AI adoption has two stages: defensive fear vs. curious experimentation¶
Insight: Developers (and designers) go through two stages of AI adoption. Stage 1 is defensive — when AI makes a mistake, you mock it and feel reassured about job security. Stage 2 is curious — when AI makes a mistake, you ask why and figure out how to prevent it (e.g., adding a Cursor Rule to specify your tech stack). The transition from Stage 1 to Stage 2 is the same curiosity mindset that drove people to software development in the first place. The key insight: AI mistakes are feedback about your instructions, not evidence of AI's inadequacy.
Detail: Ferreira frames the two reactions to the same AI error (confusing Styled Components for Tailwind): Stage 1 mocks the AI, Stage 2 investigates the confusion and fixes it structurally. This maps to the broader pattern in the knowledge base: "AI coding quality is a skill issue" (Shrivu) and "vibe coding vs. AI-assisted engineering" (Osmani). The curiosity mindset is what separates productive AI users from those who dismiss the tools. Ferreira notes this isn't a new idea — curiosity is what makes good developers good, period — but AI adoption specifically requires letting go of ego to learn a new tool.
Sources
Related: existing entry "AI is commodifying knowledge work" in external/ai-assisted-design.md — COMPLEMENTS