---
id: "prereq-generative-vs-applied-ai"
type: "prereq"
source_timestamps: ["§ Thinking About AI Capability on a National Scale"]
tags: ["technical-knowledge", "infrastructure"]
related: ["claim-energy-dictates-generative-ai", "concept-embodied-ai-specialization"]
reason: "Without this distinction, the reader cannot understand why a company would choose France for one AI task (energy for training) and Japan for another (robotics infrastructure for application)."
speakers: ["Yasuhiro Yamakawa", "Thomas H. Davenport"]
sources: ["futures"]
sourceVaultSlug: "hbr-seg-futures"
originDay: 2
articleStem: "hbr-cl-94-ai-strategy-beyond-us-china"
sourceUrl: "https://hbr.org/2025/12/your-ai-strategy-needs-to-expand-beyond-the-u-s-and-china"
sourceTitle: "Your AI Strategy Needs to Expand Beyond the U.S. and China"
---
# Generative vs. Applied AI Resource Needs

**Prerequisite knowledge:** The fundamental difference in resource requirements between (a) *training foundation generative models* — which demand massive compute and energy — and (b) *applying existing AI models* to specific business problems or physical hardware — which demands software ecosystems, robotics infrastructure, and domain expertise.

**Why it matters:** Without this distinction, a reader cannot understand why a company would choose **France** for one AI task (energy for training; see [[claim-energy-dictates-generative-ai]]) and **Japan** for another (robotics infrastructure for application; see [[concept-embodied-ai-specialization]]). The distinction is what makes the seven-factor [[framework-national-ai-capability]] actionable — different factors serve different AI workloads.