Economics and markets
Navigating the AI hype cycle: A case for active
October 21, 2026
- Commentary
- Regional outlooks
Commentary by Andrew Shuman, Director of Research, Vanguard Oversight & Manager Search
AI-related capital spending has dominated the U.S. economy and stock market. Investors have hardly had to differentiate between AI’s winners and losers as the rising tide of the data center buildout has seemingly lifted all boats across the supply chain.
Vanguard continues to believe that naively chasing the recent winners of the AI buildout creates a risk of buying high, and that companies outside traditional tech that use AI to improve efficiency and margins represent the next opportunity area. So how can investors position their portfolios for what history suggests could be a forthcoming shift in leadership?
Is semiconductor leadership poised for a shift?
Notes: This chart compares rolling three-year returns for semiconductor and semiconductor equipment companies within the Russell 1000 Index with returns for the index as a whole, which are represented by a zero baseline. Semiconductors and semiconductor equipment are as defined by the Global Industry Classification Standard. The Russell 1000 represents the 1,000 largest publicly traded stocks in the U.S.
Sources: Vanguard calculations, based on data from Bloomberg, as of September 30, 2026.
A potential echo of prior technology cycles
Thematic and sector/industry ETFs focused on semiconductors or data center construction have substantially outperformed the broader equity market in recent years (despite a 10.2% decline this month, through October 16).
These ETFs have garnered substantial inflows in the process, perpetuating their momentum. Investors pumped nearly $82 billion into technology stocks in the 12 months through August 2026, according to Morningstar. That was more than three times the net inflows into the next-leading sector, industrials. Given their cyclicality, semiconductor and semiconductor equipment stocks tend to have higher betas and volatility, often experiencing significant drawdowns—as they did in July when falling by more than 12%.
A shift in leadership would echo prior technology cycles, from the buildout of railroads in the mid-19th century to 1990s broadband. New AI-enabled business models may also outperform incumbents, as e-commerce and social media companies built atop the internet did in the 2000s and 2010s.
For investors looking to position portfolios for this potential shift in leadership, value stocks are a natural first instinct. These “old economy” companies that could benefit from AI adoption are disproportionately represented in value indexes. However, depending on the specific index, exposure to potential AI adopters may also bring exposure to companies known as “hyperscalers.” For example, Amazon and Microsoft, which are spending billions of dollars on AI data centers, are now two of the Russell 1000 Value Index’s three largest holdings.
Additionally, the productivity benefits of AI adoption among old economy or value stocks may or may not flow through to equity holders and stock prices. Potential margin gains will depend on management quality, data infrastructure, and competitive advantages. In highly competitive industries where all participants can achieve similar savings using the same AI models, competition may result in passing those savings to customers through lower prices, rather than leading to higher margins.
From hype to reckoning: Where will AI profits accrue?
Talented active managers, when accessed at reasonable costs, could prove their mettle in navigating this environment across both equities and fixed income. As the AI cycle matures and hype shifts to a real reckoning with economic profitability and return on investment, active managers could have wide scope to add alpha by identifying underappreciated long-term winners within the AI buildout, AI adopters, and AI-enabled disruptors.
Dispersion within individual stocks has widened significantly despite historically normal levels of overall market volatility, creating an ideal backdrop for stock selection.
Greater volatility within individual stocks
Note: This chart compares the Cboe S&P 500 Constituent Volatility Index (VIXEQ), which measures the implied volatility of individual stocks within the Standard & Poor’s 500 Index, with the Cboe Volatility Index (VIX), which measures the implied volatility of the S&P 500 over the next 30 days.
Source: Bloomberg, as of October 8, 2026.
Likely winners and losers among the hyperscalers
History suggests that the trillions of dollars being poured into building out AI data centers could result in overcapacity that leads to broadly disappointing profits and stock returns, but it’s still likely there will be winners and losers among the hyperscalers. To some degree, the market has already baked in some of this risk, with valuations sitting around 20 to 25 times expected earnings.
Active managers can add alpha by identifying the hyperscalers that:
- Offer the best potential return on investment based on their ability to build products for both consumers (such as internet search and personal agents) and enterprises (such as cloud computing).
- Can deliver these products efficiently and at low cost.
- Maintain strong balance sheets as they increasingly tap both public and private credit markets for capital.
Further, understanding how these hyperscalers will adapt their data center construction to work around bottlenecks and integrate leading-edge technologies to lower costs will be key to identifying the sustainable winners in the supply chain of semiconductors and industrials—versus companies enjoying shortage-driven pricing power that is likely to prove transitory.
Among AI adopters seeking to use the technology to improve efficiencies, active managers can use their knowledge of management teams, proprietary data, and industry dynamics to identify sustainable winners. This may become more important as AI token costs fall. Previously, larger companies with the biggest budgets had a clear advantage in adopting AI. As costs decrease and the ability to experiment increases, value could accrue to smaller companies with more nimble cultures and management teams willing to disrupt their existing business models.
Last but not least, active managers can seek to identify the new growth companies with business models built atop AI that are either still private or represent only very small weights in cap-weighted indexes. This is akin to investing in e-commerce or mobile internet companies whose business models would not have been possible without the dot-com bubble’s expansion of broadband internet. Ultimately, these were the most successful companies and stock market investments of that era and now sport multitrillion-dollar market capitalizations.
Active managers able to identify the Amazons and Googles of the 2030s may be well positioned to deliver attractive returns to their shareholders.
Notes:
All investing is subject to risk, including possible loss of the money you invest.
Funds that concentrate on a relatively narrow market sector face the risk of higher share-price volatility.
While skilled managers may add excess return, active strategies involve higher fees and the risk of underperforming comparable index investments; alpha is not guaranteed and may be negative.
Prices of mid- and small-cap stocks often fluctuate more than those of large-company stocks.
London Stock Exchange Group companies include FTSE International Limited (“FTSE”), Frank Russell Company (“Russell”), MTS Next Limited (“MTS”), and FTSE TMX Global Debt Capital Markets Inc. (“FTSE TMX”). All rights reserved. “FTSE®,” “Russell®,” “MTS®,” “FTSE TMX®,” and “FTSE Russell” and other service marks and trademarks related to the FTSE or Russell indexes are trademarks of the London Stock Exchange Group companies and are used by FTSE, MTS, FTSE TMX, and Russell under license. All information is provided for information purposes only. No responsibility or liability can be accepted by the London Stock Exchange Group companies nor its licensors for any errors or for any loss from use of this publication. Neither the London Stock Exchange Group companies nor any of its licensors make any claim, prediction, warranty, or representation whatsoever, expressly or impliedly, either as to the results to be obtained from the use of the FTSE Indexes or the fitness or suitability of the Indexes for any particular purpose to which they might be put.
© 2026 Morningstar, Inc. All rights reserved. The information contained herein: (1) is proprietary to Morningstar and/or its content providers; (2) may not be copied or distributed; and (3) is not warranted to be accurate, complete or timely. Neither Morningstar nor its content providers are responsible for any damages or losses arising from any use of this information. Past performance is no guarantee of future results.