What the AI spending boom means for Big Tech
Big Tech’s AI spending boom is a rapid and sustained surge in capital poured into AI infrastructure—chips, data centers and networks—that is large enough to reshape profits, shareholder returns and employment structures across the technology sector. This shift is driven by hyperscalers that now treat AI infrastructure investment as their primary use of cash, even ahead of stock buybacks. Goldman Sachs expects the largest hyperscalers to spend USD 755 billion (approx. RM3.5 trillion) on capital expenditures in 2026, an 83% increase from the previous year. That figure covers Amazon, Alphabet, Meta, Microsoft and Oracle, and makes clear that AI is no longer an optional side project. It has become the core of hyperscaler capital allocation, forcing boards and investors to accept lower immediate cash returns in exchange for long-term AI growth.
Hyperscaler capital allocation and the end of buybacks
The AI infrastructure investment wave is rewriting how hyperscalers allocate capital. For years, Amazon, Alphabet, Meta and Microsoft balanced growth with generous buybacks that supported earnings per share and signaled confidence. That balance is breaking. According to analysis cited by MarketWatch, hyperscaler buybacks fell by nearly two-thirds in the first quarter, even as capex guidance surged. Alphabet bought back no stock after repurchasing about USD 15.1 billion (approx. RM69.5 billion) in the same period a year earlier, and Meta is pushing more cash into infrastructure. Amazon has mapped a huge capital spending plan tied to AI infrastructure, custom chips, data centers and robotics, while Oracle’s inclusion in the USD 755 billion bucket underlines how broad the race has become. Shareholders now own businesses that prioritize AI buildouts over shrinking share counts, tying valuations more tightly to future AI revenue.
Cash flow, AI capex and the new risk for investors
The most important change for investors is not the headline growth story but the cash flow strain behind it. The Wall Street Journal reported that free cash flow for the five big hyperscalers is expected to drop 91% in 2026 to about USD 16 billion (approx. RM73.6 billion), even as net income is projected to rise 25% to USD 506 billion (approx. RM2.3 trillion). That gap captures the AI buildout: cash is leaving now, while accounting spreads costs over years. Nvidia and other suppliers record revenue immediately, but hyperscalers depreciate hardware slowly, creating a period where reported profits look strong even as cash is tight. With buybacks reduced, investors lose a key cushion that once supported share prices during volatility. Their returns now depend on whether AI services and cloud demand can ramp fast enough to pay for hundreds of billions in infrastructure.
Tech layoffs and the efficiency narrative at Robinhood
While hyperscalers redirect billions into AI infrastructure, other tech firms are reshaping their workforce to match an automation-first future. Trading platform Robinhood plans to cut about 10% of its staff, eliminating roughly 290 full-time roles from a base of around 2,900 employees, even as executives say the business has “never been stronger.” The company frames the move as a way to streamline operations, flatten management and maintain a “leaner organizational structure” to support growth, with around USD 28 million (approx. RM128.8 million) in expected restructuring costs. This decision fits a wider wave of tech layoffs 2026, where companies cite efficiency, management simplification and, often, AI as reasons they can maintain or raise output with smaller teams. As Oliver Voros of Gooseberry AI notes, more than half of layoffs this year reference AI-driven restructuring, even when it is not mentioned in official announcements.
Automation, jobs and the new shareholder bargain
The timing of workforce cuts alongside record AI infrastructure spending suggests a deeper shift in what roles companies see as worth paying for. AI is not erasing jobs overnight, but it is changing the mix: routine work and some specialist functions are being consolidated or automated, while new roles cluster around AI model training, data engineering and platform operations. For employees, that means rising pressure to move toward tasks that complement AI tools rather than compete with them. For shareholders, expectations are changing too. The old bargain of high growth plus steady buybacks is giving way to a new deal: accept weaker capital returns now in exchange for exposure to large-scale AI platforms. Whether this pays off depends on execution. If AI services fail to deliver strong margins, investors could face the worst mix—shrinking buybacks, thinner cash cushions and a workforce unsettled by repeated restructuring.
