AI capital expenditures (capex) are adding up fast. Technology, Media, and Telecommunications (TMT) giants are expected to spend about $2.7 trillion on data centers and AI infrastructure in the U.S. by 2030, McKinsey & Company estimates.2 Annual AI‑related capex at the largest hyperscalers is already approaching $100 billion.3 The projected spend is large enough that hyperscalers increasingly are shifting the funding for AI-related capex from self‑funded projects to cycles of capital-market‑funding.
Historically, cash‑rich firms such as Microsoft Corp., Alphabet Inc. (Google), Amazon.com, Inc., Meta Platforms, Inc. (Facebook), and Oracle Corp., funded capex from free cash flow (FCF). As the capex demands of AI growth increase, these companies are engaging in regular bond issuance and structured financing to support AI buildouts that include data centers, graphics processing units (GPUs), and power supply demands.
Corporate spending guidance in the cases of four large TMT firms reflects the size of the planned spending (See Figure 1).
Debt Funding AI Capex
Hyperscalers, historically, generate significant cash flow from operations (CFO), which has funded capex spending undertaken by the companies (See Figure 2).
Despite significant CFO, current projected capex spending for AI-related initiatives is expected to result in negative FCF for these companies, at least initially. As a result, additional funding sources are necessary to pay for capex. Funding sources include IG debt (including corporate bonds, asset-backed securities (ABS), and commercial mortgage-backed securities (CMBS), off-balance sheet financing (including joint venture financing), special project vehicles, private credit, and leasing.
In recent months, hyperscalers engaged in substantive rounds of IG debt issuance, resulting in Meta, Alphabet, Amazon, and Oracle’s collective weighting in the Bloomberg (BBG) U.S. Corporate IG Index4 to nearly double over the year ending April 1, 2026, from 2.2 percent to 4.1 percent.
AI‑linked issuance was roughly five percent of total IG corporate bond issuance in 2025 and approximately three times the prior decade’s average annual tech issuance.5 Increased debt issuance in the sector may cause material changes to benchmark composition,6 supply/demand dynamics, and investment considerations for credit investors.
Considering the Investment Implications of Debt Funding for AI Growth
Additional bond issuance to fund accelerated AI development is increasing leverage in the near term, as companies must finance data center buildouts in advance of seeing returns from customers such as OpenAI and Anthropic.
Most hyperscalers entered this period of higher IG debt issuance with extremely strong balance sheets, leverage below 1x, or net cash positions. The exception was Oracle, with net leverage above 3.5x. Among other things, hardware obsolescence, specifically from GPUs, will result in ongoing capital needs, rather than a brief period of elevated capex.
Higher leverage and periods of negative FCF could pressure credit quality ratings of some companies, depending on the tolerance a rating agency may have for higher leverage during periods of time before returns on capital investments are realized. Given the high-quality nature of most hyperscalers, the risk of a ratings downgrade may be relatively lower.7
Along with increasing leverage, hyperscalers may face added concentration risk. In early 2026, OpenAI and Anthropic are two of the largest purchasers of AI infrastructure services. These large AI customers can be pre-profit or venture backed. If AI customers cannot monetize their products/services as expected and hyperscalers face revenue collection challenges, AI infrastructure obsolescence could emerge.
Already, increased supply has come with widening spreads during the first quarter of 2026, at levels indicative of lower ratings (See Figure 3). Driven by supply technicals, sustained increased issuance may continue to impact spreads across the sector. Changing risks and wider spreads within the sector are altering risk dynamics in what has historically been a defensive sector.
While more supply will be necessary, the amount of IG debt relative to other financing sources is uncertain. Persistent supply could result in continued pressure on spreads, despite strong fundamentals. At the same time, these dynamics could lead to opportunities for stronger returns.
Conclusions and Considerations for Sector Investors
AI‑linked issuance is rising sharply. As the level of debt issued by hyperscalers increases, the relative size of the sector represented in common capitalization-weighted bond indexes is increasing. More sector concentration in indices could increase potential spread pressure from supply overhang. Variation in business models, capex intensity, and funding strategies could drive greater issuer-level spread dispersion, which could reward active selection.
While the largest hyperscalers remain IG quality issuers, smaller names could be more fragile. The risk of widespread credit deterioration may be lower for now, but the rapidly changing AI business model could result in higher risks over the longer term. Hyperscalers with strong FCF and cash balances, clear AI monetization paths, and the capacity to slow or sequence capex if returns disappoint, seem best positioned to offer risk/reward advantages.
[1] Hyperscalers operate large, highly scalable data center infrastructure designed to support large-scale cloud computing services and workloads. The data centers can quickly grow, or scale, to manage massive amounts of data, growing numbers of users, and persistently increasing processing demands. Commonly known hyperscalers include Amazon Web Services (AWS), Microsoft Azure, Google Cloud Platform (GCP), and Meta (Facebook) and sometimes Apple (for internal infrastructure).
[2]“The data center balance: How U.S. states can navigate the opportunities and challenges,” Adam Barth, Chhavi Arora, Gayatri Shenai, Jesse Noffsinger, and Pankaj Sachdeva, McKinsey & Company, August 8, 2025.
[3]“The cost of compute: A $7 trillion race to scale data centers,” Jesse Noffsinger, Mark Patel, Pankaj Sachdeva, Arjita Bhan, Haley Chang, and Maria Goodpaster, McKinsey Quarterly, McKinsey & Company, April 28, 2025. “Our research shows that by 2030; data centers are projected to require $6.7 trillion worldwide to keep pace with the demand for compute power. Data centers equipped to handle AI processing loads are projected to require $5.2 trillion in capital expenditures, while those powering traditional IT applications are projected to require $1.5 trillion in capital expenditures. Overall, that’s nearly $7 trillion in capital outlays needed by 2030—a staggering number by any measure.”
[4] The BBG U.S. IG Corporate Bond Index is an unmanaged market-value-weighted index of IG corporate fixed-rate debt issues with maturities of one year or more. You cannot invest directly in an index.
[5] From 2020 through 2025, U.S. IG corporate bond issuance averaged roughly $1.5 to $1.6 trillion annually. Total U.S. corporate bond issuance, including high-yield, often exceeded $2 trillion annually during this period, according to data from the Securities Industry Financial Markets Association (SIFMA) and Akin Gump Strauss Hauer & Feld LLP, as of March 31, 2026.
[6] As a corporate bond issuer issues more debt, the total market value of its outstanding bonds increases, leading to a higher weight in market-value-weighted indices.
[7] Oracle is an exception. The issuer is BBB-rated, as of March 31, 2026. with negative outlooks at S&P Global Ratings and Moody’s Investors Service.
BCAI-04162026-j6li9kug (4/23/2026)
DISCLAIMERS:
The content is intended for investment professionals and institutional investors.
This material provides general information and should not be construed as a solicitation or offer of services or products or as legal, tax or investment advice. Nothing contained herein should be considered a guide to security selection, asset allocation or portfolio construction.
All information and opinions are current as of the dates indicated and are subject to change. Breckinridge believes the data provided by unaffiliated third parties to be reliable but investors should conduct their own independent verification prior to use. Some economic and market conditions contained herein have been obtained from published sources and/or prepared by third parties, and in certain cases have not been updated through the date hereof.
There is no assurance that any estimate, target, projection or forward-looking statement (collectively, “estimates”) included in this material will be accurate or prove to be profitable; actual results may differ substantially. Breckinridge estimates are based on Breckinridge’s research, analysis and assumptions. Other events that were not considered in formulating such projections could occur and may significantly affect the outcome, returns or performance.
Not all securities or issuers mentioned represent holdings in client portfolios. Some securities have been provided for illustrative purposes only and should not be construed as investment recommendations. Any illustrative engagement or sustainability analysis examples are intended to demonstrate Breckinridge’s research and investment process.
Yields and other characteristics are metrics that can help investors in valuing a security, portfolio or composite. Yields do not represent performance results but they are one of several components that contribute to the return of a security, portfolio or composite. Yields and other characteristics are presented gross of advisory fees.
All investments involve risk, including loss of principal. No investment or risk management strategy, including diversification, can guarantee positive results or risk elimination in any market. Periods of elevated market volatility can significantly impact the value of securities. Investors should consult with their advisors to understand how these risks may affect their portfolios and to develop a strategy that aligns with their financial goals and risk tolerances.
Active investing generally involves more risks than laddered strategies because active managers may take on greater market risk to outperform their index. There is no guarantee that either passive or active investing will achieve their objectives. Active strategies also tend to have higher management fees and operating costs than passive strategies. Investors should consider all the differences and risks before making any investment decisions. Active management does not guarantee a profit or protect against a loss.
Past performance is not indicative of future results. Breckinridge makes no assurances, warranties or representations that any strategies described herein will meet their investment objectives or incur any profits. Performance results for Breckinridge’s investment strategies include the reinvestment of interest and any other earnings, but do not reflect any brokerage or trading costs a client would have paid. Results may not reflect the impact that any material market or economic factors would have had on the accounts during the time period. Due to differences in client restrictions, objectives, cash flows, and other such factors, individual client account performance may differ substantially from the performance presented.
Actual client advisory fees may differ from the advisory fee used to calculate net performance results. Client returns will be reduced by the advisory fees and any other expenses incurred in the management of their accounts. For example, an advisory fee of 1 percent compounded over a 10-year period would reduce a 10 percent return to a 9 percent annual return. Additional information on fees can be found in Breckinridge’s Form ADV Part 2A.
Index results are shown for illustrative purposes and do not represent the performance of any specific investment. Indices are unmanaged and investors cannot directly invest in them. They do not reflect any management, custody, transaction or other expenses, and generally assume reinvestment of dividends, income and capital gains. Performance of indices may be more or less volatile than any investment strategy.
Fixed income investments have varying degrees of credit risk, interest rate risk, default risk, and prepayment and extension risk. In general, bond prices rise when interest rates fall and vice versa.
Equity investments are volatile and can decline significantly in response to investor reception of the issuer, market, economic, industry, political, regulatory or other conditions.
Breckinridge believes the data provided by unaffiliated third parties, including rating agencies, to be reliable but investors should conduct their own independent verification prior to use. Some economic and market conditions contained herein have been obtained from published sources and/or prepared by third parties, and in certain cases have not been updated through the date hereof. All information contained herein is subject to revision. Any third-party websites included in the content has been provided for reference only, and does not necessarily indicate an endorsement.
BLOOMBERG® is a trademark and service mark of Bloomberg Finance L.P. and its affiliates (collectively “Bloomberg”). Bloomberg does not approve or endorse this material or guarantees the accuracy or completeness of any information herein, or makes any warranty, express or implied, as to the results to be obtained therefrom and, to the maximum extent allowed by law, neither shall have any liability or responsibility for injury or damages arising in connection therewith.
The S&P500 Index (“Index”) and associated data is a product of S&P Dow Jones Indices LLC, its affiliates and/or their licensors and has been licensed for use by Breckinridge. © 2025 S&P Dow Jones Indices LLC, its affiliates and/or their licensors. All rights reserved. Redistribution or reproduction in whole or in part are prohibited without written permission of S&P Dow Jones Indices LLC. For more information on any of S&P Dow Jones Indices LLC’s indices please visit www.spdji.com. S&P® is a registered trademark of Standard & Poor’s Financial Services LLC (“SPFS”) and Dow Jones® is a registered trademark of Dow Jones Trademark Holdings LLC (“Dow Jones”). Neither S&P Dow Jones Indices LLC, SPFS, Dow Jones, their affiliates nor their licensors (“S&P DJI”) make any representation or warranty, express or implied, as to the ability of any index to accurately represent the asset class or market sector that it purports to represent and S&P DJI shall have no liability for any errors, omissions, or interruptions of any index or the data included therein.
Subscribe to Insights
Sign up to receive curated insights directly in your inbox.
