From Buyback Machines to AI Infrastructure Builders
Big Tech’s AI infrastructure spending refers to the massive shift of cash flow from stock buybacks and dividends into data centers, chips, and cloud capacity needed to run advanced AI workloads at scale. For most of the past decade, the largest hyperscalers offered a clear deal: strong growth plus steady share repurchases that supported earnings per share and gave investors technical support when markets turned. That era is fading. Goldman Sachs figures cited by MarketWatch show the major AI hyperscalers are expected to spend about USD 755 billion (approx. RM3.48 trillion) on capital expenditures in 2026, an 83% jump from the prior year and spread across Amazon, Alphabet, Meta, Microsoft and Oracle. This is no longer an incremental tech upgrade cycle; it is a structural reallocation of cash that is changing what shareholders effectively own.
The AI Capex Arms Race and Hyperscaler Investment Plans
The AI capex arms race is now the defining feature of Big Tech capital expenditure. Amazon has outlined roughly USD 200 billion (approx. RM920 billion) of capital spending for 2026, tying it directly to AI infrastructure, custom chips, data centers and robotics. Analysts have guided Alphabet into the USD 175 billion to USD 185 billion (approx. RM805 billion to RM851 billion) range, while Meta has lifted its 2026 capex guidance to USD 125 billion to USD 145 billion (approx. RM575 billion to RM667 billion), citing higher infrastructure costs and memory pricing. Microsoft is spending above the USD 88.2 billion (approx. RM405 billion) it recorded in fiscal 2025. According to the Wall Street Journal, free cash flow for the five leading hyperscalers is expected to drop 91% in 2026 to about USD 16 billion (approx. RM73.6 billion) even as net income climbs, underscoring how cash is being pulled into AI buildouts.
Stock Buyback Alternatives: When Cash Meets Concrete and Silicon
This surge in AI infrastructure spending is directly crowding out stock buyback programs. Goldman Sachs data reported by MarketWatch show buybacks by hyperscalers fell by nearly two-thirds in the first quarter, with Microsoft a notable outlier. Alphabet bought back no stock in its latest quarter after repurchasing about USD 15.1 billion (approx. RM69.5 billion) in the same period a year earlier. Amazon has not been a regular repurchaser, and Meta is pushing more cash into infrastructure. Instead of returning capital through buybacks, these firms are choosing stock buyback alternatives such as long-dated data center leases, power contracts, custom silicon orders and, increasingly, debt financing. For investors who once counted on shrinking share counts, the investment thesis is shifting toward a longer, riskier payoff tied to AI revenue that must eventually cover hundreds of billions of depreciating assets.
Microsoft’s Legal Headwinds Expose AI Spending Tensions
Microsoft’s latest earnings shock shows how fragile confidence can be when AI capex outpaces clear disclosure. The company reported USD 51.5 billion (approx. RM237 billion) in cloud revenue for fiscal Q2 2026, yet its stock fell about 10%, erasing an estimated USD 357 billion (approx. RM1.64 trillion) in market value in a single day. A securities class action filed by the City of St. Clair Shores Police and Fire Retirement System alleges Microsoft softened the reality of capacity constraints and the scale of its AI infrastructure spending. Capital expenditures hit USD 37.5 billion (approx. RM172.5 billion) in one quarter, up 66% year on year and above analyst expectations, while gross margin slipped to just over 68%, its lowest in about three years. The suit argues investors were not adequately warned how aggressively capacity was being diverted toward AI and OpenAI-linked workloads.
A Structural Reset in Big Tech Capital Allocation
Behind the headlines, the AI capex boom signals a structural reset in how hyperscalers balance growth, competition and shareholder value. Free cash flow is taking the hit first: Amazon’s 12‑month free cash flow through March 31, 2026 fell to USD 1.2 billion (approx. RM5.52 billion) from USD 25.9 billion (approx. RM119.1 billion) a year earlier, driven by a USD 59.3 billion (approx. RM273 billion) jump in property and equipment purchases. Accounting smooths the optics because hardware is depreciated over years, but the cash leaves now. As buybacks shrink, valuations lean harder on AI revenue materializing fast enough to justify sunk costs in GPUs, memory and data centers. Founders and partners dealing with these clouds are facing companies where every new AI commitment competes with power contracts, debt service and existing infrastructure, reinforcing that the era of easy, cash-rich buybacks is over.
