Stratechery argues that reasoning models have reintroduced marginal costs to AI services, breaking the zero-cost-scaling assumption that underpinned internet-era tech economics. This forces hyperscalers to optimize compute allocation based on opportunity costs rather than marginal costs, reshaping how companies like Microsoft, Amazon, and Google balance customer services against internal AI products and strategic investments.
Strategy
Mythos, Muse, and the Opportunity Cost of Compute
Reasoning models reintroduce marginal costs to AI services, shattering the zero-cost-scaling assumption that defined internet economics and forcing hyperscalers like Microsoft, Amazon, and Google to prioritize compute allocation by opportunity cost rather than abundance.
Monday, April 13, 2026 12:00 PM UTC2 MIN READSOURCE: StratecheryBY sys://pipeline
Tags
strategy
/// RELATED
InfrastructureApr 22
NVIDIA and Google Cloud Collaborate to Advance Agentic and Physical AI
NVIDIA and Google Cloud cut agentic AI inference costs by 10x with new A5X GPU instances, pairing Vera Rubin compute with Gemini and Nemotron for enterprise deployment at scale.
StrategyApr 22
Elon Musk thinks college is ‘basically for fun’—but his former Tesla HR chief tells Gen Z even their liberal arts degrees are more valuable than ever
As early-career job postings drop 16% year-over-year, Tesla's former HR chief argues liberal arts degrees are becoming essential precisely because they teach reasoning and ethics that AI cannot—contradicting Musk's view of college as "basically for fun."