Wall Street re-examines AI investment: The bubble debate is meaningless; the key is how to address the misalignment and lag between investment and monetization

Wall Street is reshaping the investment narrative around artificial intelligence (AI).

Top financial institutions, including BlackRock and Bank of America, believe that it is pointless to get entangled in whether AI is a "bubble"; the real focus should now shift to how to address the misalignment and lag between massive upfront investments and commercial monetization, a phase that may bring new turbulence to the market.

Recently, Jean Boivin, head of BlackRock Investment Institute, clearly stated that at this stage, using a "bubble" framework to evaluate AI investments is unhelpful. He believes that given the unprecedented scale and speed of AI infrastructure development, any retrospective assessment based on historical indicators is "incomplete." BlackRock pointed out that the scale of capital expenditure in the AI field is enormous, constituting a macroeconomic event in itself.

Bank of America, on the other hand, issued a more specific warning. Savita Subramanian, head of U.S. equity and quantitative strategy, also denied the "bubble theory," but she warned that the market may face an "air pocket." She described it as a phase where capital expenditure growth exceeds revenue growth, and this lag between investment and monetization, especially under the constraints of power and infrastructure bottlenecks, could shake investor confidence in the short term.

Despite the short-term risks, both institutions maintain an optimistic long-term outlook. They believe that the current AI boom is driven by real corporate profits, investments, and productivity growth, fundamentally different from the internet bubble of the early 21st century.

BlackRock: The "AI bubble theory" is misleading; the real challenge lies in the gap between spending and revenue

Wall Street giants are guiding the market to shift its attention away from the debate of "Is AI a bubble?" Jean Boivin of BlackRock pointed out in a media roundtable on Tuesday that simply defining the current AI boom as a bubble is misleading.

He emphasized that there is a healthy skepticism in the market, stating, "There is so much discussion about potential bubbles... people are aware of the risks," which is in stark contrast to the lack of risk awareness during the irrational exuberance period.

In its outlook report, BlackRock stated that the ambition for AI-related capital expenditure is so vast that "micro is macro." The institution believes that this scale of investment is expected to drive U.S. GDP growth to consistently break the 2% trend line that has dominated for decades. This signifies that AI is not only a technological innovation but also a macro driver capable of reshaping the overall economic landscape.

Behind the grand narrative of AI lies astonishing capital expenditure plans and real physical constraints. BlackRock estimates that by 2030, global corporate spending plans in the AI field will range between $5 trillion and $8 trillion, with most occurring in the United States. The report stated, "The challenge for investors is: how to align the enormous capital expenditure plans with potential AI revenues. Will their magnitudes match?"

This "upfront investment" model, while bringing about transformation, also makes it vulnerable. The report pointed out the physical limits from computing power to the power grid and predicted that by the end of this decade, AI data centers could account for 15% to 20% of total electricity consumption in the United States. BlackRock believes that this upfront spending is necessary to achieve eventual returns, but it also constitutes a structural pressure that investors must confront

Bank of America: Beware of the "Air Pocket" - The Lagging Risks of AI Investment and Monetization

Bank of America has issued a clearer warning regarding the next phase of the AI investment cycle. Savita Subramanian believes that while AI's leadership will not continue "unconstrained," it is more like a "pause" rather than the beginning of a collapse. Her concept of the "air pocket" precisely describes the temporary disconnection between investment and returns.

Data shows that this risk has already manifested on balance sheets. According to Bank of America, in the past year, the capital expenditures of hyperscalers have risen to 60% of their operating cash flow, up from 30% a decade ago. Although this ratio is still far below the peak of 140% during the internet bubble, the rapid growth of expenditures cannot be ignored. The bank predicts that the massive spending from companies like Microsoft, Amazon, Google, Meta, and Oracle will reach $400 billion in 2025 and $510 billion in 2026.

A Different Era: Why the Current Situation is Not a Repeat of 2000

Despite the narrowing market breadth and high valuations being "similar" to those in 2000, Savita Subramanian emphasizes several key differences that explain why history will not simply repeat itself. She points out that, compared to the internet bubble period, current investors have a much lower allocation to stocks, overvaluation is supported by corporate earnings growth, the IPO market is smaller, and speculative behavior towards unprofitable companies is far less extreme than in the late 1990s.

These fundamental differences support Bank of America's long-term bullish outlook. The bank forecasts that the S&P 500 index (^GSPC) will reach 7,100 points by the end of 2026. This position indicates that, despite the short-term turbulence risk of an "air pocket," Wall Street generally believes that the AI revolution, driven by real fundamentals, remains on a solid long-term upward trajectory