brain/
← all entities
entityAMZNstock-market

Amazon

Notes

Amazon

One-line summary: Amazon (AMZN) is the AWS-hyperscaler whose Q2 2026 print became the key refinement of the ai-roi-reckoning FCF-discrimination rule — FCF-negative and spending ≥$200B on capex, yet rewarded (+8–9%) because AWS reaccelerated to +37% with +64% operating income, proving the discriminator is visible ROI on the capex, not the sign of FCF.

What it is

Amazon operates AWS (cloud/AI infrastructure), e-commerce, advertising, and devices. For this project it matters primarily as one of the four hyperscalers whose capex fills the AI-infrastructure demand that sells HBM/CoWoS/power — and as a live test of the market's reaction function to AI capex.

Why it matters to stock-market

Amazon's Q2 2026 reaction is the load-bearing datapoint refining ai-roi-reckoning: the market does not simply punish FCF-negative capex raisers. It rewards capex whose incremental return is visible in accelerating high-margin revenue and punishes capex where the return is not yet visible. AWS reacceleration is the read-through demand for the asset-heavy consumption-compute leg (hbm-cowos-as-binding-bottleneck).

Key facts (Q2 2026 — 2026-07-30)

  • Revenue $200.6B (+20% YoY); operating income +43% to $27.5B; EPS $5.75 (vs ~$1.81 est).
  • AWS +37% YoY to $42.2B — fastest growth in ~18 quarters; operating income +64% to $16.6B (~39% margin).
  • FCF negative −$7.6B (TTM, vs +$18.2B a year earlier); TTM operating cash flow $161.4B < TTM capex ~$169B; Q2 capex $54.2B.
  • 2026 capex ≥$200B (one source reports a raise to ~$220B — disputed). Jassy: "Even at that amount, we will still not have enough capacity to meet all the demand we have in 2026, and I believe this dynamic will also be true in 2027 too."
  • Reaction: +8–9% after hours (~$257) — the market rewarded demonstrated AWS ROI over the FCF drain.

Corroboration — the cloud-reaccel read (Compound, 2026-07-31)

The Compound & Friends panel, taping as AMZN reported, put the AWS reacceleration alongside Azure and Google Cloud as the same signal — the cloud is the unambiguous AI winner, and AWS is the least model-dependent of the three:

  • michael-batnick in 2026-07-31-podcast-the-compound-and-friends-why-demand-for-compute-is-about-to-explode-with: "the cloud is a clear winner from AI" — AWS net sales +20% to $206B; AWS +36.7% YoY, "our fastest growth in 18 quarters"; AWS exceeded a $25B annual revenue run-rate for its AI business. Alongside Azure +43% (highest since 2022) and Google Cloud +82%.
  • alex-kantrowitz in the same source on AMZN's strategy: "basically do very little, build a lot of compute, profit" — the bring-your-own-model / most-of-the-world's-compute-to-license position is what makes AWS less exposed than Oracle (OpenAI-dependent) to any single lab. Note AMZN trailing FCF was flagged as −$7.6B, "and the market seems to be able to look past it."

Related

Sources

Referenced by