On June 16, 2026, the US stock market experienced a landmark session. The Dow Jones Industrial Average surged 468.77 points to close at 51,671.03, setting a new all-time high. The Nasdaq Composite jumped 795.10 points, or 3.07%, closing at 26,683.94. The semiconductor sector led the rally—NVIDIA climbed 3.54%, pushing its market cap back above $5 trillion. Micron Technology soared 10.84%, and Western Digital surged 16.10%.
However, the more significant signals lie beneath these headline gains. The Philadelphia Semiconductor Index rose over 5% that day, closing at a record high. At the same time, AI-related crypto tokens collectively gained about 20%, with Bittensor (TAO) hitting a three-week high of $283. The market’s AI narrative is evolving from a single-minded "compute arms race" to a multi-layered capital rotation spanning infrastructure, platform, and application layers.
This shift aligns with BlackRock’s three-phase AI investment framework. As the world’s largest asset manager, overseeing more than $13 trillion, BlackRock’s 2026 annual outlook highlighted "AI, income, and diversification" as its three pillars. At the core is a key insight: AI investing is not a short-term trade, but a decade-long structural cycle.
Phase One: The Compute Arms Race Continues, But an Inflection Point Emerges
BlackRock’s three-phase framework—Buildout, Adoption, and Transformation—maps out a clear evolution for AI investment. The market is still in phase one: a "super-boom" of massive capital expenditures focused on chips, data centers, power, and supporting infrastructure.
How big is this phase? Hyperscalers are expected to invest over $500 billion in 2026 to expand AI-ready infrastructure. Tech giants—Microsoft, Alphabet, Amazon, Meta, and Oracle—will collectively spend between $660 billion and $725 billion in capital expenditures in 2026, with about 75% directed toward AI infrastructure. BlackRock notes that AI-related capex will contribute to US economic growth at three times the historical average in 2026.
NVIDIA’s numbers illustrate this trend vividly. In Q1 of its fiscal 2027, the company posted $81.615 billion in revenue, up 85% year-over-year, with data center revenue reaching $75.246 billion, up 92%. CEO Jensen Huang described this as "the largest infrastructure expansion in human history." The company also completed a $25 billion bond issuance, attracting $85 billion in subscriptions.
Yet, BlackRock’s Spring 2026 investment outlook sent a key message: while still bullish on AI infrastructure, the focus for 2026 is shifting toward "diversification beyond mega-cap tech." The subtext: phase one’s valuation concentration risks are building. The "Magnificent Seven" now account for over 40% of the S&P 500’s market cap—a record high. Jay Jacobs, BlackRock’s Head of Thematics and Active ETFs, put it bluntly: this concentration is "either a feature or a bug."
Phase Two: From Selling Shovels to Digging for Gold
In BlackRock’s framework, phase two focuses on "companies implementing AI into their businesses." Jay Jacobs, speaking on the Earn Your Leisure podcast, drew a vivid analogy: during the 1990s telecom bubble, $500 billion in fiber overbuild ultimately handed the remaining value to Google and Amazon. A similar rotation is now underway in AI.
What supports this view? A structural feature of AI investment is "front-loaded investment, back-loaded returns"—infrastructure spending is concentrated early, while revenue lags behind. This mismatch creates a financing "camel hump," meaning that as infrastructure comes online, the first to realize commercial value may not be chipmakers, but rather platform and application companies that turn AI capabilities into products and services.
Market data from 2026 is already showing early signs. JPMorgan notes that over the past 18 months, ETF investors seeking AI exposure have overwhelmingly favored semiconductor and hardware funds, while broader software ETFs have been relatively overlooked. This imbalance is drawing attention—investors are starting to look beyond chips and data centers, searching for companies that can truly monetize AI adoption.
BlackRock’s 2026 investment direction report goes further: in AI, the firm favors high-conviction alpha strategies, aiming for innovation and dynamic allocation across the AI stack. This signals a shift from passive "buy the chip stocks" to active AI allocation strategies. BlackRock’s iShares A.I. Innovation and Tech Active ETF (BAI) has already grown to over $8 billion in assets.
Phase Three: Real-World AI and Long-Term Expectations
The third phase is "real-world AI"—applications like autonomous vehicles and robotics. BlackRock notes that AI buildout may be faster and larger in scale than any previous tech revolution. But market predictions remain cautious: as of June 2026, the probability of Tesla’s robotaxi launch in California that month is just 3%. This suggests that meaningful returns from phase three are still some way off.
In its 2026 global outlook, BlackRock highlights three core investment themes: "Micro is Macro," "Leverage," and "Diversification Illusion." "Micro is Macro" emphasizes that micro-level AI investment decisions are reshaping the macroeconomic landscape—AI infrastructure investment alone will contribute to US economic growth at three times the historical average in 2026. "Leverage" points to risks: AI builders are taking on more leverage, and a highly leveraged financial system is more vulnerable to shocks like surging bond yields.
The Path of Capital Rotation and Beneficiary Sectors
Based on BlackRock’s framework, the path of capital rotation in AI investment is clear:
Phase One (currently dominant): Chips (NVIDIA, AMD), data center infrastructure, power and energy infrastructure, semiconductor equipment. Capital expenditures in this phase are expected to grow over 34% in 2026.
Phase Two (now emerging): AI platforms and cloud service providers, enterprise AI software and applications, AI-enabled data services, edge computing and inference infrastructure. BlackRock notes that US corporate earnings expectations are being steadily revised upward, even amid increased volatility and price declines. Upward earnings revisions for 2026 and 2027 are among the most significant since 1988.
Phase Three (long-term): Autonomous driving, robotics, AI agent economy. Gartner projects that by the end of 2026, 40% of enterprise applications will incorporate task-specific AI agents.
Notably, BlackRock’s Spring 2026 outlook emphasizes that emerging markets outside the US are playing a central role in AI buildout, leading the firm to prefer emerging market equities over developed markets. This view offers important guidance for global asset allocation.
Shifting Institutional Allocations
Institutional capital flows show AI allocations moving from "concentrated" to "diffuse." BlackRock’s surveys indicate that in 2026, investors seeking AI exposure prefer energy and infrastructure providers over Wall Street’s mega-cap tech stocks. Global energy ETPs attracted $16.9 billion in inflows in 2026, ending three consecutive years of net outflows. Emerging market equity ETPs have already drawn $21.7 billion in 2026, the second-largest annual inflow on record.
These data point to a trend: institutional investors are moving downstream in the AI value chain, expanding from upstream chip manufacturing to energy supply, emerging market infrastructure, and ultimately the application layer.
Conclusion
BlackRock’s three-phase framework offers a systematic perspective on the AI investment cycle. The market remains at the peak of the infrastructure buildout phase, but phase two—the penetration of AI capabilities into enterprises and industries—is accelerating. For investors, the key is to recognize the rhythm and signals as capital rotates from the "compute arms race" to the "application explosion."
As Jay Jacobs put it, "AI trading is short-term, but AI investing is a decade-long journey." Within this ten-year framework, the list of beneficiaries extends far beyond chipmakers. From enterprise AI software to energy infrastructure, from emerging markets to decentralized compute networks, AI’s value creation is expanding from a narrow track into a structural trend that spans the entire economy.
On June 16, 2026, with the Nasdaq up 3.07%, semiconductors led, software followed, and AI tokens resonated—a real-world preview of the three-phase framework playing out in the market.
