Morgan Stanley has taken the lead among Wall Street banks for underwriting AI-related debt, benefiting significantly from what its CEO, Ted Pick, terms an "AI capex super cycle." This surge is driven by tech giants financing the massive build-out of data centers. Pick noted that the forecast for data center capital expenditure for 2026, initially around $575 billion, has already jumped to $850 billion, with projections for 2027 rising from $700 billion to $1.3 trillion, and potentially reaching $1.5 trillion in 2028. Morgan Stanley expects total AI-related capital expenditure to hit $10 trillion over several years. This robust demand is generating substantial fees for investment banks through capital raising, loans, and advisory services.
Investment banks across the board are seeing a significant boost in earnings due to this AI-driven infrastructure spending. Goldman Sachs CEO David Solomon emphasized that the industry is in the midst of an AI capital expenditure super cycle, demanding the utilization of every financing instrument. Goldman Sachs reported a record $3.4 billion in investment banking fees in the second quarter, a 55% increase year-on-year. Similarly, JPMorgan reported $3.3 billion, up 30%, and Morgan Stanley's investment banking revenue soared 58% to $2.44 billion, exceeding analyst expectations and contributing to record net income of $5.58 billion. Bank of America and Citigroup also saw substantial increases in their investment banking revenues.
Morgan Stanley's research projects that AI-related bond issuance alone could reach approximately $570 billion in 2026, more than double the previous year. This massive debt issuance is a key mechanism for hyperscalers like Amazon, Alphabet, Microsoft, and Meta to fund their data center expansions, rather than relying solely on operating cash flow. Such activity creates a continuous pipeline of underwriting mandates, advisory work, and related trading activities for banks. Morgan Stanley has already secured significant roles in major AI-related deals, including an underwriting role in the expected IPOs of Anthropic and OpenAI, and was involved in large debt financings like the roughly $13 billion package for Meta's data center in El Paso, Texas.
The widespread impact of AI infrastructure development extends beyond direct tech companies. This demand for financing is seen across various industries and regions, boosting equity issuance, M&A activity, and debt financing. Even companies not directly involved in AI are benefiting from indirect links, such as increased demand for plumbers and electricians for data center construction. The "AI capex super cycle" is transforming Wall Street's landscape, leading to record revenues and profits for major banks, and analysts like Stephen Biggar of Argus Research confirm that this multi-year investment cycle will continue to drive elevated levels of strategic activity and capital formation across markets.
