The latest financial reports from major tech companies reveal a continued increase in capital expenditure focused on artificial intelligence (AI). Microsoft (MSFT, Financial), Google (GOOGL), and Meta (META) reported a combined year-over-year growth in capital spending of 48%, indicating an unwavering commitment to AI investment. However, the return on these massive investments remains mixed, suggesting that the market may need to wait longer for significant results. Key trends include a shift towards inference over training, and Microsoft’s strategy in reducing OpenAI's importance.
Google's third-quarter capital expenditure reached $13 billion, driven by robust spending on AI servers. This figure was slightly above market expectations, with the company forecasting similar spending levels in the fourth quarter of 2024 due to a projected 60% year-on-year increase. The management anticipates significant growth in AI infrastructure investment in 2025.
Microsoft reported third-quarter capital expenditure of $15 billion, boosted by the acceleration of its H200 AI server project and increased infrastructure investment. Next quarter, capital spending is expected to rise by 32% year-over-year, supported by investments in Azure and AI infrastructure.
Meta's capital expenditure for the third quarter was $8.3 billion, falling short of Bloomberg's forecast. However, the company revised its 2024 spending guidance to between $38 billion and $40 billion, up from a midpoint growth of 41%. This adjustment is primarily fueled by AI infrastructure investments.
Amazon (AMZN) has also increased its spending, with third-quarter capital expenditure on real estate and equipment reaching $22.6 billion, an 81% rise from the previous year. For 2024, Amazon expects capital spending to grow by 55% to $75 billion, mainly driven by its AWS cloud division.
Despite the optimism in capital expenditure and robust core and AI business prospects, the return on AI investments remains a topic of concern. Dialogue-based AI services show some positive growth, but not at the anticipated levels. Microsoft noted a slight decline in Azure's growth, not due to reduced client demand but to ongoing capacity constraints. Analysts foresee accelerated Azure growth beginning in the second half of 2025 as more AI servers become operational.
Meanwhile, Google's cloud computing division reported accelerating growth with a 35% year-over-year increase in revenue, driven by strong core and AI Google Cloud Platform demand. However, some factors unrelated to AI contributed, such as contracts requiring clients to spend a minimum amount on server rentals.
Three emerging trends were highlighted: first, companies are focusing more on purchasing technology rather than hiring, as seen with Amazon's layoffs since 2022. Second, the emphasis is shifting from AI training to inference, with Microsoft predicting $10 billion in annual AI revenue from inference-driven services. Lastly, Microsoft is reducing OpenAI's prominence by introducing alternative models like Claude 3.5 Sonnet and Gemini 1.5 pro, signaling a strategic pivot akin to past cloud computing advancements.
