Amazon reports earnings after-hours on February 5. Alphabet already reported on February 4. Together, these results carry outsized importance for every AI infrastructure position in the market.
If you’re trading cloud stocks (SKYY), positioned in AI hardware (NVDA, SMCI), or exposed to mega-cap tech (QQQ), this is the moment that resolves the biggest question of Q1 2026:
Is massive AI capital spending converting into profitable demand - or is infrastructure being built faster than customers can monetize it?
Over the past week, markets absorbed a series of conflicting signals. Oracle announced plans to raise $45–50 billion for AI cloud expansion. Gartner collapsed 20% after warning that enterprises are “deferring everything possible” on AI spending. Super Micro Computer raised guidance to $40 billion on record GPU orders. Palantir posted 70% AI revenue growth, driven largely by government contracts.
These datapoints cannot all coexist indefinitely.
AWS and Google Cloud earnings decide which narrative breaks.
What Actually Happened
From February 1-5, investor focus narrowed to two earnings events: Alphabet reporting February 4 after-hours, and Amazon scheduled for February 5 after-hours. In both cases, the spotlight is squarely on their cloud segments - Google Cloud and AWS - where analysts are looking for proof that AI infrastructure utilization is translating into real revenue growth and sustainable margins.
For Amazon, expectations center on 25-28% AWS revenue growth and whether margins remain intact. Historically, AWS has operated at 28-32% operating margins, generating the bulk of Amazon’s operating income despite being a minority of total revenue. That profitability funds everything else Amazon does. If AI infrastructure capex — now estimated at $50-60 billion annually - pushes AWS margins down toward 24-26%, Amazon’s overall profit engine weakens even if revenue grows.
For Alphabet, the focus is different but equally critical. Google Cloud has spent years scaling near break-even. Analysts expect 30–35% revenue growth and are looking for clear evidence that operating margins can move toward 10–15%as AI workloads improve pricing power. If Google Cloud fails to show margin inflection, it raises doubts about whether AI cloud infrastructure can ever deliver software-like returns.
The timing makes these earnings uniquely powerful. They arrive within 72 hours of Oracle’s capital raise announcement, SMCI’s guidance increase, and Gartner’s enterprise spending warning — giving markets immediate confirmation or rejection of whether infrastructure spending is aligned with demand.
Why This Matters
These earnings will reprice far more than Amazon or Alphabet stock. They will reset expectations for the entire AI infrastructure complex.
If AWS reports 28%+ revenue growth with margins holding above 28%, it validates the bullish thesis across the board. Oracle’s $50B raise looks strategic rather than defensive. SMCI’s guidance appears sustainable rather than peak-cycle. Palantir’s growth begins to look transferable beyond government use cases. In that scenario, cloud and AI infrastructure stocks could rally 3–5% in short order.
The bearish scenario carries more weight. If AWS shows sub-22% growth or margin compression below 26%, it breaks multiple assumptions at once. It would imply that AI workloads are cannibalizing existing cloud revenue rather than adding incremental demand — a dangerous outcome when capital intensity is rising. That scenario would likely pressure cloud and infrastructure stocks 8–12%, validating Gartner’s warning that enterprise AI spending isn’t materializing fast enough to absorb $200+ billion in annual industry capex.
The less obvious but critical detail is who is driving demand. AWS and Google Cloud results will reveal whether AI infrastructure is being used primarily for customer workloads (enterprise AI, AI labs) or internal workloads (retail optimization, advertising models, Gemini/Alexa). Internal usage does not create durable, recurring cloud revenue. Customer usage does.
Watch management commentary on GPU utilization and AI workload mix. Utilization above 70% across diverse customers signals healthy demand. Utilization below 50%, or concentration in internal projects, signals speculative buildout.
What to Watch Next
For the February 4-5 earnings window, five metrics determine whether AI infrastructure economics work:
- AWS Revenue Growth & Margin (Amazon, Feb 5)
Growth above 25% with margins holding above 28% confirms success. Below 22% growth or margins under 26% signals trouble. - Google Cloud Operating Income (Alphabet, Feb 4)
Movement toward 10–15% margins with strong growth confirms scaling. Flat or widening losses question long-term economics. - 2026 Capital Expenditure Guidance
Rising capex signals confidence in demand. Flat or reduced guidance signals weakening visibility. - Enterprise vs Startup Customer Mix
Enterprise-driven demand is sticky and sustainable. Startup-driven demand is volatile and funding-dependent. - Geographic Cloud Growth
Expansion beyond the U.S. into Europe and Asia validates a multi-year runway. U.S.-only growth limits upside.
The Cross-Story Connection
AWS and Google Cloud earnings act as the resolution mechanism for every conflicting AI signal this week.
Infrastructure suppliers sell servers to hyperscalers.
Hyperscalers deploy data centers and GPUs.
Customers must rent that compute at scale for revenue to flow back.
Gartner’s warning breaks the final step. Palantir and SMCI suggest it still works for government and hyperscale-backed buyers. AWS and Google Cloud will show which customer type dominates — and whether that demand is sufficient to justify the spending.
The Question Traders Should Ask
Do these earnings prove that AI infrastructure spending converts into profitable, recurring revenue — or do they show that the industry is building capacity ahead of monetization?
If AWS and Google Cloud validate margins, infrastructure strength continues.
If they disappoint, the market pivots toward defense and capital discipline.
Either way, these earnings don’t just report results - they decide the AI narrative for the next quarter.