Goldman Sachs estimates AI infrastructure spending across three phases: $633 billion from 2023 to 2025, $1.73 trillion from 2026 to 2027, and $4.14 trillion from 2028 to 2030. To break even, the six tracked hyperscalers (Alphabet, Microsoft, Amazon, Meta, Oracle, and SpaceX) would need approximately $300 billion in annual AI revenue, while $1 trillion annually would be required for satisfactory returns. AI infrastructure capex is projected at $800 billion in 2026, $1.2 trillion in 2027, and $1.4 trillion in 2028, figures above current Wall Street consensus estimates. The combined backlog for AWS, Azure, and Google Cloud reached approximately $1.69 trillion in Q2 2026, a 152% year-over-year increase. Using a 15% return on invested capital benchmark, Goldman Sachs estimates these companies would need to generate roughly $1.42 trillion in cumulative revenue between 2028 and 2030.
Source: Read the original article
