Goldman Sachs Projects Big Tech AI Infrastructure Spending to Hit $1.2 Trillion in 2027

Goldman Sachs analysts predict that the spending on artificial intelligence infrastructure by the five leading U.S. cloud computing companies will reach $1.2 trillion by 2027, compared to $800 billion in the current year, according to a Bloomberg report on Friday, September 25. The forecast from Goldman Sachs exceeds the consensus estimate of $1.1 trillion for the following year.

The five companies included in the projection are Amazon, Alphabet, Microsoft, Oracle, and Meta, as reported. Goldman Sachs strategists, headed by Ryan Hammond, noted that the capital expenditures of the hyperscalers in 2027 will constitute "a larger share of GDP than any technological investment cycle since the railroad expansion in the late 1800s," according to the report.

The forecast anticipates a 12% rise in 2028, bringing the total to $1.4 trillion. JPMorgan Chase Chairman and CEO Jamie Dimon told CNBC on September 21 that hyperscalers may spend $1 trillion next year, up from $700 billion this year and $300 billion in 2022. He stated that this spending represents a "1% increase to GDP each year" and could contribute to inflation due to hiring, factory construction, equipment purchase, copper wire acquisition, and power plant/building.

Nonetheless, Dimon believes that the technology holds significant potential for reducing deflationary factors down the line. Nvidia's CEO, Jensen Huang, stated during a May earnings call that hyperscaler AI infrastructure spending alone is expected to surpass $1 trillion in 2027, with the potential to reach $3 trillion to $4 trillion annually by the end of the decade.

Huang emphasized that "compute is revenue. Compute is profit." Citigroup projected in September 2025 that AI-related infrastructure spending by Big Tech could exceed $2.8 trillion by 2029, citing early investments by hyperscalers and increasing demand for enterprise AI applications. The bank noted that Big Tech firms are now utilizing debt financing to meet the growing demand for AI infrastructure.