Promise and Pressure: The road to year end

(DESCRIPTION)

Gold embossed text writes: J.P. Morgan. Ideas and Insights.

(SPEECH)

[MUSIC PLAYING]

(DESCRIPTION)

Kriti Gupta. GLOBAL INVESTMENT STRATEGIST J.P. MORGAN PRIVATE BANK

(SPEECH)

KRITI GUPTA: Welcome to our third quarter market update. My name is Kriti Gupta. I'm a global investment strategist here at JP Morgan Private Bank. I'm joined here today by Jake Manoukian. Our head of US investment strategy. This is a great time to check in because when we started this year, we came in with very different expectations. Our overall 2026 outlook was so different. We put out a new major outlook a couple of months ago. I just want to get your sense of just the sea change we've had from maybe where we started the year and where we are now.

JAKE MANOUKIAN:

(DESCRIPTION)

Jake Manoukian, U.S. HEAD OF INVESTMENT STRATEGY J.P. MORGAN PRIVATE BANK

(SPEECH)

Sure. I think that's a great place to start. Thanks for having me. Let's have some fun. Do a Q3 update. I think this is where we came into the year.

(DESCRIPTION)

Text: Outlook 2026. What we thought. Position for the A.I. Revolution. A.I. has delivered for investors. Now you must consider a phase where froth is building, and disruption could have consequences. Capture the upside of this transformation while avoiding the risks of over-exuberance. Own: Hyperscalers, Semiconductors, Power, Applications in Private Markets, Select China Tech. Avoid: Small Cap, Legacy Software as a Service. Think Fragmentation, Not Globalization. The global economy is being reconfigured by a fundamental shift that prioritizes reliability and security over efficiency. Investment Implications. Look for investment opportunities where security, energy, and supply-chains converge. Own: Gold, Global Industrials, Security, National Champions, Natural Resources, India. Avoid: Manufacturing Exporters, The Euro. Prepare for Inflation's Structural Shift. We expect a higher floor for inflation, with more volatility. It is once again a central variable for portfolio construction and an even greater risk to wealth. Plan with intent to maintain purchasing power and diversify into real assets. Own: Infrastructure, Hedge Funds, Hybrid Capital, EM Corporate Debt, Financials. Avoid: Cash, Long-Term Sovereign Debt

(SPEECH)

So I think markets have had a sea change. But the way I might characterize it is that markets have come around to our view of thinking about the world. We've always been focused on three key themes. The AI revolution, fragmentation not globalization, and inflation structural shift.

What we've seen throughout the year is that the AI trade has been defined by a complete mismatch between demand for AI products and the ability to supply those products. So

(DESCRIPTION)

A graph shows What we got: Demand for Al is outstripping supply. Token demand is rising much faster than computing supply... Growth since 2024 of token processing, data center capacity, and Al bottleneck capex. A graph. Sources: (LHS) EpochAl, Google internal disclosures. Data as of August 2026. Note: Al bottleneck capex. uses the constituents of the J.P. Morgan Al bottleneck basket. Outlooks and past performance are no guarantee of future results. It is not possible to invest directly in an index. Please refer to "Definition of Indices and Terms" for important information.

(SPEECH)

what I'm showing on this chart is that token output is up something like 330x over the last two years. Data center capacity is up 14x, which is historic in terms of an infrastructure build out but still lagging token output. And at the same time, the bottleneck companies that actually fueled the data centers have only increased their CapEx by less than 2x.

So that's the bottleneck that this entire trade is running through. And

(DESCRIPTION)

Therefore, competing capacity bottleneck have outperformed. A chart.

(SPEECH)

what markets have priced that as is a huge benefit to those bottleneck companies in the semiconductors that actually fill the data centers. We've seen material outperformance from both segments over the S&P 500.

(DESCRIPTION)

A chart. What we got: The conflict in Iran exacerbates global energy and defense fragility Defense inventories and energy supply chains are pinched. Patriot Missile inventories and vessel crossings through the Strait of Hormuz. Sources: (LHS) CSIS, Bloomberg Finance L.P. Data as of August 5, 2026. Outlooks and past performance are no guarantee of future results. It is not possible to invest directly in an index. Please refer to "Definition of Indices and Terms" for important information.

(SPEECH)

With fragmentation, you're right. We didn't foresee the war in Iran, but we knew that the structural underpinnings of this era of globalization and free trade were really changing. And the war in Iran has really exacerbated a lot of those stress points that we saw.

So you can see it in missile stockpiles being depleted. You can see it in disruption of flows through chokepoints like the Strait of Hormuz.

(DESCRIPTION)

therefore, companies that improve resilience have outperformed. YTD performance index (Jan 1, 2026 = 100)

(SPEECH)

And what that has resulted in within markets is that companies that benefit from this idea of sovereign security spending-- security in a holistic sense, not just traditional defense-- have outperformed the market. And then finally, with inflation, we've been dealing with an inflation overshoot for 5 and 1/2 years.

(DESCRIPTION)

A chart. What we got: The energy shock exacerbates an already sticky inflation dynamic. Inflation has overshot by 1140bps over the last 5 years... Fed funds rate, FOMC and our expectations for the Fed funds rate, percent. Sources: (LHS) Bloomberg Finance L.P. Fed funds rate data as of September 2026. PCE inflation data as of July 2026. Outlooks and past performance are no guarantee of future results. It is not possible to invest directly in an index. Please refer to "Definition of Indices and Terms" for important information.

(SPEECH)

That red area in the chart is the overshoot above the Fed's target.

Finally, we saw the Fed respond to this inflation overshoot by raising interest rates. And now a lot of the debate is how far will they go with this interest rate hiking cycle. But where the rubber really meets the road for us as investors is that real assets like infrastructure and less correlated strategies like hedge funds have outperformed aggregate bonds. So

(DESCRIPTION)

...therefore, bonds continue to underperform real assets. Total return, year-to-date

(SPEECH)

when we came into the year, a key focus for us was making sure that portfolios were insulated across a wide dispersion of risks.

And what we saw within client portfolios is that there wasn't enough protection against an inflationary environment, which is why we focused so much on things like hedge funds and infrastructure.

KRITI GUPTA:

(DESCRIPTION)

Gupta.

(SPEECH)

I think what's really fascinating about these three key topics is that when we did come into this year, I think our expectation-- and correct me if I'm wrong here, but our expectation was largely that, look, we've had a pretty good run, inflation is calming down, the economy is strong, it's resilient, the consumer is spending. But they might be starting to decelerate a little bit. So you still have positive growth but you don't necessarily have the enormous gains of 2022, 2023.

And I think maybe where the message has changed at that AI boom in particular, has become even more powerful than we expected. And that's not just showing up in the equity market. It's showing up in the bond market too.

JAKE MANOUKIAN:

(DESCRIPTION)

Manoukian.

(SPEECH)

Absolutely. And that's one of the reasons why bond yields have been the big story in markets over the last couple of weeks.

KRITI GUPTA:

(DESCRIPTION)

Gupta.

(SPEECH)

So take us underneath the hood of that AI trade in particular.

JAKE MANOUKIAN:

(DESCRIPTION)

Manoukian. Charts: AI, Fragmentation, and Inflation have all placed upward pressure on interest rates. The economy has been surprising to the upside, which is the latest pressure pushing up rates. Citi Economic Surprise Index, U.S. 10-year Treasury yield and future Fed policy rate pricing, percent. Sources: (LHS) Bloomberg Finance L.P. Data as of September 29, 2026. (RHS) Bloomberg Finance L.P. Data as of September 29, 2026. Note: Fed pricing is a 1 Y T Y. OIS forward swap yield which is good proxy of Fed funds rate 1-2 years from now.

(SPEECH)

So with the AI trade in particular, what you have is the hyperscalers. So the largest companies that are spending the most money on data centers have issued something like $370 billion worth of debt into the long end of the capital market-- into the long end of the fixed income markets. And what that's doing is it's creating a lot of supply. So just to size it, it's around half of what Treasury issuance is. So when you have a lot of supply coming to the bond market, the price is going to have to fall to reflect that. The price of the bond is the inverse of the yield. So the yields are going to rise.

What we've also seen within the bond market recently, and for the balance of the year, but especially recently is economic growth that surprised the upside. So the chart on the left here is the Citi Economic Surprise index. It hasn't been below 0 all year. That's the first year that I could find on record that it has not been below 0 for an entire year. And that's really remarkable because that's supposed to be a mean reverting index. Economists are supposed to change their forecasts for incoming economic data based on the prior trend, and they just haven't been able to raise their forecasts enough consistently.

When we look at the chart on the right, what we're really grappling with now is like, OK, we have AI that's putting upward pressure on interest rates. We have global fragmentation and oil price rises that are catalyzed by the war in Iran pushing up energy prices. And we have the Fed responding to the inflation overshoot, which is pressuring interest rates. And now the question we have to ask is, how far can this go? Because what we're seeing on the chart on the right is that the 10-year Treasury yield hasn't been this high since the mid 2000s.

And the market expectations for where the Fed might go in terms of a terminal interest rate are even higher than they were at the peak in 2022, when we were all grappling with 7% inflation.

KRITI GUPTA:

(DESCRIPTION)

Gupta.

(SPEECH)

And I love that you've outlined so many of those factors, because another key piece of this is also going to be the debt component. We're in a world-- we're not just in the United States, but internationally, people are really worried about these growing deficit concerns. And it feels like we have this conversation almost every autumn, frankly. But it feels like this time, the fact that it's coming, as you point out, so rightly, in tandem with this massive hyperscaler issuance.

From an investor standpoint, this is a very simplistic view. But if you have the opportunity to invest in a government bond of name your country for almost a similar yield, if not perhaps a little bit lower, than what a super safe, rare to market hyperscalers offering you, that feels like a very easy choice to make for a lot of global investors. And that is certainly part of the yield move. Which brings me back to the AI story, because one of the key concerns going into the next 12 months, and arguably the next couple of years, is going to be simply whether this massive AI-driven story can sustain.

JAKE MANOUKIAN:

(DESCRIPTION)

Manoukian.

(SPEECH)

Absolutely. And we've been tracking this concept of the AI wall of worry.

(DESCRIPTION)

What next? AI, Can AI continue to climb the wall of worry?

(SPEECH)

Ever since ChatGPT was released, the outperformance of AI-linked stocks has really been remarkable in that it's powered through a consistent wave of skepticism.

(DESCRIPTION)

The AI trade has powered through intense scrutiny. I A I Q index, May 2023 equals 100. A chart with headlines. ChatGPT hallucinates, some researchers worry it isn't fixable. Will AI be a bust? A Wall Street skeptic rings the alarm. Is the flurry of circular AI deals a win win or a sign of a bubble? Anthropic CEO says AI could wipe out half of all entry-level white-collar jobs. Anthropic pulls its most powerful AI models after US bars foreign access. House passes bill taking aim at data center electricity costs. AI could possibly end humanity, how are humans supposed to process that? tember 18, 2026. Note: IAIQ Index is the Indxx Artificial Intelligence and Big Data in of Artificial Intelligence technology in their products and services, as well as companies that produce hardware used in AI applied for the analysis of bit data.

(SPEECH)

And we've just called out a couple of examples of that skepticism. I remember the memes about LLMs not being able to count the number of R's in the word strawberry, for example. And now, today, we're worried that I could possibly end humanity.

So which one is it? Can it not even count the number of bars in strawberry? Or is it possible to end humanity? What we're focused on in investors is like, what does the ultimate ROI for the AI build out mean in terms of corporate profits, in terms of the token layer and the companies that actually sell the outputs of AI products, whether that's from the data centers or from the model companies or from the hyperscalers? And then what is the ultimate return for the application layer?

So everyone who's thinking about building products on top of the AI infrastructure that exists. What could that possibly be? So this question of can I continue to climb the wall of worry is critical to us. Because AI is going to be the primary driver of how your equity portfolio feels over the next five years, whether good or bad.

(DESCRIPTION)

A chart: #1: "Al infrastructure spending is unsustainable". Hyperscalers have exhausted their free cash flow... 12 month trailing free cash flows, US dollars in billions. Sources: (LHS) Michael Cembalest, EOTM, JPMAM, Bloomberg Finance L.P. Data as of August 4, 2026. Note: Estimates for Q3 2026 and use 12m blended forward. (RHS) Empirical Research Partners Analysis, Epoch Al, Pitchbook. Note: Aggregate gross cashflow based on aggregate of current analyst expectations for each company. Cumulative funding raised since inception based on cumulative funding to date for each private company aggregated; includes both debt and equity. Hyperscalers include Amazon, Microsoft, Alphabet, Meta, and Oracle. Equity raises included company expectations as well as reported estimates from Anthropic and OpenAl IPOs.

(SPEECH)

And you're right, that part of the concern coming into this year is all these hyperscalers are spending all of their free cash flow to buy all of the semiconductor chips that they can and build data centers.

And what you can see on the chart on the left is the massive transfer of free cash flow from the hyperscalers in red to the semiconductor companies in green. And one of the issues that the market was grappling with is, how are they going to keep spending? Their out of free cash flow. And the answer is that the capital markets are here to help. And we talked about the issuance that we've seen from hyperscalers.

(DESCRIPTION)

But capital markets will finance the free cash flow gap. Calendar year 2027.

(SPEECH)

What I'm showing on this chart in the green is their CapEx intentions for next year. The red is what they're expected to generate in free cash flow. So you can see that the CapEx is subsuming all the free cash flow. They've already started to raise debt. So that's the blue circle. But what we can get is continued equity raises from the AI labs. We can get dry powder from private credit and infrastructure funds. And if the hyperscalers change their debt profiles to be an average investment grade company, they could borrow an additional trillion dollars.

So we don't think we're at the financial limits of the build out. But it makes that question of where we are in terms of the j-curve of cost versus benefit all the more important.

KRITI GUPTA:

(DESCRIPTION)

Gupta.

(SPEECH)

And I think it's such an important moment in time because the capital markets, both in the equity space and the debt space, have not been open for the last couple of years to navigate some of the volatility post-pandemic and, arguably, even leading into the pandemic. So I think this is an especially crucial time to see that open up.

JAKE MANOUKIAN:

(DESCRIPTION)

Manoukian.

(SPEECH)

100%. And the opportunity that we're seeing as investors, it should be good for bank stocks, who are the kind of intermediary between these companies and the capital markets. But

(DESCRIPTION)

A chart. #2: "Data Center ROI will never materialize". Current spending trounces current revenue... 12 month trailing free cash flows, USS billions. Sources: (LHS) Bloomberg Finance L.P., BEA, Ramp, The Economist. Data as of June 2026.

(SPEECH)

the next question or the next worry that we get is, OK, we're spending $1 trillion on AI. We can point to maybe, I don't know, $200 billion of current revenues. This formula doesn't make sense. You don't have to be an accountant to understand that this isn't sustainable. So we understand that worry.

But

(DESCRIPTION)

...but reasonable GPU pricing supports attractive returns. Annual revenue of a 1GW data center by GPU hourly rates, USD bn

(SPEECH)

when you aggregate up the potential returns on a micro level for a data center, it starts to become compelling. So let's just walk through this chart. A one-gigawatt data center costs around $38 billion, $40 billion. That's what you have to put into the ground. The question then becomes, what do you have to sell your GPUs for to cover that and to cover the cost of capital? And what we're showing in these bars is that annual revenue generated by a certain GPU price per hour.

The Neocloud average, which is companies that are selling older GPUs is like $270 an hour. So that's probably close to break once you account for cost of capital. It's probably not making anybody a lot of money. Where we're focused is this $7 to $12 range, which is where reports say that frontier inference is being sold. So by frontier inference, I mean, one of the AI labs is renting computing space from one of the hyperscalers. So they're selling those GPUs from around $7 to $12 an hour.

And what you can see at that $7 per hour point is your revenues are meeting your capital costs in a year. So of course, have other costs to factor in. You have the cost of capital. So it's not a true break measure. But I think this helps to put into context the ROI calculation. Now two other numbers here frontier model API. That's what some providers think that some of the leading AI labs can generate of revenue with just one gigawatt. So some pretty stellar numbers of around $100 billion a year per gigawatt.

And then the final number is that $48 an hour. And the reason that's there is because if you're a white collar worker who makes $100,000 per year, you're making $48 an hour. So if we're thinking about some theoretical ceiling of where GPU hour prices can go, in theory, you'd be willing to pay up to $48 an hour if you could perfectly replicate a worker. So this is the work that we're doing to justify the amount of CapEx that's being put into the system.

KRITI GUPTA:

(DESCRIPTION)

Gupta.

(SPEECH)

And just put this into perspective in terms of maybe the narrative shift that's happened, over the last couple of months, in particular, because I think we started off this year and I want to say maybe even the summer by saying, well, the way we measure ROI is the minute a data center is constructed. If a data center takes 10 years to construct and become operable and have all the people making it kind of work, then that's the ROI time frame. But I feel like this summer we've heard very different things from the hyperscalers in terms of what that time frame is.

JAKE MANOUKIAN:

(DESCRIPTION)

Manoukian.

(SPEECH)

100%. And we'll get into that. If you don't mind, I'm going to save that because I want to talk about that on a different chart. The last kind of bear thesis is related to the timeline, which is like, OK, we say I should be productivity. We

(DESCRIPTION)

A chart. #3 "We should be seeing productivity gains". Productivity has not picked up so far... Real output per hour worked,% change 5-year annualized. Sources: (LHS) Bureau of Labor Statistics, Haver Analytics. Data as of June 30, 2026. (RHS) Solow, Robert M. (1987). "We'd Better Watch Out." Review of Manufacturing Matters: The Myth of the Post-Industrial Economy, by Stephen S. Cohen and John Zysman. The New York Times Book Review, July 12, 1987, page 36.

(SPEECH)

should be seeing productivity gains and we're not. So the chart on the left is just productivity growth, real output per hour. There's not much there. And if we're aggregating up all of the traditional economic data, yeah, there's not much evidence that AI is driving a lot of productivity growth within the economy today.

But on the other hand, there's a lot of historical examples of productivity growth taking some time.

(DESCRIPTION)

...but as Robert Solow said, "You can see the computer age everywhere but in the productivity statistics". We'd Better Watch Out.

(SPEECH)

So what this snippet is, it's not for you to read on the screen. I'm sure the text is very small. But this is an op-ed that an economist named Robert Solow wrote in 1987. And as an aside in this op-ed, he threw out a quip that's become quite well known, which is "you can see the computer age everywhere but in the productivity statistics." And this is 1987.

So if you're an equity investor and you're waiting for the productivity gains to show up, to be invested in a trend, you're probably going to wait too long. Because you missed 13 years of the computing and internet bull market that marked the 1990s. So we're having this debate internally. Eventually, will we need to see productivity gains? Absolutely. To justify the spending. But from an equity market investor have to look at more bleeding-edge data than just waiting for the real traditional economic data sources.

KRITI GUPTA:

(DESCRIPTION)

Gupta.

(SPEECH)

So, Jake, I'm curious about your perspective on this. It's been four years about since the launch of ChatGPT, which is really, I think, the kind of game changing moment, even though I feel like the development has been in the works for longer. Why do you think, despite the fact that we haven't seen more productivity gains, as you point out, are people still so excited about this technology? Kind of feels like a big leap of faith to make without real visibility into the future.

JAKE MANOUKIAN:

(DESCRIPTION)

Manoukian.

(SPEECH)

I think what people have seen is that it really transforms the way that you work and live. And I think there's an interesting dynamic where the users of AI-- there's a bell curve, where the people who are researching AI are getting worried about the capabilities that it can do. And the people who don't use AI are kind of like, what is the point of this chatbot? But I think there's a huge distribution in the middle where it's automating tasks, it's taking away no joy work. It's doing in minutes what would have taken hours in terms of automatable tasks.

And I think you're seeing more and more evidence of enterprises at a large scale in traditional sectors really find use cases for this technology that are going to result in cost savings and potential new markets for revenue. So there's clearly something there, and it's going to take time to see that in the productivity statistics. But at least for now, the market signals are telling us that this is valuable to somebody.

KRITI GUPTA:

(DESCRIPTION)

Gupta.

(SPEECH)

Walk us through maybe some of the other factors that are driving the trade. Because this AI story, arguably, is a very US-driven story, and it's something that we focused on in the public markets. But there's obviously a private market component to that as well.

JAKE MANOUKIAN:

(DESCRIPTION)

Manoukian. We are focused on investing in components, clouds and private markets. An image shows a cloud, chip, data center, and factory connected by glowing lines. Source: Illustration created using CharGPT.

(SPEECH)

Absolutely. I think what's interesting about the private markets-- and we recently saw another valuation markup for one of the leading AI labs today-- is the question of who ultimately derives the benefit. So I don't know if we're going to continue to spend trillions of dollars per year on data centers in perpetuity. We think we will in the near term, which is why we're excited about the companies that make some of these components.

What

(DESCRIPTION)

A chart. Chips, components, and clouds can all outperform. "And even though our demand is much greater than 70%, our supply allows us to confidently deliver 70%." Jensen Huang, Nvidia CEO. Al bottlenecks have superior fundamentals Sales growth. Sources: (LHS) FactSet. Data as of June 30, 2026. Note: For companies that don't report adjusted operating margin, operating margin was used. Based on J.P Morgan Al Bottlenecks basket (including companies like Marvell, Credo Technology, Coming, Eaton, and Sandisk). Data used was from most recently available quarter. (RHS) FactSet. Data as of August 28, 2026.

(SPEECH)

we've done here on the chart on the left is we've just shown those AI bottlenecks that are selling components into data centers in the green. And you can see whether on a rule of 40 or a rule of 80 frontier, which is kind of a traditional private equity type metric that ads sales growth and adjusted operating margin to just see how attractive you are as a company. 19 of the 21 that we track meet that rule of 40 threshold. So the fundamentals of these businesses are incredibly strong.

Now, the worry is that this spending can't last forever and their margin profiles can't hold forever. But we think they can hold through 2027. And they're trading at a valuation discount to the market right now, which is compelling to us. The quote I have here from one of the CEOs of the leading semiconductor company is that, "even though our demand is much greater than 70%, our supply allows us to confidently deliver 70%." And what that means to us is that demand is still well in excess of the ability to supply a lot of these components, which is where we want to be.

(DESCRIPTION)

Gupta.

(SPEECH)

KRITI GUPTA: And I think you made a really crucial point as well on just the repricing-- and you're going to hit this in a moment, but just the repricing of the discount that's offered at. One of the other key pieces that I think has changed since the start of the year, and even since the start of the summer, is this compression in multiples. So just walk us through that.

JAKE MANOUKIAN:

(DESCRIPTION)

Manoukian. A chart. Cloud revenue growth is accelerating Year on Year, percent.

(SPEECH)

Yeah. So it goes back to the rates conversation. If the 10-year Treasury rate is rising, that's becoming more and more competitive with earnings from the equity market. So if you can get over 5% in a risk-free asset, your equities have to reprice to compensate you to own them. So what we saw at the beginning of the year where the S&P 500 was trading at a 22 times forward multiple, that's compressed to around 19.

And what that 19 times forward price to earnings multiple means is that you're making a 5% yield on every dollar of earnings that the S&P 500 is generating, or a little bit more than 5%, which is about in line with the 10-year Treasury. Now, of course, earnings are growing over time. So you have to account for that within your multiple framework. But I think that balancing act is important for investors to consider.

KRITI GUPTA:

(DESCRIPTION)

Gupta.

(SPEECH)

And we're already seeing some of those flows drive the next, really, allocation conversations even with some of our clients.

JAKE MANOUKIAN:

(DESCRIPTION)

Manoukian.

(SPEECH)

Yeah, absolutely. I mean, for some clients, the opportunities that you're seeing in fixed income to immunize your portfolio and just lock in the yield is attractive. And for others who continue to believe in the capital appreciation potential of the equity market, it's an opportunity to buy the S&P at a valuation that we haven't seen in a couple of years.

KRITI GUPTA:

(DESCRIPTION)

Gupta.

(SPEECH)

So we've got a bond market with yields that are rising for a different set of issues. We've got AI growth that's growing. We have stock market that is thriving, but also kind of factoring in some of these other inputs. And on top of that, you have this kind of revolutionary technology that people are trying to make more cost effective. If I have that all right.

JAKE MANOUKIAN:

(DESCRIPTION)

Manoukian.

(SPEECH)

Yeah, absolutely. And I think another debate that we're having is like, can the hyperscalers continue? And are they seeing the signs that they need to continue? And the chart on the right shows that. Cloud revenue growth, which is the bread and butter business for these companies, is accelerating. So when they're sitting around in they're c-suite office thinking about where does the next dollar of capital go, they're going to look at the sales growth and say, hey, maybe this is working.

And that's what the Amazon CEO said after Q2 earnings. It's like, we did cloud 1.0. We're monetizing AI at a little bit faster pace than we did cloud 1.0. And I think that was really reassuring to investors.

KRITI GUPTA:

(DESCRIPTION)

Gupta.

(SPEECH)

And I think, again, and we hit this earlier, but the capital markets component is such a crucial one.

JAKE MANOUKIAN:

(DESCRIPTION)

Manoukian.

(SPEECH)

Yes.

KRITI GUPTA:

(DESCRIPTION)

Gupta.

(SPEECH)

I mean, we've seen this help a lot of the hyperscalers already listed. But

(DESCRIPTION)

A chart. We can not ignore private markets In the internet/cloud cycle, value accrued to public markets in earlier phases while privates play a larger role in the application layer. Estimated cumulative value creation, $ trillions, 1995-2020. Sources: (LHS) J.P. Morgan Private Bank, Bloomberg Finance L.P., Company reports, September 2025. Value creation indicative of value accrued to companies. Physical infrastructure refers to foundational physical assets (e.., major telecom players), digital infrastructure refers to core hardware/networking (e.g., semiconductors, servers, routers), platform technology refers to enabling software/services (e.g., search, operating systems, cloud), and applications refers to end-user products/services (e.g., e-commerce, social media, streaming). Value creation estimated by public companies' peak market capitalizations of key companies in each stage of the internet / cloud cycle. The split between public/private derived by a top-down estimate informed by private companies value creation estimated from valuations for key companies, where available, and blended forecasts from historical reports. Examples of companies included: physical infrastructure (AT&T, Level 3 Communications), digital infrastructure (Cisco, Lucent Technologies), platform tech (Google, Oracle, Microsoft), application (Apple, Facebook, Netflix). Chart meant to be illustrative of value creation across phases, not exhaustive. (RHS) J.P. Morgan Private Bank, Bloomberg Finance L.P., Company reports, September 2025. Value creation indicative of value accrued to companies. Physical infrastructure refers to foundational physical assets (e.g., power, data centers), digital infrastructure refers to core hardware (e.g., semiconductors), platform technology refers to enabling software/services (e.g., LLM, APls), and applications refers to end-user products/services (e.., Al copilots, Al-enabled software). Physical infrastructure based on change in market capitalization of global utilities and key datacenter REITs since launch of ChatGPT in November 2022. Digital infrastructure based on change in market capitalization of global semiconductors and key hyperscalers (Amazon, Meta, Alphabet, Microsoft) since launch of ChatGPT in November 2022. The split between public/private derived by top-down estimate informed by private companies' value creation, where data is available. Platform technology incorporates latest reported valuations of key companies such as OpenAl, Anthropic, and xAl. Application layer incorporates latest reported valuations of key companies such as ByteDance, Revolut, Databricks. Chart meant to be illustrative of value creation across phases, not exhaustive.

(SPEECH)

you don't actually need the capital markets to be open and thriving maybe the way they are in this moment in time to actually grow some of this technology.

JAKE MANOUKIAN:

(DESCRIPTION)

Manoukian.

(SPEECH)

Yeah, I think the distribution of who wins and who loses if capital markets close is an interesting question to ask, because there might not be the huge spending surge on actual data center capacity. But if AI is already really valuable, in part because we're in a state of scarcity, like slowing down the build out might actually not be bad for everybody if there's only so many GPUs that can achieve the work that a lot of end users are finding valuable.

(DESCRIPTION)

Gupta.

(SPEECH)

KRITI GUPTA: So walk us through what's on what's on the screen right now.

JAKE MANOUKIAN:

(DESCRIPTION)

Manoukian.

(SPEECH)

So you asked about private markets.

KRITI GUPTA: I did.

JAKE MANOUKIAN: This is tech 1.0. And it's the share of value that was created in both public and private. In tech 1.0. And the point I want to make here is that the application layer is the largest bar. So regardless of public and private. And what we mean by application layer is like, what companies are valuable because the internet and computers exist? So you can see the iconic names here.

(DESCRIPTION)

In the Al cycle, value has accrued to the infrastructure players, but we believe we are entering the platform tech/application phases. Estimated cumulative value creation, dollars in trillions

(SPEECH)

What's interesting is that in the AI phase, where value has accrued so far as digital infrastructure. So it's the companies that are building the data centers and putting the components in data centers.

We're starting to see the platform technologies start to gain share. But if you think that the experience from the computer and internet phase is a good analogy, then we should keep looking towards that kind of platform technology and application layer for the next big idea. And the concept which companies are going to exist because of AI that are the next trillion dollar company, and they might not even exist yet.

We've put some examples here of what the growth equity and venture capital managers that we focus on are kind of looking at. But the whole idea here is that this innovation is providing a foundation from which more innovation and more economic value creation can flourish.

(DESCRIPTION)

Gupta.

(SPEECH)

KRITI GUPTA: This is a really fascinating concept. I feel like again-- and we've talked about this in the context of the rates conversation-- we focus so much on the US story, the idea that you have to be a hyperscaler to really succeed in this market. And this is a great example of layers of-- I mean, not to repeat the word, but elements of that I trade and the AI ecosystem that exists in other elements in growing parts of the way our financial system works. I think that's a really interesting one.

I think the moment in time that this conversation is happening is crucial, because there's also a question of not just making things easier in a bigger part of our lives, but things like cybersecurity, things like sovereign technology, sovereign AI. So much of this innovation is happening in the United States. Is that something that is as exportable as it was 20 years ago?

JAKE MANOUKIAN:

(DESCRIPTION)

Manoukian.

(SPEECH)

It's such a good point. And we're doing a lot of work on this as we look ahead to the 2027 outlook. And it leads into the next section, which is this idea of global fragmentation. AI is going to fragment along these sovereign lines. And if you think about a country running its own sovereign AI system, they're not going to want to run that compute in a data center that's not in their own jurisdiction. What's really interesting is that open models can be portable across borders, but the actual physical elements that create AI technology cannot.

So if you think about the AI ecosystem might have to be duplicative because of this idea of global fragmentation, it adds a really interesting element to the discourse around how much AI infrastructure do we need to build.

(DESCRIPTION)

What next? Can Al continue to climb the wall of worry? Global Fragmentation. Will the midterms mark a structural shift in geopolitics?

(SPEECH)

I think that other near-term catalysts that we have to watch is midterms. So what's really going to change after midterms? Markets are pricing in a wave election. I'm not going to spend too much time making the case for or against.

But if there is a wave election, is anything going to change? And we would argue that, not as much as you might think, at least from an investor's point of view. We talked about the war in Iran.

(DESCRIPTION)

A chart. The conflict seems to be expanding, and is the key near term risk to markets Crude supply has collapsed, and the Red Sea is now a risk. 7-day rolling average number of vessels through the straits. Source: (Both) Bloomberg Finance L.P. Data as of September 18, 2026.

(SPEECH)

That's still a key near-term risk to markets that might start to move after the midterms. But I think the salient point for investors here is watching the flow of goods through the Strait of Hormuz.

Now, the interesting thing this is just the ships that are tracked from public sources. There's alternative data out there that says that oil flows are actually pretty close to pre-war levels, which is really interesting. So we're monitoring that divergence within the data. But it's clear that there's disruption within the energy space. And

(DESCRIPTION)

which adds stress to crude oil and refined product prices

(SPEECH)

when you look at crude oil prices and you look at diesel prices, this is becoming a friction point for economic output in the US and the world.

So this is a risk that we're monitoring. It's a risk that we're watching. We don't think it will derail the cycle. But for at least the next three months, the war is going to be a key driver of asset prices both in equities and fixed income.

KRITI GUPTA:

(DESCRIPTION)

Gupta.

(SPEECH)

And I think this is a really crucial point to make here because, again, when we started this year, we did not expect this particular conflict to flare up. We certainly didn't expect this kind of disruption to flare up. This is in line with the rolling shocks theme that we've been discussing for the last 12 months that I think had a lot of experience with really post pandemic. And I think especially when it comes to energy shocks, that's such a crucial piece of things like the AI build out, things like the fragmentation theme. You need energy to be affordable for a lot of this expansion.

The fact that it's not is something, to your point, is being reflected in bond yields. But I think what is so striking to me is that even with this disruption, we're still looking at a resilient economy. We're still looking at a CapEx build out that requires an energy input, and an AI trade that is still very much on track to continue to expand. Is that fair?

JAKE MANOUKIAN:

(DESCRIPTION)

Manoukian.

(SPEECH)

Absolutely fair. And the energy input is critical. This idea that the whole world is going to have to be focused on choke points now in a way that maybe they weren't ready to be even after the war in Ukraine. Which also catalyzed NATO to raise their spending targets on infrastructure and defense.

(DESCRIPTION)

National defense and next generation security spending is accelerating globally. European defense spending targets have been revised higher. Defense spending as a % of GDP. Sources: (LHS) NATO. Data as of June 2024 and June 2025. (RHS) PBS, Reuters. Data as of August 7, 2026. But now the Gulf and Asia are driving defense spending. Headlines. Japan's Cabinet approves record defense budget aiming to deter China as tensions grow. Saudi Arabia, Turkey, Pakistan pledge mutual defence as Middle East turmoil escalates

(SPEECH)

But that trend is going global. When you look at Japan, when you look at other Middle Eastern countries locking arms around sovereign defense, we think that this source of demand and this source of spending is here to stay.

Then when you think about where we're going next in terms of AI, we've developed a lot of industrial policy that incentivizes semiconductor production within the US. But this is increasingly going to go to the model layers themselves. So

(DESCRIPTION)

A chart. Artificial Intelligence will be a target for sovereign intervention Industrial policy has focused on critical inputs, but now could focus on the Al models themselves. Sources: (LHS) FRED. Capacity data as of January 2025. Production data as of August 2026. (RHS) Artificial Analysis. Data as of September 18, 2026.

(SPEECH)

the idea of the government intervening within the AI ecosystem, we just had a meeting of the president with the AI labs yesterday about regulating and intervening in some way, shape, or form with these models is going to be all the more critical.

Because on the one hand, you're going to have the political backlash. And on the other hand, you're going to have the kind of wing that says, the US needs to win AI to maintain its global relevance from both a financial markets perspective and a influence and defense perspective. What it means for us is that this idea of national champion companies across the four Cs, compute, conflict, climate and choke points, are crucial. And you can see that companies that traffic in these areas have grown their earnings per share at almost 1,000% from January 2024, and their prices haven't kept up with it.

(DESCRIPTION)

A chart. Focus on National Champions across four C's: compute, conflict, climate, and chokepoints. Earning expectations validate outperformance for strategic companies. 10 strategic companies (MU; GFS; INTC; GD; TXN; MP; UUUU; LEU; SGML; HII).

(SPEECH)

It's a key area of focus for us and a key theme that we're monitoring within markets.

(DESCRIPTION)

Icons appear. A chip, shield, ship and tornado. Compute, Sovereign Al strategies, electrification. Conflict, Defense systems & technology Chokepoints, Energy, natural resources, agriculture. Climate, Renewable power, energy storage, adaptation.. Gupta.

(SPEECH)

KRITI GUPTA: I love that you--

JAKE MANOUKIAN: Picked those icons?

KRITI GUPTA: Picked those icons, first of all. We're visual learners here. But also I think when we think about national champions and, specifically, the defense conversation, I get this question a lot. Are we talking about a build out of military infrastructure? And in the United States, it's a very normal conversation to have. We have the largest defense spend in the world. But the fact that you are seeing other countries, countries like Germany, countries in Europe, countries in Japan, as you pointed out, rightly, leaning into that defense conversation is not something we've seen in decades.

But that also security and national champions doesn't just mean physical military infrastructure anymore. It does mean energy security, and it does mean things like climate control. And even just the fact that I think Japan was talking about their building up their Naval capacity-- again, something we have not seen in decades-- at a time when fiscal budgets are already squeezed is fascinating. So I think the expanding definition of what the AI trade is, the expanding definition of what defense is such a crucial one.

(DESCRIPTION)

Manoukian.

(SPEECH)

JAKE MANOUKIAN: And you're right. And then to bring it back to midterm. Well, I'll react to demand a little bit. Big

(DESCRIPTION)

A Chart. The key political risk for investors is the rise of populist policies. Both Democrats and Republicans have embraced populist policies. Sources: (LHS) FRED. Capacity data as of January 2025. Production data as of January 2026.

(SPEECH)

picture, what that means is there's more demand in the economy because you have spenders that have other constraints, whether it's chasing AGI or whether it's security. And there's a limited pool of resources. So what that means is that the cost of capital is likely to rise, which is one of the reasons we've seen the sell off that we've seen in 10-year bonds.

And then, I mean, it also ties into the risk we're watching around midterms, which is populism. And I think this chart on the left is pretty striking, just because it shows that Democrats and Republicans agree on a lot of policies that I think 10 years ago would have been viewed as populist. I mean, just think about capping credit card interest rates at 10%. A lot of agreement there. So that's what we're watching in the midterms is like how well do populist candidates do.

And the reason it matters is because when economic studies have shown that in the years after populists take power in a country, you get rising debt to GDP and you get more inflation. Which have obviously posed challenges to portfolios over the course of this decade. And those could become exacerbated if more populist policies do come into being.

KRITI GUPTA:

(DESCRIPTION)

Charts. The risk of raising government debt and inflation. Gupta.

(SPEECH)

I mean, look, affordability was the key voting issue in the last presidential election, is the key voting issue of midterms. And

(DESCRIPTION)

Inflation. Will inflation move back towards target, or will the target move towards inflation?

(SPEECH)

I think, again, it's a question that we get a lot here in the States, in particular, where the wealth effect is such a crucial part of the investor conversation. And the fact that we are at levels of exposure just from the retail crowd and household wealth that is directly tied to the stock market, which is directly tied to the AI trade, I think is a really interesting connection that we haven't seen-- I want to say since like the 1970s when Americans had this much exposure.

That doesn't even account for some of the things that we're seeing on the international space. So affordability is such a key issue. But I don't think that's one that's going to go away after the midterms, especially if we're talking about the rising cost of capital.

JAKE MANOUKIAN:

(DESCRIPTION)

Manoukian.

(SPEECH)

Absolutely. And that brings us to the final section here, which is inflation. You're absolutely right to point out that the cost of living is the most salient political issue. And the Fed, for their part, has finally noticed this. And listen, they have a dual mandate. They have a really hard job. But they were dealing with inflation overshoot over the last 5 and 1/2, coming on 6 years.

(DESCRIPTION)

Charts. We expect a higher floor for inflation, with more interest rate volatility Inflation has been well above Fed target, but slightly higher rates likely won't solve the inflation problem. Sources: (LHS) Bloomberg Finance L.P. Data as of July 2026. (RHS) Bloomberg Finance L.P. Data as of September 18, 2026.

(SPEECH)

They've raised interest rates once. The market goes back and forth between pricing 4 more and 3 more interest rate hikes.

And the question that we have to ask ourselves as investors is, what is the cost of bringing inflation back to target? Because if you are serious about bringing inflation back to target, it probably means that you need to prick some of the enthusiasm around AI spending and the stock market along the way. Because the wealth effect and the AI investment boom is a large part of what's driving the growth within the economy that then leads to inflation.

So it's our view that they're going to get some help from an alleviation of the conflict of Iran. They're going to get some help from an underlying trend of the economy, which looks stable to us, but not necessarily overheating. And they're probably only going to have to hike another 2 times for them to feel better about the inflation environment. And that should lead to continued positive performance in risk assets. But that's a risk to our view is that the Fed decides, hey, we actually have to take this seriously and we need to hike rates farther than the market thinks so that we can get inflation down to actual 2%.

KRITI GUPTA:

(DESCRIPTION)

Gupta.

(SPEECH)

I think what's so fascinating about that chart in particular that you're talking about is when it comes to the Federal Reserve's approach to inflation and what's driving economic growth broadly, again, another sea change just in the last couple of months has been a new Fed Chair. And I think that's really crucial because, suddenly, we are now talking about AI at the Federal Reserve in a way that we hadn't previously and we hadn't even accounted for.

And I want to say, in the last couple of years, there have been multiple press conferences where AI has come up and has been approached as a completely different issue relative to economic growth and to the labor markets. Here we are talking about, as you said, can a chatbot replace a human? And suddenly, that's now the conversation at the Federal Reserve. And I think that's a really fascinating one, in addition to the CapEx build out. And that's something that's really only come up, I want to say, in the last couple of months.

JAKE MANOUKIAN:

(DESCRIPTION)

Manoukian.

(SPEECH)

Yeah. I mean, the macroeconomic question is fascinating. AI build out certainly seems inflationary today. It's keeping the stock market elevated. You're competing for scarce semiconductors and cooling equipment and electrical equipment and skilled labor. But in the future, you might get this productivity benefit where you actually raise the potential output of the entire economy, which should alleviate some of the inflationary concerns. But that's the part that no one can see yet.

(DESCRIPTION)

Gupta.

(SPEECH)

KRITI GUPTA: And it's seeping in on the government level. It's also seeping into the regular diaspora. If you see mainstream media now talking about AI every day, I don't think we could have said that 12 months ago. It would have been maybe deemed a little too sophisticated for the masses. And that's so not the story today.

JAKE MANOUKIAN:

(DESCRIPTION)

Manoukian.

(SPEECH)

100%. And it also relates to this question, which is like, how do you deal with the debt pile? How do you deal with rising government debt to GDP? Now, the CBO comes out every year with a report where they list all the ways you can do it.

(DESCRIPTION)

Charts and reports. Political drivers increase inflation risk. The U.S. debt path is on a dangerous trajectory, and the legislative options are unpalatable. Sources: (LHS) Haver Analytics, Congressional Budget Office. Data as of February 2026. (RHS) CBO. Data as of December 2024.

(SPEECH)

The problem is, none of them are politically palatable. Because what they sound like is, we're going to raise taxes, we're going to cut spending, we're going to cut entitlements. So it's hard to build the political alignment and the political capital to actually enact that.

Now, what the AI proponents would say is like, guess what? We're going to grow our way out of it because AI is going to lead to this era of really booming economic growth without a lot of inflation. Maybe. I don't think we can count on that. I think the risk that we're watching to investors is that the US learns some lessons from Japan. And what happened in Japan is the debt to GDP ratio peaked five years ago, and it's declined. And the question we have to ask ourselves is like, OK, what was Japan up to?

Were they making solid fiscal decisions to consolidate the budget? No. They enacted yield curve control and they ran inflationary policies. And the outlet for that has been the yen, which has seen continued weakness over the last several decades. And is currently trading at some of the lowest levels in 30 years.

KRITI GUPTA:

(DESCRIPTION)

Charts. The U.S. may learn a lesson from Japan's policy path. Japan drove nominal growth to reduce their debt load, but the Yen paid the price. Sources: (LHS) Haver Analytics, World Bank, International Monetary Fund. Data as of March 31, 2026. (RHS) Bloomberg Finance L.P. Data as of August 31, 2026.

(SPEECH)

I mean, this is maybe the scariest part for Americans and for some of our clients who say, is this the only way to do it? Especially because if you go to Japan, I mean, that yen piece is so important. Their purchasing power has completely dropped and declined. I mean, housing is a problem in Japan. Demographics are a problem in Japan, especially in light of the conflict in the Middle East and rising oil prices. Japan is a major energy exporter.

There frankly-- an importer, excuse me-- an island, and they have those additional costs anyway. So I think this is a fascinating one. But to your point, they don't have the same innovation in their economy that we do here in the United States.

JAKE MANOUKIAN:

(DESCRIPTION)

Manoukian.

(SPEECH)

Or energy independence or status as the world's Reserve currency. So there's a lot of things balancing it out. It's just that I think the lesson here is, OK, one of the ways to deal with the debt is drive a little bit of inflation and keep interest rates a little bit lower than you normally would. And that brings us to the implication.

KRITI GUPTA:

(DESCRIPTION)

Plan with intent to protect against the risks that are hard to see. Dots appear on an illustration of a plane. Gupta.

(SPEECH)

You're going to have to explain this one. Walk us through this picture.

JAKE MANOUKIAN: So this is a famous example from World War II, which is the Allies would do bombing runs over Germany and they would fly the planes back to England, and they would map where the Planes were getting shot. And what you can see is you have a lot of bullet holes in the tip of the wings and in the tip of the tail. So what do you think the engineers did with this information?

KRITI GUPTA: They probably fortified it.

JAKE MANOUKIAN: They fortified the spots where the planes were getting shot. Seems logical, right? But that was a huge mistake. And the reason it's a huge mistake is because they were measuring the planes that made it back to England. So every plane that actually went down was getting shot in the blank spots, in the dark spots. And the lesson for investors, or I think the salience for investors is that we spend so much time talking about AROI, hyperscale CapEx plans.

When is the Strait of Hormuz going to open? What's the impact of the midterm elections? What's going on with interest rates? How high is the Fed going to hike? And I think what the black spots on the plane are for many of our clients is, what do I do in a world where inflation is going to overshoot and my portfolio is built for a world where inflation was contained? Because when inflation is overshooting, your purchasing power is eroding faster, your fixed income allocations are more volatile, and your fixed income and your stocks are more highly correlated. So you're more likely to have episodes where stocks are selling off at the same time as bonds.

(DESCRIPTION)

A chart. Consider diversifying into real assets to hedge against portfolio level drawdowns Investors should consider commodities, real assets, and alternative strategies to hedge inflation risk. Volatility. Source: J.P. Morgan Asset Management Long-Term Capital Market Assumptions. Data as of September 30, 2025.

(SPEECH)

So the question is, what do we do about it? And I love this constellation chart because it breaks down portfolio resilience in a very clear way. Where we started at the beginning was AI. AI to us, it's becoming a credit theme and there's interesting ways to play it in the credit space. But it's going to be the determinant of capital appreciation within equity markets, which is the blue circle at the top of this chart, which means we're going to endure some volatility. We're not going to have much correlation to inflation, but we're going to drive capital appreciation in the portfolio.

With core fixed income, we're actually more excited about core fixed income today than we have been in quite some time, just given the move in interest rates. And for a US taxpayer, when you can get above 5% tax free in a municipal bond, it's a pretty compelling trade. So we feel good about investing and earning that carry over time. From a portfolio construction perspective, we want to own fixed income when we're in a recession, when inflation is falling, when the Federal Reserve is cutting interest rates and bonds are appreciating and protecting against equity volatility that's happening.

What a lot of folks might be missing if they built their portfolio for a world where inflation was below Fed target is everything in the green circle on the right. Which is assets that have a lower volatility than equity, similar volatility to core fixed income, but a positive correlation to inflation. So what's going to work in a world where equities are a little bit volatile

(DESCRIPTION)

Gupta.

(SPEECH)

and inflation is still sticky? So that's where we come back to things like core real estate infrastructure, pockets of leveraged loans and diversified hedge funds to help balance out that portfolio exposure.

KRITI GUPTA: This is a really fascinating one, because it also feels like some of these key investments are also taking on a new life of their own. I don't think we could have said or even characterize maybe some of the infrastructure the way we do now five years ago. It's a completely changing dynamic. And I think this is really interesting because some of these assets play multiple roles in a portfolio. So even though they are on these various spectrums, you kind of need a little bit of everything.

JAKE MANOUKIAN:

(DESCRIPTION)

Manoukian.

(SPEECH)

Absolutely. And if you only have the blue or the red, you're not complete. And I think that's the most prudent way that we think investors can design the portfolio based on the outlook that we have.

KRITI GUPTA:

(DESCRIPTION)

Gupta.

(SPEECH)

So Jake, we've covered a lot in a very short amount of time, it feels like. Are there any key messages you want to wrap with?

JAKE MANOUKIAN:

(DESCRIPTION)

Manoukian.

(SPEECH)

Yeah, I'll just wrap with what's shown on this slide. Can AI continue to climb the wall of worry? We think the answer is yes. We think it's the most important global theme for equity markets. Is anything really going to change with the theme of global fragmentation after the midterms? We don't think so. And we think security spending and government intervention will endure. And then finally, with inflation, we just think we're in this more rangebound and volatile world for inflation and bond yields. It doesn't mean there's not opportunities.

Again, as I said, we're getting paid for a lot of the risks that are in the system based on the latest move in bond yields. But making sure you have a full suite of complements within a portfolio to protect against the disparate range of outcomes that might come to fruition is critical.

KRITI GUPTA:

(DESCRIPTION)

Gupta.

(SPEECH)

I think this one's so fascinating here because you put these into three buckets, and that's really how we started this year. These are three key themes to watch. And I think going into 2027, we are seeing the blurring of the lines. We seeing the intersections, which I think you've laid out beautifully. Thank you so much for joining.

JAKE MANOUKIAN: It was fun. Thanks for having me.

KRITI GUPTA: There's a lot to go through. We have so much more to share. Reach out to the JP Morgan Private Bank team. We have plenty more analysis to help navigate the uncertainty. Thank you for joining.

(DESCRIPTION)

IMPORTANT INFORMATION. Key risks. Alternative investments. Investing in alternative assets involves higher risks than traditional investments, including, without limitation, limited liquidity and valuation risk, and is suitable only for investors with sufficient knowledge and sophistication to evaluate the merits and risks of such investments. Alternative investments should not be deemed a complete investment program and distributions are not guaranteed. They may not be tax efficient, and an investor should consult with their tax professional prior to investing. Alternative investments often have higher fees than traditional investments and they may also be highly leveraged and engage in speculative investment techniques, which can magnify the investment loss or gain--including risk of loss of the entire investment. For comprehensive details around unique set of risks for specific alternative investments, please consult the offering memorandum. Bonds. Bonds are subject to interest rate risk, credit and default risk of the issuer. Bond prices generally fall when interest rates rise.

Commodities. Investments in commodities may have greater volatility than investments in traditional securities, particularly if the instruments involve leverage. The value of commodity-linked derivative instruments may be affected by changes in overall market movements, commodity index volatility, changes in interest rates, or factors affecting a particular industry or commodity, such as drought, floods, weather, livestock disease, embargoes, tariffs and international economic, political and regulatory developments. Use of leveraged commodity linked derivatives creates an opportunity for increased return but, at the same time, creates the possibility for greater loss. Cryptocurrency. Investments in digital assets and cryptocurrencies, or other investment vehicles holding or referencing digital assets, are highly speculative and involve significant risks, including, but not limited to, increased volatility, ongoing regulatory uncertainty including lack of clear precedent in various jurisdictions surrounding digital assets, cybersecurity risk relating to digital asset holdings, and increased sensitivity to news, speculation and manipulation. Likewise, various digital assets may differ from one another in their technological characteristics, regulatory treatment, market convention and performance. There is limited data on the performance of digital assets and products referencing digital assets (although in no case should historical performance be taken as an indication of future performance). You should consider these unique characteristics and whether these assets are suitable for you when making an investment decision. Diversification. Diversification and asset allocation does not ensure a profit or protect against loss. Emerging markets. Investing in emerging markets involves a greater degree of risk and increased volatility compared to developed markets. Changes in currency exchange rates and differences in accounting and taxation policies outside the investor's jurisdiction can raise or lower returns. Some markets may not be as politically and economically stable, in addition to differences in taxation policies, and legal systems outside the investor's jurisdiction may create additional risks. Investors should carefully consider these risks and consult with financial and legal advisors before investing in emerging markets.

Equities. The price of equity securities may rise or fall due to the changes in the broad market or changes in a company's financial condition, sometimes rapidly or unpredictably. Share values can rise with strong earnings or positive market expectations, but they can also fall due to weak earnings or negative sentiment, and dividends are not guaranteed. Fixed income. Investing in fixed income products (such as bonds) is subject to certain risks, including, but not limited to, interest rate, credit, inflation, call, default, prepayment and reinvestment risk. Any fixed income security sold or redeemed prior to maturity may be subject to substantial gain or loss. High yield bonds. High yield bonds (rated at or below BB+/ba1 or unrated) are speculative, non-investment grade securities with increased risk of default and loss. These investments are suitable only for investors able to bear higher risk. International investments, International investments may not be suitable for all investors. International investing involves a greater degree of risk and increased volatility. Some international markets may not be politically or economically stable. Foreign holdings are subject to currency risk, as fluctuations in exchange rates between the investment's foreign currency and the holder's domestic currency can affect the value of the investment.

Municipal bonds. Investors should understand the potential tax liabilities surrounding a municipal bond purchase. Certain municipal bonds are federally taxed if the holder is subject to alternative minimum tax. Capital gains, if any, are federally taxable. The investor should note that the income from tax-free municipal bond funds may be subject to state and local taxation and the altemative minimum tax (AMT). Preferreds. Preferred securities share characteristics of both stocks and bonds. Preferred securities are typically long dated securities with call protection that fall in between debt and equity in the capital structure. Preferred securities carry various risks and considerations which include: concentration risk; interest rate risk; lower credit ratings than individual bonds; a lower claim to assets than a firm's individual bonds; higher yields due to these risk characteristics; and "callable" implications meaning the issuing company may redeem the stock at a certain price after a certain date. Small capitalization equities. Investments in smaller capitalization companies often involve significantly greater risks than the securities of larger, better-known companies because they may lack the management expertise, financial resources, product diversification and competitive strengths of larger companies. The prices of the securities of smaller companies may be subject to more abrupt or erratic market movements than larger, more established companies, as these securities typically are traded in lower volume and the issuers typically are more subject to changes in earnings and prospects. In addition to liquidity risks, when selling large positions in small capitalization securities, the seller may have to sell holdings at discounts from quoted prices or may have to make a series of small sales over a period of time, and dividends are not guaranteed.

This material is for information purposes only, and may inform you of certain products and services offered by private banking businesses, part of JPMorgan Chase & Co. ("JPM"). Products and services described, as well as associated fees, charges and interest rates, are subject to change in accordance with the applicable account agreements and may differ among geographic locations. Not all products and services are offered at all locations. If you are a person with a disability and need additional support accessing this material, please contact your J.P. Morgan team or email us at accessibility-support@ipmorgan.com for assistance. Please read all Important Information. GENERAL RISKS & CONSIDERATIONS. Any views, strategies or products discussed in this material may not be appropriate for all individuals and are subject to risks. Investors may get back less than they invested, and past performance is not a reliable indicator of future results. Asset allocation/diversification does not guarantee a profit or protect against loss. Nothing in this material should be relied upon in isolation for the purpose of making an investment decision. You are urged to consider carefully whether the services, products, asset classes (e.g. equities, fixed income, alternative investments, commodities, etc.) or strategies discussed are suitable to your needs. You must also consider the objectives, risks, charges, and expenses associated with an investment service, product or strategy prior to making an investment decision. For this and more complete information, including discussion of your goals/situation, contact your J.P. Morgan team. NON-RELIANCE. Certain information contained in this material is believed to be reliable; however, JPM does not represent or warrant its accuracy, reliability or completeness, or accept any liability for any loss or damage (whether direct or indirect) arising out of the use of all or any part of this material, No representation or warranty should be made with regard to any computations, graphs, tables, diagrams or commentary in this material, which are provided for illustration/ reference purposes only. The views, opinions, estimates and strategies expressed in this material constitute our judgment based on current market conditions and are subject to change without notice. JPM assumes no duty to update any information in this material in the event that such information changes. Views, opinions, estimates and strategies expressed herein may differ from those expressed by other areas of JPM, views expressed for other purposes or in other contexts, and this material should not be regarded as a research report.

Any projected results and risks are based solely on hypothetical examples cited, and actual results and risks will vary depending on specific circumstances. Forward-looking statements should not be considered as guarantees or predictions of future events. Nothing in this document shall be construed as giving rise to any duty of care owed to, or advisory relationship with, you or any third party. Nothing in this document shall be regarded as an offer, solicitation, recommendation or advice (whether financial, accounting, legal, tax or other) given by J.P. Morgan and/or its officers or employees, irrespective of whether or not such communication was given at your request. J.P. Morgan and its affiliates and employees do not provide tax, legal or accounting advice. You should consult your own tax, legal and accounting advisors before engaging in any financial transactions. In the United States, bank deposit accounts and related services, such as checking, savings and bank lending, are offered by JPMorgan Chase Bank, N.A. Member FDIC. LEGAL ENTITY, BRAND & REGULATORY INFORMATION In the United States, bank deposit accounts and related services, such as checking, savings and bank lending, are offered by JPMorgan Chase Bank, N.A. Member FDIC. JPMorgan Chase Bank, N.A. and its affiliates (collectively "JPMCB") offer investment products, which may include bank managed investment accounts and custody, as part of its trust and fiduciary services. Other investment products and services, such as brokerage and advisory accounts, are offered through J.P. Morgan Securities LLC ("JPMS"), a member of EINRA and SIPC. Insurance products are made available through Chase Insurance Agency, Inc. (CIA), a licensed insurance agency, doing business as Chase Insurance Agency Services, Inc. in Florida. JPMCB, JPMS and CIA are affiliated companies under the common control of JPM. Products not available in all states. References to "J.P. Morgan" are to JPM, its subsidiaries and affiliates worldwide. "J.P.Morgan Private Bank" is the brand name for the private banking business conducted by JPM. This material is intended for your personal use and should not be circulated to or used by any other person, or duplicated for non-personal use, without our permission. If you have any questions or no longer wish to receive these communications, please contact your J.P. Morgan team. Copyright 2026 JPMorgan Chase & Co. All rights reserved.