Global spending on artificial intelligence infrastructure is tracking toward a new headline figure. According to a Morgan Stanley report, worldwide AI-related capital expenditures are projected to reach approximately $1.58 trillion by 2027, a figure commonly rounded to $1.6 trillion. The more striking detail is where that money is expected to land: Nvidia could collect roughly $690 billion of it, or about $700 billion in some assessments. If the projection holds, a single chipmaker would emerge as the biggest beneficiary of one of the largest corporate building cycles on record.

Tracing the Spending

The bank's methodology functions like a funnel. Roughly 65% of total AI capex—about $1.03 trillion—is expected to flow to companies in the S&P 500. Nvidia is then projected to take about two-thirds of that S&P 500 share. The remainder would flow to companies outside the index.

The overall forecast has grown as well. Morgan Stanley raised its 2027 total by about 14% after incorporating lease obligations that had not previously been counted. Lease commitments are a less visible form of spending: a company that rents data capacity rather than building it still commits substantial funds to AI, but the expense shows up on a different line of the financial statements. Once those obligations entered the tally, the total climbed.

Broken down by category, semiconductors and servers are projected to account for about 42% of 2027 AI capex, or approximately $672 billion.

Nvidia's Guidance Points in the Same Direction

The projection lines up with what Nvidia itself has been telling investors. Speaking on the company's fiscal Q2 2027 earnings call, CFO Colette Kress indicated that hyperscaler capital spending is expected to approach $800 billion in 2026 and reach around $1.3 trillion in 2027. Kress also highlighted a cloud industry backlog exceeding $2 trillion. The two forecasts are not the same measure: Morgan Stanley's total covers worldwide AI-related spending of all kinds, while Kress's figures describe hyperscaler budgets alone.

The hyperscalers doing the heavy lifting are familiar names: Microsoft, Alphabet, Amazon, Meta, and Oracle. These companies build and operate the giant data centers that train and serve AI models, and Nvidia's GPUs sit at the center of most of that buildout. Nvidia's chips and associated platforms are considered critical to this infrastructure spending, and upcoming technology such the Blackwell platform is a key part of how the company expects to keep serving that demand.

There is less agreement on the near-term pace, however. Estimates for 2026 hyperscaler capex have ranged from about $650 billion to more than $800 billion, with forecasters adjusting their figures throughout the year.

What It Means for the S&P 500

The spending wave is also expected to register in the broader U.S. equity market. Morgan Stanley expects AI-related capex to contribute about 5% of S&P 500 revenue in 2027 and to drive roughly 25% of the index's sales growth that year. Earnings tell a similar story: AI “enablers” and “adopters” are expected to drive 78% of S&P 500 earnings growth in 2027.

The Takeaway

The clearest theme for investors is concentration. Should Nvidia capture approximately $690 billion, one company would absorb a substantial share of the AI capex wave.

The lease adjustment is also worth watching. A 14% upward revision driven by previously uncounted commitments suggests the true scale of AI spending may be harder to track than headline capex figures imply. The more than $2 trillion cloud backlog Kress cited offers some reassurance, since it reflects demand that customers have already committed to.

From here, the key markers are whether hyperscaler capex guidance keeps climbing, how quickly Blackwell and related platforms reach customers, and whether Morgan Stanley or other forecasters adjust their 2027 totals again.