Apollo Chief Economist Torsten Slok has raised underinvestment as an open question—not declared that the AI economy has a proven, market-wide infrastructure shortfall. His June 2026 framing is a test of compute demand, price and supply, alongside the risk that new capacity could outpace profitable demand.
What did Apollo actually say about underinvestment?
In Apollo’s June 30, 2026 mid-year outlook, Partner and Chief Economist Torsten Slok asks: “Are there areas where we may have seen underinvestment?” He describes the AI cycle through three indicators: “what demand for compute will look like, what the price of compute will look like, and what the supply of compute will look like.” That is a question to investigate, not a finding that the market has a quantified infrastructure gap.
The attribution matters. Olivia Wassenaar’s name appears in Apollo’s 2026 infrastructure outlook, but the cited formulation about possible underinvestment is from Slok’s June interview. Apollo’s company-published outlooks and commentary express the firm’s views; they are not, by themselves, independent measurements of market-wide supply or consensus forecasts.
What has to be built for AI capacity to reach users?
AI infrastructure is a chain of connected assets. A data center needs compute hardware and networking, but also power, a grid connection, suitable sites, permits, skilled labor and equipment such as transformers. A shortage at any one layer can delay the usable capacity promised by investment elsewhere. Spending announced for chips or buildings is therefore not the same as compute that is installed, powered and available to customers.
Recommended Free Tools
#1 Best Overall
Digital infrastructure and compute
Apollo’s July 2026 infrastructure outlook groups data centers, semiconductors, cell towers and fiber networks under digital infrastructure. These assets work together: chips process workloads, while networking and fiber move data between facilities and users. Apollo identifies rising compute needs as one reason these linked assets may require further investment.
Power, storage and grid connections
Data centers need reliable electricity, and new generation alone does not guarantee that a particular facility can draw power where and when it needs it. Apollo’s outlook also identifies renewable energy, storage and grid modernization as connected investment needs. Permitting, grid connections and transformers can constrain delivery even where demand for compute is strong.
Rank #2
- Ideal for Gifting
- Ideal for a bookworm
- Compact for travelling
Financing the long-lived assets
Buildings, power equipment and related infrastructure require substantial capital before all projected revenues arrive. Apollo’s March 20, 2026 analysis asks who bears the balance-sheet risk if revenues lag. It argues that recurring revenue and free cash flow can make the buildout more self-funding, but that depends on those cash flows materializing.
What figures is Apollo using—and what do they mean?
The figures below are claims or projections published by Apollo, not independently verified measures of an AI infrastructure shortfall.
The Tool Desk
Outbyte PC Repair FREEClear out junk files and repair common Windows errorsFree Scan →Outbyte Driver Updater FREEScan for outdated or missing drivers - takes under a minuteDriver Scan →| Figure | Apollo’s claim | How to read it |
|---|---|---|
| U.S. electricity demand | Apollo’s 2026 outlook says demand grew 0.8% annually from 2000 to 2024 and projects a 2.2% compound annual growth rate through 2050. | The 2.2% figure is a projection, not a measured growth rate or a guarantee of future demand. |
| New energy capacity | Apollo’s 2026 outlook says more than 90% of new energy capacity built in 2024 was clean energy and says the trend continued into 2025. | The more-than-90% figure refers to 2024, not 2025. |
| Data centers and related infrastructure | Apollo’s July 2026 discussion gives a range of $5 trillion to $6 trillion through 2030 for data centers and related power infrastructure, chips and networking, depending on whose estimates are used. | This is a range reported in Apollo’s discussion, not one audited or universally accepted forecast. |
| Broadcom AI XPV Platform | Apollo’s June 9, 2026 transaction announcement says Apollo-managed funds and affiliates led a $35 billion initial capital solution with Blackstone and leading global banks, targeting more than 20 GW of compute capacity through 2028. | The announcement describes an initial financing solution and a capacity target. It does not establish that the target capacity has already been built or is operating. |
The size of an estimate or financing announcement does not answer whether supply is sufficient. To assess that, distinguish projected spending from capacity delivered, and capacity delivered from capacity that is powered, contracted and generating revenue.
Why could supply be tight while some projects still lose money?
The supply and return questions are related but not interchangeable. Apollo’s July 2026 infrastructure article says compute demand is outpacing available supply and points to constraints including power, skilled labor, permitting, transformers, semiconductors and grid connections. If these bottlenecks persist, projects may take longer to complete and capacity could arrive too slowly in particular places or for particular uses.
Rank #4
But a shortage today does not prove that every planned project will earn an adequate return. In his July 29, 2026 Daily Spark, Slok asks, “Will there be unlimited demand for compute, or will compute demand peak?” Apollo also warns that efficiency gains, commoditization or slower enterprise adoption could leave demand below planned capacity. If expected customer use or pricing does not arrive before assets depreciate, investment can fall short even where construction was difficult.
Financing adds another condition. Apollo’s July commentary identifies debt financing and the cost of capital as factors that can constrain marginal investment. A project can be technically useful yet uneconomic if its expected revenues do not cover its financing and operating costs. These are risks Apollo raises, not proof that a particular project is mispriced or that a downturn is imminent.
Outdated Drivers Are Slowing You Down
One free scan finds every outdated or missing driver and matches the right update for your exact hardware.Free scan · exact hardware matchPC Slower Than It Used to Be?
A free scan shows the junk files, broken settings and background clutter dragging Windows down - then fixes them in one click.Free scan · Windows 10 & 11Best Value
- It can be a gift option
- Comes with secure packaging
- Helpful in various ways
How can investors evaluate an AI infrastructure claim?
Compare the project’s assumptions rather than relying on a single market-growth estimate or spending headline. The following questions synthesize the issues Apollo raises:
- Compute demand: Who is expected to use the capacity, and what supports the demand forecast?
- Delivered capacity: Is the figure planned, under construction, installed or actually available for workloads?
- Power access: Is generation available, and does the project have a practical grid connection, permits and needed equipment?
- Customer commitments: Are revenues supported by recurring contracts, or depend on future customers and usage?
- Financing: What debt, equity or other capital funds the project, and how sensitive is its viability to the cost of capital?
- Utilization and asset life: How much use does the business need to earn an adequate return, and can that use persist over the assets’ useful life?
These checks help separate a genuine bottleneck from a compelling but uncertain forecast. They do not establish that a project will succeed; actual outcomes depend on utilization, pricing, delivery and financing conditions.
What does Apollo’s view mean for a personal-finance reader?
Apollo’s outlook is not a personal investment recommendation, and its capacity and spending figures do not establish that a particular stock, fund, utility bill or AI-related investment will rise or fall. For a household investor, the useful distinction is between a broad theme and the economics of the specific asset or investment being considered: demand forecasts, contracts, financing and the ability to deliver capacity all matter. Apollo’s infrastructure outlook also says opinions may change and offers no assurance that the trends it describes will continue.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
Free tools Windows power users keep installed
One-click scans. No signup required.
