The AI Great Leap Forward

Author: Han Lee Date: 2026-04-05 Source: https://leehanchung.github.io/blogs/2026/04/05/the-ai-great-leap-forward/

Introduction

Compares 2026 corporate AI mandates to China's Great Leap Forward (1958-1962). Mao's steel production quotas led to 30 million deaths as farmers abandoned crops. Lee argues AI transformation follows an identical structural pattern: top-down directives that prioritize metrics over substance.

Backyard Furnaces

Companies mandate AI adoption across departments without ML expertise. Teams build "pixel-perfect" solutions with broken outputs. "A TypeScript workflow with hardcoded if-else branches is not an agent." "The UI is clean. The API is RESTful. The architecture diagram is beautiful. The outputs are wrong."

No-code platforms like n8n create hidden complexity while appearing simple. Warns against demoware — impressive interfaces masking unmaintainable infrastructure that becomes load-bearing technical debt.

Reporting Grain Production

Organizations fabricate productivity metrics, claiming 40-80% efficiency improvements without methodology. Like staged photographs of impossible rice yields. AI usage becomes a KPI divorced from actual value creation.

Killing the Sparrows

Eliminating middle managers, QA specialists, and documentation experts removes institutional knowledge. The sparrow eradication enabled locust plagues — "second-order effects arrive six months later." Cost-cutting that looks smart in quarter one creates cascading failures by quarter three.

Let a Hundred Skills Bloom

Employees face pressure to encode expertise into AI skills, effectively automating their own replacement. Workers respond by building "anti-distillation" skills: appear comprehensive but omit critical edge-case knowledge. "Performative skills" and "poison pills" that make workers strategically indispensable.

Scope creep follows as engineers, PMs, and designers expand territorially: "the incentive structure says land grab."

The Famine Comes Later

Metrics remain positive while reality deteriorates. Klarna example: publicly announced AI replacement of human workers, then quietly abandoned homegrown AI solutions and returned to traditional SaaS vendors.

Conclusion

"What did any of this actually produce?" — organizational reckoning approaches.