AI Markets · · 7 min read

Big Tech Sheds 23,000 Workers While Doubling Down on $700B AI Capex

Meta and Microsoft workforce cuts expose the paradox driving Tech's earnings season: can infrastructure spending at industrial scale justify valuations when free cash flow is collapsing?

Meta and Microsoft announced approximately 23,000 combined workforce reductions this week while simultaneously accelerating AI infrastructure spending toward $700 billion across Big Tech in 2026, crystallising the sector’s defining tension between operational efficiency and capital intensity.

Meta disclosed a 10% workforce reduction—approximately 8,000 employees—effective May 20, plus 6,000 unfilled roles left open, according to CNBC. Microsoft offered voluntary retirement buyouts to 8,750 U.S. employees under a ‘sum 70’ formula combining age and tenure, the company’s first-ever program of its kind. Both announcements came within 48 hours ahead of April 29 Earnings calls, timing that amplifies investor scrutiny on whether record capital expenditure translates to revenue growth and margin recovery.

Big Tech AI Capex Landscape
2026 Combined Spending (Meta, Microsoft, Amazon, Alphabet)
$700B
Meta 2026 Capex Guidance
$115-135B
Microsoft Q2 FY2026 Capex
$37.5B
Tech Layoffs Year-to-Date
92,000

The workforce reductions arrive as Big Tech transforms from asset-light software businesses into capital-intensive infrastructure operators. Meta’s capex guidance of $115-135 billion for 2026 nearly doubles the $72 billion spent in 2025, per Meta Investor Relations. Microsoft reported $37.5 billion in capex for Q2 FY2026 alone—a quarterly figure that exceeds many companies’ annual budgets—putting the company on track for $120-145 billion full-year, according to Microsoft Investor Relations.

The Free Cash Flow Reckoning

Capital intensity at this scale produces immediate effects on financial flexibility. Amazon is projected to generate negative free cash flow of $17-28 billion in 2026 due to $200 billion in capex, while Alphabet’s FCF is expected to plummet 90% to $8.2 billion from $73.3 billion in 2025, according to CNBC citing Morgan Stanley and Pivotal Research estimates. Mizuho analysts noted that the potential doubling of capex leaves ‘limited FCF with uncertain return on investment’.

Microsoft’s capex as a percentage of revenue has risen to its highest level in over a decade, data from RBC Wealth Management shows. The shift marks a structural departure from the asset-light software model that drove two decades of margin expansion. Forward P/E multiples reflect the tension: Microsoft trades at roughly 1.2x versus a 1.4x ten-year average, compression that mirrors investor concern over returns on deployed capital.

“If you’re going to pour all this money into AI, it’s going to reduce your free cash flow.”

— Jake Dollarhide, CEO, Longbow Asset Management

The labour cuts serve dual purposes: immediate margin relief and narrative cover. Meta’s workforce reduction follows aggressive automation initiatives enabled by internal AI systems. Microsoft’s voluntary program targets employees whose age plus tenure exceeds 70, a demographic profile that correlates with higher compensation and legacy skillsets less aligned with AI-centric operations.

Revenue Conversion Pressure

Investor patience for AI spending on faith alone is exhausting. Microsoft’s 365 Copilot—the flagship monetisation vehicle for Azure AI Infrastructure—has achieved just over 3% adoption among 450 million 365 users, per Channel IAM. That adoption rate generates annualised revenue in the low single-digit billions against infrastructure spending approaching $140 billion, a payback horizon measured in decades at current trajectories.

Market reaction to the announcements reveals the operative logic: Meta’s stock rose 4.2% and Microsoft gained 3.1% in post-announcement trading, according to AI Business Review. Wall Street rewards operational discipline even as it underwrites capital intensity, a combination that squeezes headcount while expanding data centre footprints.

Capex vs Free Cash Flow Trajectory
Company 2026 Capex Estimate FCF Impact
Amazon $200B -$17-28B (negative)
Alphabet ~$150B $8.2B (-90% YoY)
Meta $115-135B Compressed
Microsoft $120-145B Under pressure

The Substitution Thesis

Dan Ives of Wedbush Securities framed the cuts as enabling companies to “automate tasks that once required large teams, allowing the company to streamline operations and reduce costs while maintaining productivity,” per Al Jazeera. This substitution narrative—replacing human payroll with machine capex—offers political cover for a structural bet: that capital deployed at industrial scale will eventually generate returns justifying current valuations.

The timing ahead of earnings calls is deliberate. Analysts will press management on three variables: capex guidance for the second half of 2026, revenue conversion timelines for AI products, and margin trajectories under continued infrastructure buildout. Over 92,000 tech workers have been laid off in 2026 year-to-date, with nearly 900,000 exits since 2020, data from Layoffs.fyi via CNBC shows.

Context

Big Tech capex estimates have been revised upward repeatedly. Initial 2025 projections of $250 billion for 2026 have climbed to $562 billion in current consensus, according to Bloomberg data cited by RBC Wealth Management. The upward revisions reflect both competitive dynamics—no hyperscaler can afford to cede infrastructure advantage—and technical reality: training frontier models and serving inference at scale requires compute capacity that doubles annually.

What to Watch

The April 29 earnings calls will reveal whether management teams maintain or revise capex guidance in response to margin pressure. Key indicators: any language around ‘optimising’ or ‘rightsising’ capital deployment, specific revenue figures from AI products rather than aggregate growth rates, and commentary on whether labour substitution has reached its natural limit or represents the opening phase of a multi-year restructuring.

Longer horizon risks center on ROI timelines. Analysis from WNDYR shows 61% of business leaders report heightened pressure to demonstrate AI returns, while 53% of investors expect positive results within six months—expectations that align poorly with infrastructure payback periods measured in years. If capex continues expanding while revenue conversion stalls, the current workforce reductions may prove the opening act in a broader operational reset across the sector.