Resolution criteria

This market resolves based on the combined total capital expenditure (capex) reported by Google (Alphabet), Amazon, Meta, and Microsoft for calendar year 2027. Resolution will use official financial statements and SEC filings from each company, with capex defined as cash capital expenditures (excluding finance leases unless explicitly included in company guidance). The market resolves YES if the combined total exceeds the specified threshold, and NO if it does not.

Resolution sources:

• Estimated spending is $427 billion in 2025, $620billion in 2026, and $637 billion in 2027 according to RBC Wealth Management's latest projections

• Official 10-K filings and earnings call transcripts from each company

• SEC EDGAR database for verified financial data

Background

Alphabet, Meta, Microsoft and Amazon each lifted their guidance for capital expenditures and now collectively expect that number to reach more than $380 billion this year (2025). In all, Amazon, Microsoft, Meta, and Google plan to spend about $650 billion in 2026, representing a dramatic acceleration in AI infrastructure investment.

Goldman Sachs sees hyperscaler capex increasing sharply through 2027 – capex is projected to be $1.15 trillion from 2025 through 2027, more than double the $477 billion spent from 2022 through 2024. The spending surge is driven primarily by data center buildouts for AI model training and inference, with approximately 75% of aggregate hyperscaler capex in 2026 will fund AI-related infrastructure.

Considerations

Analyst estimates for 2027 capex remain uncertain and subject to revision. Consensus capex estimates have proven to be too low for two years running. At the start of both 2024 and 2025, consensus estimates implied capex growth of roughly 20% for the year. In reality, it exceeded 50% in both years. Companies may adjust spending based on AI monetization progress, energy constraints, or macroeconomic conditions. Additionally, the medium term carries significant risks of overcapacity. If AI application rollouts prove slower than expected, the industry could face a glut of underutilized computing power by 2027.

This description was generated by AI.