Recommended Articles
Markets & Policy

The Hyperscaler Earnings Week Ahead: What Founders Should Read Into AI Capex Scrutiny

By Angeled · October 25, 2026 · Markets & Policy
The Hyperscaler Earnings Week Ahead: What Founders Should Read Into AI Capex Scrutiny
The Hyperscaler Earnings Week Ahead: What Founders Should Read Into AI Capex Scrutiny

The week of October 26, 2026 is the earnings season's centre of gravity. Alphabet and Microsoft report after the market close on Wednesday, October 28, and Amazon follows after the close on Thursday, October 29. Eli Lilly reports before the open on October 29, and Exxon Mobil and Chevron report before the open on Friday, October 30. Caterpillar also appears in earnings listings for this window in some calendars, while other calendars place it on November 4 — a reminder to check the company calendar rather than relying on a single listing. Roughly one-third of S&P 500 members report in the same week, and one market tracker puts about 44% of the US index's market value reporting across these five days.

For founders, the question hanging over the hyperscaler reports is not whether AI demand is growing. It is whether the enormous spending on AI infrastructure is turning into revenue, margin, and cash flow fast enough to justify itself. That question now shapes funding, hiring, and pricing decisions far beyond the hyperscalers themselves.

The setup: high expectations, sharper questions

S&P 500 companies are on track to increase third-quarter earnings by more than 32%, according to LSEG IBES data cited by Reuters, yet the market wobbled in the week before these reports. The Nasdaq Composite posted its biggest one-day drop since June on Thursday, October 23, as investors questioned the financing costs behind heavy spending on AI data centres. Reuters described Microsoft, Alphabet, and Amazon as three AI hyperscalers whose capital-spending forecasts will be closely watched, and the Federal Reserve's policy meeting lands in the same week, adding a rate backdrop to a crowded calendar.

Consensus figures ahead of the reports put Microsoft at $4.71 earnings per share with Azure growth near 45%, Alphabet at $2.94, and Amazon at $1.96. These are previews, not results, but they show where attention will fall: cloud growth, capex, depreciation, and free cash flow, not just headline earnings.

The capex numbers behind the scrutiny

Guidance cited in previews puts combined 2026 capital expenditure for Amazon, Alphabet, Microsoft, and Meta at roughly $720-$745 billion, with Amazon alone at about $220 billion. Recent cash-flow signals underline the questions: Alphabet reported negative free cash flow of $5.9 billion in Q2 2026, and Amazon's trailing-twelve-month free cash flow was an outflow of $7.6 billion. Meta guided to $135-$145 billion in 2026 spending with negative quarterly free cash flow of about $2.12 billion.

Analysis: these figures do not prove overspending — data centres are long-lived assets, and companies measure capex differently. But they explain the mood shift: investors now ask not just "how fast is cloud growing?" but "how much of that growth is profitable, and how is it financed?" Founders should expect the same questions at their own scale.

What founders should listen for

First, listen for cloud growth versus capex growth. If cloud revenue grows strongly but capex grows faster, the market will ask how long that gap can persist. For a startup, the analogue is revenue growth versus infrastructure cost growth: a company whose compute bill grows faster than its revenue is on a path that needs a correction, whether through efficiency, pricing, or a smaller model.

Second, listen for inference economics. Jassy has said AI inference remains roughly ten times too costly for widespread business use, and that Amazon's Trainium chips offer 30-40% better price-performance than comparable processors. Whether or not those specific claims hold up in the earnings calls, the direction is clear: the hyperscalers are competing on cost per query, not just on capability. Startups that compete on capability alone, without a cost story, will find it harder to hold margin as inference prices fall.

Third, listen for what the non-tech reports say about the real economy: Eli Lilly on healthcare demand, Exxon and Chevron on energy prices and capital discipline, and Caterpillar on industrial demand. Founders selling into those sectors should read them as pipeline signals.

Practical takeaways before the prints land

Do not make irreversible decisions on preview numbers. Wait for the transcripts, capex guidance, and free-cash-flow commentary, and compare guidance with prior guidance, not just consensus.

For fundraising, prepare a capex story of your own: how much compute you need, how that need scales with revenue, and where efficiency gains come from. For product, revisit pricing: if inference costs are falling but still high, usage-based pricing and efficiency features are not optional extras. For hiring, note that a scrutiny phase tends to favour teams that can show unit economics, not just growth.

The hyperscaler earnings week will not settle the AI investment debate in one set of prints. But it will set the terms of that debate for the next quarter — and founders who listen for capex, cash flow, and cost per query, rather than just for revenue beats, will hear it first.

Angeled.com connects founders with investors.

Sources: https://alphabetastock.com/microsoft-alphabet-amazon-earnings-preview/ ; https://www.tradingkey.com/analysis/stocks/us-stocks/262237761-weekly-fed-microsoft-alphabet-amazon-earnings-tradingkey ; https://globaleconomycharts.com/2026/10/25/ai-big-tech-earnings-cash-flow-2026/ ; https://www.worldmarketstats.com/earnings-calendar.html

Read Next