Workers in China worry over being replaced as they adapt to the growing impact of AI on jobs

TEXT START: Computer programmer Fei Zhaojun's boss asked him if artificial intelligence (AI) could soon replace humans in coding jobs.

The Dissection

This is a field report on labor bargaining power being amputated in real time, disguised as a story about adaptation. Programmer layoffs, translator pay collapsing by more than half, declining foreign-language programs, industrial AI adoption rising from 9.6% to 47.5%, robotic logistics, and shrinking human-produced media all point toward the same mechanism: fewer workers are required to produce economic output.

The article then redirects the reader from systemic displacement to individual coping experiments—vlogging, training AI, independent studios, and classroom augmentation. It converts a structural crisis into a collection of personal adjustments and presents demographic decline as a possible escape hatch.

The Core Fallacy

The article treats AI primarily as a productivity tool that will complement workers or compensate for China’s aging population. Under the Discontinuity Thesis, that is the wrong unit of analysis. The decisive question is not whether AI raises output or helps an aging society. It is whether human labor remains economically necessary.

If AI achieves durable cost and performance superiority, productivity gains reduce the demand for labor. A shrinking workforce does not recreate the jobs AI eliminates; it merely reduces the number of humans competing for work while machines continue expanding capacity. Nor does the claim that humans remain decision-makers prove mass employment will survive. Residual selection, supervision, and taste can form niches without supporting the wage-to-consumption circuit for the majority.

Hidden Assumptions

• Productivity gains will be distributed through wages rather than concentrated by owners of AI capital.
• AI will remain a task-level assistant instead of replacing entire roles and departments.
• Human creativity, judgment, and trust will remain scarce enough to support large-scale employment.
• Demographic aging will offset displacement before AI eliminates more labor than population decline removes.
• Individual entrepreneurship can absorb workers expelled from conventional employment.
• Headline unemployment rates adequately capture underemployment, wage compression, labor-force exit, and declining career value.
• Government promotion of AI can preserve social stability without restoring productive participation.

Social Function

Partial truth functioning as transition management and ideological anesthetic. The article documents real displacement, so it is not pure propaganda. But its framing channels fear into self-adaptation: if workers cannot defeat automation, they should help deploy it, monetize a niche, or accept temporary usefulness.

Its productivity and demographic arguments reassure institutions that the system can absorb the shock. They do not establish that displaced workers will retain bargaining power or purchasing power. The article makes the victims responsible for adapting to the machine that is removing their economic necessity.

The Verdict

The article accidentally supports the Discontinuity Thesis more than it refutes it. The layoffs, wage collapse, credential devaluation, and robotic encroachment are early P1 and P3 symptoms. China’s state-backed diffusion of AI accelerates P2 by making human-only economic domains harder to preserve.

The anecdotes describe transition niches, not a replacement for mass productive participation. Aging may slow the visible unemployment shock, but it cannot restore the wage-consumption circuit once AI makes labor broadly surplus. This is not evidence that the system has already finished dying. It is evidence that the corpse has begun moving under automation.