This analysis is part of Amazon’s AI Business Model Pivot, a deep dive by The Business Engineer.

AWS’s value proposition was “elastic compute and storage”—pay for what you use, scale infinitely. It’s becoming “enterprise AI workforce”—agents that perform work previously requiring human labor.
The Fundamental Shift
Before: Elastic Infrastructure. Pay for compute hours used. Scale up/down on demand. Compete on price per unit. Commodity infrastructure. Competition was a race to the bottom on pricing.
After: Enterprise AI Workforce. Pay for work completed. Agents work 24/7 autonomously. Compete on value per outcome. Differentiated AI workforce. Competition shifts to value capture.
The Agent Product Portfolio
The product lineup tells the story: Kiro (Autonomous Developer, 200K+ developers), Quick Suite (Business AI Teammate), Security Agent (Virtual Security Engineer), DevOps Agent (Virtual SRE), and Connect (AI Contact Center, $1B ARR).
The New Unit Economics
Old model: Revenue = Hours × $/hour. New model: Revenue = Tasks × Value/task. The customer buying decision shifts from “How much compute do I need?” to “How many tasks can agents complete?” and from “What’s my monthly AWS bill?” to “What’s my cost per outcome?”
Instead of competing on price-per-compute, AWS competes on value-per-outcome. This fundamentally changes the unit economics and captures more value per customer.









