now with very little interference from
human beings, and so technology has already taken that. Over
on the other side, you have a bunch of humans
running around and there has been for a couple of
decades trying to do better than that. And this is
(19:59):
the part that it's entirely plausible that AI will come
in and replace those people. So now AI is going
to try to do better than the market. And the
thing that we have to understand about that is that
the vast majority of these humans running around trying to
do this are wasting their time because the vast majority
of them will fail at it. Over any reasonable period,
more than ninety percent of them will not be able
(20:19):
to do it, And so you can stop and ask
yourself philosophically, why are they doing this if they can't win.
Let's leave that aside for one second, although we can
come back to it if you want. Now inject AI
into those people's chairs and ask yourself a couple of questions.
One is, Okay, if the AI replaces the humans and
the humans can't do it, is there a reason to
believe that AI is going to do it any better
(20:41):
than them? And the truth is, we don't know the
answer to that question. But let's give AI the benefit
of the doubt, and let's say that AI is so
much smarter than what humans can do that it will
now be able to beat the market to do what
humans have always wanted to do. The problem with that
is the first person in the AI is going to
have an advantage, clearly, but as AI becomes more adopted,
(21:04):
because ultimately profits are going to attract competition, everybody has
an AI, then no one can beat the market.
Speaker 1 (21:11):
In other words, you lose the first mover advantage over time.
Speaker 3 (21:14):
Exactly if it turns out that AI will be as
good as an investor. As we think, then what will
happen is markets will become more efficient. Whatever advantages AI
finds in the markets will disappear, and ultimately no one
will be able to win.
Speaker 1 (21:30):
You know that ties in. I guess to the argument
you made earlier that people are better off investing in
index funds because anything that tracks a broad market will
generally outperform what individual stock pickers or people trying to
beat averages are going to be able to do over time.
Based on the performance data. We already know that's out there,
(21:51):
and you would say that that's going to occrue to
AI as well.
Speaker 3 (21:56):
Well, Yes, one of two things is going to happen.
Either AI is as good as the human but no better,
in which case that's a fail. Or AI is a
lot better than the humans, in which case the competition
among all the AI bots is going to basically the
profits from that activity will disappear, which, by the way,
it's worth just saying we have a history here active managers'
(22:16):
humans have gotten better over time. If you went and
you looked at the average stock picker in the nineteen
sixties and nineteen seventies, that person was a lot less
skilled than the active managers today. And so we have
a history of humans becoming better at their craft, and
we also have a track record. And what do we
know about that track record is as the humans became
better at their craft, counterintuitively, their ability to generate excess returns,
(22:39):
in other words, to beat the market went down.
Speaker 1 (22:42):
Because so many other people were developing the same expertise contemporaneously.
Speaker 3 (22:46):
Exactly. Competition makes it harder to make money, not the reverse.
I think what that history tells you is that AI
is likely to do the same thing. If it turns
out that AI is better than the humans.
Speaker 1 (22:57):
I was going to say, we could put Warren Buffett
off to the side in this equation, although even more
recently he hasn't had the same kind of fantastic returns
he had early in his career.
Speaker 3 (23:05):
That's right, and that might explain part of it.
Speaker 1 (23:07):
So, Aaron, you work for a large hedge fund, come
to the defense of hedges and active stock picking. Here.
If no active stock pickers can reliably beat the market,
then maybe AI can either Or am I wrong headed
in that perspective?
Speaker 2 (23:22):
Yes, you and near are wrong headed. And this is
a fundamental issue. And I think the view expressed byoneer
is deeply pessimistic and ignores the connections between finance and
the real economy. Here's how I see it. The reason
the index funds are both so cheap and efficient and
(23:43):
produce such a great return, it's not magic. It is
ordained by God. When the world began, it's because of
work by more aggressive investors. That's individuals who are aggressive
and talented, and also hedge funds. And the way I
think about it is, these investors go out, they look
for exotic assets, liquid assets, They trade things in new ways,
(24:08):
and they generate alpha. They beat the market. Of course,
as they do this, you get more liquidity in those assets,
you get more familiarity and information, and so what I
call hedge fund beta comes up. These are people who
perform sort of well known hedge fund strategies like merger
arbitrage or convertible arbitrage or something like that. But they
(24:28):
learned how to do it, you know, from a paper
or from working at some other place, and they kind
of do it mechanically, and that's worth some fees to
investors more than index funds. But it's nothing special, and
eventually it turns into beta. It turns into what anybody
can get in an index fund, but it's a continuous process.
Speaker 1 (24:45):
I told you at some point i'd stop you. I
knew you'd say beta. So let's just tell our listeners
right now what is beta and what is alpha.
Speaker 2 (24:52):
Well, for the purposes of this discussion, beta is something
anybody can get cheap or free. So beta is what
you get from an index fun. Alpha is what it
takes work and money to get. So it's an additional
return you can get, but only by investing more in
systems and people, in information whatever. But the point is
this is a continuous process, and this does two things.
(25:15):
It makes the economy more efficient, and it gives index
funds eventually, not immediately, but eventually the index funds improve
their risk adjusted returns because they're embracing more and more
strategies that used to be alpha and now have migrated
all the way to beta. The promise of AI and
investing is in making the economy better, in directing capital
(25:39):
to the firms that are going to use it to
grow and succeed, and directing capital away from those people
who are going to waste it. And you know, if
it can do this, then a long term expected return
on the market above inflation could move up from you know,
the six percent it was in a twentieth century to eight, ten,
twelve percent. And what this does is it means more
(25:59):
and more people, ordinary people with median or below median
incomes can save money from their own work and have
financial security and retire and comfort. The higher the return
you can get from real economy, the more financially secure
people could be. So this is how we will measure
the return of AI. It may be absolutely the case
(26:21):
that in fifty years AI is just beating the index funds,
but only because AI made the index funds so much better.
Speaker 1 (26:28):
Near you, you had a broad smile on your face
at a couple different moments here, tell me what's on
your mind.
Speaker 3 (26:34):
Well, you know, I have a column for Bloomer Opinion
about A and investing, and one of the things that
I mentioned in there half jokingly is that one of
the things that indexers, people who are fans of index
funds often here from active managers, is that active managers
are very important because it is their function. It is
what they do that allows index funds to work. Along
(26:55):
the lines of what Aaron laid out. And while all
of that is true, what amuses me, is this idea
that AI could come in and do that function along
the lines of what Aaron is describing. And the reason
that's humorous to me is because indexers can finally stop
listening to the winding from active managers that what they
do is so important that they didn't go away, and
if they went away, that it would be difficult for
(27:15):
indexers to do what they do. But that's where I
am optimistic. I think that to the except that AI
can come along and do the job of active managers
and can take over from Wall Street, then perhaps that
function can be done more cheaply, perhaps markets can become
even more efficient than they are, and perhaps we can
stop having these discussions about the fact that if there
weren't any active managers, then indexers would ruin the world.
(27:38):
So from that perspective, I'm relatively optimistic.
Speaker 1 (27:41):
Okay, on that note of optimism, let's take another break.
I want to hear from our sponsor, and we will
come right back. We're back with Aaron Brown in their
CaSR and we're talking about whether AI is going to
upend professional investing near You've gone so far as to
say that large language models like chat, GPT looking at
(28:06):
FED statements on inflation or looking at financial news aren't
really going to make a difference because both of those
things FED statements and financial news have very little or
perhaps no predictive value. So it doesn't matter who's analyzing
them that there's no inherent predictive value in them. Am
(28:26):
I interpreting your thoughts on that correctly? Yeah, that's right.
Speaker 3 (28:29):
I mean, I think the point of departure question that
we should ask ourselves about anything that is being automated.
Is that process worth automating to begin with? And you have,
and you have long had humans on Wall Street and
elsewhere looking for advantages, looking for ways to outsmart the market,
(28:50):
reading headlines, mining headlines for information, looking at the pronouncements
of central bankers and other policy makers to see what
that would portend markets. And you know, the best evidence
that we have is that none of this is useful,
because at the end of the day, most of those
people don't do any better than the market. And so
you have to stop and ask yourself, why are we
(29:11):
automating something that, by all available evidence, has no use
The only answer that you can really come up with,
I think, is to say, well, there's information there humans
may just not be smart enough to glean it, whereas
AI may come along and be able to do a
better job of that.
Speaker 1 (29:28):
In fact, it's quite possible that AI could be better
at pattern recognition than humans are, and all of that
documentation and data may have a predictive value in it
that we haven't discovered yet for sure.
Speaker 3 (29:40):
But I guess the point is we should be skeptical
about that. We should be open minded about that, but
we should be skeptical, and we should be mindful that
ultimately what we're asking AI to do is something that
we have really very little evidence that that activity has
any value today.
Speaker 1 (29:56):
Erin, I want to jump in here with a counter
argument to near that. I will ask you to voice
for me, because you're smarter than I am. But I
think of a firm like Renaissance. They've had one fund
that specializes in quantitative analysis. They've beat the market for
decades with that fund. It's a boutique fund. It is
not a large fund by many hedge fund firms, sizes
(30:17):
maybe ten billion dollars in assets. Why couldn't AI eventually
just be a version of that Renaissance fund on steroids?
Speaker 2 (30:26):
I think not only could it it might already be.
We don't. Renaissance, of course, is one of the most
intensely secretive firms, and we don't really know exactly what
they do. But that's almost just a change of degree.
Renaissance has always done the kinds of models that are
similar to artificial intelligence, and in fact, what I would
say is that the key difference or the key you know,
(30:47):
the rubicon you cross, is when you let the model
start updating itself. So Renaissance, what it seems to be
is that people are using very similar kind of pattern recognition.
Many of the original algorithms they used were based on
speech recognition, which is an area of artificial intelligence, but
they did not allow the models to update themselves. They
(31:09):
did not use machine learning. They used human learning to
monitor these models. They may already have switched over, and
I would be surprised if they haven't at least looked
very hard at it to let the models update themselves.
But I think Renaissance moving to artificial intelligence would be
sort of a minor tweak to their operation.
Speaker 1 (31:28):
I guess you've raised an interesting point in that Renaissance's
outperformance existed in a black box and their own internal
model that no one else can see, and we're lying
on their own reporting that they've outperformed. No one's thought
that they've been lying about that, including their very happy
investors for decades. But if that black box essentially comes
(31:50):
out into the open through AI and becomes a tool
for the masses, you no longer have to have vast
computing power, You no longer have to have spoke inputs
to have a quantitative model that produces superlative returns. Doesn't
that mean then that over time everything will revert to
(32:10):
the mean, and that your advantage in data processing and
data analytics gets shaved because everyone has the same tool
you do.
Speaker 2 (32:20):
Well. Yes, that's true. So Renaissance may be out of
business in ten years because there's a Renaissance GPT out
there that anybody can get free on the Internet. But
if that's true, I think that will mean the economy
is growing faster than it otherwise would have. I think
that means there will be generally smarter investment decisions.
Speaker 1 (32:42):
And I suppose if an important role in your worldview
of money managers is they allow people to sleep soundly
at night, because it nurses you to sleep with the
thought that your money will be all well, and good
bots probably can't do a great job of that part
of investment advice. Right maybe two hundred years from now,
some will to listen to this podcast and laugh at
(33:03):
us because there's a loving robot sitting next to someone's
bed patting their wallet. But at least in the near
term right now, that's really not on the horizon, the
sort of EQ side of this.
Speaker 2 (33:13):
Equation, right, No, I would disagree with that. I mean,
chat GPT is very reassuring. The problem is it's not
right often enough, and so we have to solve that
problem eron.
Speaker 1 (33:25):
I think chat GPT might be reassuring to you because
you're an expert and you're in the markets every day.
But I don't know that someone at home who's worried
about their mortgage or their IRA or their kids' college
savings would feel an answer coming back out of a
machine is the same as a human being saying you
might want to do this with your money, you might
want to do that with your money.
Speaker 2 (33:44):
I don't know. I mean, it produces much more reassuring
answers than a human If you called up your stockbroker
and say, how come the stock you recommended last week
went down. You know, he's going to scramble around. He's
going to give you some platitudes. Chat Gep is going
to say, well, you know, there was dis announcement this day,
but I'm not really worried about it because of this
and that, And it will give you a very You
(34:05):
will feel that you are speaking to an expert who
really knows and is honest and never admits any kind
of doubt about things. Whether or not people are going
to trust that. I mean, granted, a lot of people
are going to say, well, that comes from a computer,
I can't trust it, just like many people would be
nervous if they knew that a computer was flying their
airplane or a computer was running their electrical grid or
things like that. But that's been true for decades, and
(34:28):
many essential functions that your life depends on are decisions
made by computers.
Speaker 1 (34:33):
So then that aspect of this that you raised being
in the show that an important part of money management
is padding people on the hand, is actually completely dispensable.
Speaker 2 (34:42):
Well except that some of those very convincing answers are true,
and until you solve that problem that it's equally convincing
when it's right and when it's wrong, people are going
to have trouble trusting it.
Speaker 1 (34:57):
Neier. Would you turn your money over to a bot?
Speaker 3 (35:00):
Well, I mean I would argue that I mostly do,
because you know, like I said, most index funds, I
would argue, are bots. And one of the things that
we haven't touched on yet, but it's I think worth mentioning,
is that a lot of the methods that historically humans
used to try to beat the market to generate what
(35:23):
Aaron called alpha, and alpha just define as returns above
the market. It's a little bit more complicated than that,
but I think it's a good, useful, simple definition. A
lot of those methods have already been automated and now
are indexes. So a lot of the ways that we
traditionally thought of as alpha are now just wrapped in
an index fund. You can buy it and a bot
(35:43):
will do it for you, and that's great too. And
so one of the things just thinking about this sort
of from a financial evolution perspective, what's going to happen
is AI is going to come along and is going
to say, hey, I found alpha, and that alpha will
be automated delivered on mass, and all of a sudden,
it will no longer be alpha, and you could imagine
(36:03):
a scenario where that just keeps going endlessly. But the
thing is, the more likely outcome is that and we're
seeing this already. When you look at traditional sources of alpha,
what you see is that the excess return that they
generated historically is higher than the alpha that they produce today.
And that's just because of adoption. You could imagine a
(36:24):
world where both the number of sources of alpha diminishes
and their impact also diminishes. And so you have these
bots phonetically going out there trying to find new sources,
new sources, new sources, and their efficacy diminishes over time.
Speaker 1 (36:39):
Because ALPA is always getting absorbed and then replicated.
Speaker 3 (36:42):
Right, and ask yourself, I mean, is there an infinite
amount of alpha in the universe? I don't know the
answer to that. My guess is probably not. But what
does that do to the pricing power of the industry?
In other words, thirty forty years ago, what was alpha?
People charged one to two percent of your money a year.
For now they're charging zero point one percent four Right, Well,
(37:02):
there's two reasons. One is it's become automated, but also
because it just doesn't generate as much alph as it
used to, right, And so you could imagine a world
where AI becomes ubiquitous, where AI is awesome in terms
of investing, and it just brings down the cost of everything.
So that this world that we inhabit now that Aaron
is describing, where you have hedge funds and other people
(37:25):
at the top of the mountain who are deemed to
be worth paying two and twenty two percent of your
money a year plus twenty percent of the profits, where
they just can't demand those kinds of fees anymore. And
that's quite an optimistic picture, indeed, from my perspective.
Speaker 1 (37:40):
So infinite alpha would be a good name for a
hedge fund since you brought it up, or an AI
or an AI.
Speaker 2 (37:46):
But I'd like to say we can measure the amount
of potential alpha. It's not infinite. It's roughly three times
the current value of everything in the world. And the
way I get to that figure is if we had
infinite alpha, if we had you know, perfect AI systems,
then every investment would return the risk free rate because
there would be no risk. And in that case, the
(38:07):
world is worth about three times what it's worth today,
at least I'm talking about the financial assets of the world,
so that makes a lot of people rich. I mean,
that would be a really really good world if we
could do it, and we've got lots and lots of
work to make that happen. And if that's the case,
and if computers are running everything and everybody is three
times as rich as they are now, well maybe that's
an okay world and we can stop worrying about investing.
Speaker 1 (38:30):
Arin. We don't let anybody escape the show without telling
us what they've learned from some of the collisions we're
observing and the disruptions we've observed. What have you learned
watching the advent of chat GPT and AI's future coming
a little closer to the present as it pertains to
(38:51):
money management.
Speaker 2 (38:52):
Well before last fall, before the chat GPT and the
reaction to it, I would have said humans are going
to be essential to the money management process for decades
at least, because they can explain what they do, because
they can have conversations with their investors. Chat GPT convinces
me that's no longer a barrier, both the quality of
(39:14):
the program itself and how people react to it. We
still have the problem of getting artificial intelligence and machine
learning to make good investment decisions, or better investment decisions
than humans. That has not been solved that I haven't
learned much about in the last six months, but what
I previously saw as the main barrier to adoption and
has gone away.
Speaker 1 (39:34):
No, how about you, what have you learned?
Speaker 3 (39:36):
I mean, I would echo largely what Aaron has said,
which is, it's shocking how human like chat GPT can be,
and it's shocking how early in this whole process we are,
and it's as good as it is, and so, you know,
having no other information, I'm inclined to extrapolate into the
future and say I don't see why it couldn't do
(39:59):
everything that that everyone in the money management business does,
from back office people on Wall Street to client facing advisors.
But this is my real concern about it from an
investing perspective. If I can bring it up here, which
is I think all new technology eventually, and I think
we're probably closer to that day than even I think,
(40:22):
takes on an air of a can't miss investing opportunity,
and I worry as I listen to these kinds of conversations,
as I read a lot of the coverage about AI
and chat, GPT, et cetera, that that hype is quickly
revving up in the AI space. What that means is
that a lot of people, and those people are already
invested in AI. The early investors, the people with their
(40:43):
chips already on the table, are going to make a
fortune on it. But what's going to happen is this
going to attract a lot more investment later on, and
those people, inevitably a lot of those people are going
to get in late in the game, just before sort
of all the hype pops and the prices for AI
as it's rest in financial markets come down, and people
will lose a lot of money. I mean, we've seen
(41:03):
this just in the last couple of decades. We've seen
this with Internet, with crypto, with almost everything. And I
worry that from that perspective, AI is going to cost
people more money that's going to make.
Speaker 1 (41:13):
Them Aaron and Nere, we're out of time. Thanks for
helping us sort out a complex topic, the collision of
bots with humans.
Speaker 2 (41:21):
Thank you very much, Miir Tiem.
Speaker 1 (41:23):
Thank you.
Speaker 3 (41:23):
Erin.
Speaker 1 (41:24):
You can find Aaron brown in near Kasar's columns on
the Bloomberg Opinion website, and you can follow Near on
Twitter at near Kasar and if you want to learn
more about AI chatbots, we did another episode with fellow
columnists parme Elsen and Tyler Cowen. You can find that
in the crash Course show feed here at crash Course.
We believe that collisions can be messy, impressive, challenging, surprising,
(41:49):
and always instructive. In today's crash Course, I learned that AI,
while it seems to have an advantage that's unbeatable, may
just like humans, loses it over time. What did you learn?
We'd love to hear from you. You can tweet at
the Bloomberg Opinion handle at Opinion or me at Tim
(42:09):
O'Brien using the hashtag Bloomberg crash Course. You can also
subscribe to our show wherever you're listening right now and
leave us a review. It helps people find the show.
This episode was produced by the Indispensable Animasarakus, moses On,
Dam and Me. Our supervising producer is Magnus Hendrickson, and
we had editing help from Stage Bauman, Katie Boyce, Jeff Grocott,
(42:33):
Mike Nizza, and Christine Banden Bilart. Blake Maples says our
sound engineering and our original theme song was composed by
Luis Gara. I'm Tim O'Brien. We'll be back next week
with another crash course.
Popular Podcasts
NFL Daily with Gregg Rosenthal
Gregg Rosenthal and a rotating crew of elite NFL Media co-hosts, including Patrick Claybon, Colleen Wolfe, Steve Wyche, Nick Shook and Jourdan Rodrigue of The Athletic get you caught up daily on all the NFL news and analysis you need to be smarter and funnier than your friends.
On Purpose with Jay Shetty
I’m Jay Shetty host of On Purpose the worlds #1 Mental Health podcast and I’m so grateful you found us. I started this podcast 5 years ago to invite you into conversations and workshops that are designed to help make you happier, healthier and more healed. I believe that when you (yes you) feel seen, heard and understood you’re able to deal with relationship struggles, work challenges and life’s ups and downs with more ease and grace. I interview experts, celebrities, thought leaders and athletes so that we can grow our mindset, build better habits and uncover a side of them we’ve never seen before. New episodes every Monday and Friday. Your support means the world to me and I don’t take it for granted — click the follow button and leave a review to help us spread the love with On Purpose. I can’t wait for you to listen to your first or 500th episode!
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations. Follow now to get the latest episodes of Dateline NBC completely free, or subscribe to Dateline Premium for ad-free listening and exclusive bonus content: DatelinePremium.com