跪拜 Guibai
← All articles
Frontend · Backend · Programmers

AI Won't Kill Your Job Title — It'll Kill the Busywork You Hide Behind

By 勇宝趣学前端 ·
Read original on juejin.cn ↗ Google Translate ↗ Alt translation

The taxonomy cuts through the noise about which jobs AI will replace and reframes the threat as behavioral, not occupational. For any developer, the six patterns map directly onto common career stagnation modes — and the four counter-capabilities describe exactly what keeps an engineer relevant when tooling changes weekly.

Summary

The danger in the AI era is not a specific profession but a specific posture toward work. Six archetypes are singled out: the information porter who cannot judge what matters, the person who confuses activity with output, the executor who never asks why, the worker who refuses to learn new tools, the advisor who offers safe non-advice, and the guardian of old rules who blocks change. Each pattern shares a reliance on mechanical, repeatable tasks that AI can now perform faster.

The counterweight is four capabilities: finding real problems beneath surface tasks, making judgments with incomplete information, collaborating with AI as a force multiplier, and bearing the consequences of decisions. These are not technical skills but cognitive stances — a willingness to think, decide, and own results.

The argument is not that AI replaces people but that it exposes the gap between performing a role and producing value. A competent person who refuses tools shrinks; an ordinary person who wields AI well expands. The unit of competition shifts from individual ability to the ability to orchestrate AI toward a solved problem.

Takeaways
Information porters — people whose main job is finding, copying, and forwarding data — lose value as AI makes information retrieval nearly free.
Busyness is not output; AI automates the repetitive tasks that create the illusion of productivity, exposing who actually drives results.
Pure executors who follow orders without understanding the goal become replaceable because execution is the easiest link to tool-accelerate.
Refusing to learn new tools turns a decade of experience from an asset into inertia; experience only retains value when it can be recombined with new workflows.
Offering vague, non-committal advice — "correct nonsense" — is a liability because AI generates more of it, faster, for free.
Guarding old processes by proving why things cannot change is a narrowing niche; the competitive move is creating new positions, not defending old ones.
Four capabilities resist obsolescence: finding real problems, making judgments, collaborating with AI, and bearing consequences with fast course-correction.
The contest is no longer human versus AI but who can better wield AI to solve problems — an ordinary person with AI leverage can outperform a skilled person who refuses tools.
Conclusions

The framework reframes AI risk from a skills gap to a responsibility gap. The people most exposed are not those who lack technical ability but those who avoid owning outcomes, which is a cultural and psychological problem, not a training problem.

The six types are not distinct personalities; they are overlapping failure modes that can coexist in the same person. A busy executor who refuses to learn and offers safe advice is a composite that describes a large fraction of middle-management behavior.

The four counter-capabilities are essentially a definition of engineering judgment — problem discovery, decision-making under uncertainty, tool leverage, and ownership — which suggests that AI does not threaten engineering as a discipline but threatens role-players who never practiced it.

Source: juejin.cn ↗ Google Translate ↗ Backup ↗