Wall Street finally wants to see a return on the biggest capital spending spree in tech history.
The Summary
- Microsoft and Meta report earnings Wednesday, facing investors who are losing patience with massive AI capex and shrinking cash reserves
- Microsoft is the key company to watch this earnings season, according to DA Davidson analyst Gil Luria, because AI spending and capex levels will set the tone
- The question isn't whether AI is the future anymore—it's whether these companies can justify burning billions before the revenue shows up
The Signal
Big Tech's AI spending binge has become a balance sheet problem. Microsoft and Meta are testing how much longer investors will tolerate "profligate spending" on AI infrastructure while cash positions dwindle. This isn't skepticism about AI's potential. It's skepticism about the timeline between writing checks and cashing them.
The earnings calls Wednesday will reveal whether these companies are still accelerating AI capex or starting to show restraint. More importantly, they need to demonstrate revenue conversion, not just model benchmarks and developer adoption metrics.
"Microsoft will be the key company to watch during this earnings season and it all has to do with spending on artificial intelligence and capex."
Here's what makes this moment different from past tech spending cycles:
- The capital requirements for frontier AI are orders of magnitude larger than previous infrastructure buildouts
- Investor patience is finite when cash reserves shrink quarter after quarter
- The gap between AI investment and AI revenue is widening, not closing
DA Davidson's focus on Microsoft makes sense. Microsoft has Azure, GitHub Copilot, and enterprise AI products already generating revenue. If they can't show a clear path to ROI on AI spending, the thesis gets harder for everyone else. Meta's position is different: they're building infrastructure for products that don't exist yet, betting on a future where AI agents are as common as mobile apps.
The market is shifting from "how much are you spending on AI?" to "what are you getting for it?" That's a healthy correction. The AI buildout was always going to hit this inflection point where investors demand proof of concept at scale, not just demos and developer interest.
The Implication
Watch the guidance on capex for the rest of 2026. If Microsoft and Meta signal continued aggressive spending without corresponding revenue acceleration, expect a broader repricing of AI infrastructure plays. The companies that can demonstrate unit economics and customer adoption will separate from those still pitching potential.
For builders in the agent economy: this earnings scrutiny creates opportunity. As Big Tech faces pressure to show returns, they'll be more selective about what they build in-house versus what they buy or partner on. Specialized AI tooling companies with real traction just became more attractive acquisition targets.