Meta Vs. Oracle: The Laggards of AI Capex Payback Velocity

Meta Platforms and Oracle are facing heightened investor scrutiny over the velocity of returns on their massive artificial intelligence capital expenditures. Meta reported a 27.96% year-over-year revenue increase to $60.80 billion for Q2 2026, yet missed earnings expectations with diluted EPS of $6.18 versus $7.22 projected. Total quarterly costs surged 55% to $42.03 billion, while capital expenditures reached $30.12 billion, causing free cash flow to drop over 91% to $784 million. Meta's full-year 2026 capex is projected between $130 billion and $145 billion. In comparison, Oracle posted a massive $664 billion backlog alongside 121% IaaS revenue growth to $7.39 billion in Q1 FY2027, but incurred $28.5 billion in quarterly capex resulting in negative $5 billion in free cash flow. While Oracle monetizes AI by renting compute infrastructure to third parties, Meta embeds AI internally into its core digital advertising platform, driving a 14% increase in ad impressions and a 12% increase in average price per ad. Despite near-term cash flow compression, Meta is viewed as having a faster internal monetization mechanism through its core advertising flywheel.