China's AI strategy: Increase productivity, stabilize the labor market

China's AI strategy: Productivity gains at a social risk

Robots, new professions, fewer jobs? China's difficult AI course

China aims to advance artificial intelligence and automation without abandoning the labor market to the consequences of technological change. New guidelines focus on further education, digital skills, improved employment services, and social security. At the same time, companies should be able to make greater use of AI – especially where there is a labor shortage or where jobs are dangerous. In labor-intensive sectors, however, the pace should be cautious.

The balancing act is challenging: productivity gains do not automatically lead to more employment. AI can take over individual tasks, supplement human labor, or create new demand. Which effect prevails depends, among other things, on how quickly companies and employees adapt – and whether new jobs actually offer sustainable incomes. The starting point is also complex: while China's high-tech industry is growing strongly, consumption is developing much more slowly.

The guidelines promise neither a general job guarantee nor a comprehensive ban on automation. Their implementation will therefore be crucial: Do training programs reach employees, are transitions to new roles successful, and do productivity gains benefit a wider segment of society? Only when job quality, income, and social support keep pace with technological development can it be assessed whether China's AI strategy is economically viable.

The machines are supposed to deliver – but who pays for the progress?

China is attempting to combine AI-driven productivity gains with a controlled adjustment of the labor market, rather than fundamentally limiting technological modernization. The economic viability of this approach depends on whether training, new business models, and social security facilitate the transition into productive employment.

Employment becomes the measure of modernization

The guidelines, published in October 2026, comprise 18 tasks across six areas: job creation, employment-friendly AI adoption, digital skills, labor market services, employee rights and social security, and monitoring and addressing employment risks. The document is dated October 16, 2026, and was made public on October 23.

The guidelines are not equivalent to a general job guarantee or a comprehensive ban on automation. They contain various steering instruments, including support measures, risk assessments, and recommendations for companies. Of particular importance is the distinction between the recommended early notification of affected employees and a binding, universally identical statutory notice period, which the document does not specify.

The initial data requires differentiation

At the end of 2025, 725.04 million people were employed in China, including 475.35 million in urban areas. In the first three quarters of 2026, real GDP grew by 4.6 percent compared to the same period of the previous year; the average urban survey unemployment rate was 5.2 percent. This rate is not a comprehensive indicator of all forms of insufficient employment nationwide, but rather refers to the defined urban survey.

The value added of statistically recorded larger high-tech industrial companies increased by 14.3 percent in real terms during the first nine months of 2026, while retail sales of consumer goods, including food service, rose by 1.1 percent in nominal terms. Broader retail sales of goods and services increased by 2.6 percent, with service sales included therein rising by 5 percent. Due to differing definitions and price adjustments, these growth rates cannot be directly equated. Nevertheless, analytically, they illustrate the tension between technologically dynamic production and comparatively subdued consumer spending.

At the end of 2025, 323.38 million people were at least 60 years old, representing 23 percent of the population. The 16-to-59 age group comprised 851.36 million people, or 60.6 percent. This age group is not the same as the working-age population or the number of people actually employed.

Automation is not merely a question of the future

In 2025, approximately 354,000 industrial robots were installed in China, roughly 20 percent more than in 2024. China accounted for 59 percent of new installations worldwide; Chinese manufacturers achieved a 55 percent share of newly installed systems in their domestic market. These figures were published in September 2026.

The figures reflect robot installations, not job losses. Therefore, converting them to job losses using a uniform multiplier would not be reliable. Similarly, industrial robotics, traditional software automation, and generative AI must be considered separately because they involve different tasks and investment decisions.

Technical impact is not a loss forecast

The international study on generative AI from 2025 classifies approximately one-quarter of global employment as belonging to jobs with at least partial technological impact. The highest category comprises 3.3 percent of global employment. The study measures potential job impact and unobserved or unforeseen job losses.

These figures cannot be directly extrapolated to China's total employment figures. This would require, among other things, Chinese occupational and activity data, as well as information on actual implementation. Furthermore, the study focuses on generative AI and not on all industrial automation technologies.

A job consists of several tasks. Automating individual components can therefore change a job without eliminating it entirely. Conversely, simply taking over particularly time-consuming tasks can significantly reduce an organization's staffing needs. The remaining demand for labor depends on the reorganization of processes and the development of demand for the respective services.

Industry focus areas: B2B, digitalization (from AI to XR), mechanical engineering, logistics, renewable energies and industry

More information here:

• Expert Business Hub
• China Blog / Insights

A thematic hub offering insights and expertise:

• Knowledge platform covering global and regional economies, innovation and industry-specific trends
• A collection of analyses, insights, and background information from our key areas of focus
• A place for expertise and information on current developments in business and technology
• A hub for companies seeking information on markets, digitalization, and industry innovations

Productivity and employment do not follow an equation

Analytically, a distinction must be made between direct labor replacement, the supplementation of human skills, and additional demand resulting from new or cheaper services. With unchanged production volume, higher productivity can reduce the need for labor hours. Conversely, with sufficiently growing demand, it can be compatible with stable or increasing employment.

The speed of the transition is therefore crucial. If existing jobs become obsolete before new activities become economically viable, income and adjustment costs arise. China's promotion of further education and career changes aims to address this transition gap.

A study of 5,172 customer service employees found that, on average, the introduction of an AI assistant resulted in 15 percent more successfully resolved customer service requests per hour. Less experienced employees benefited particularly, while the advantages were smaller for highly experienced staff, and certain quality metrics declined slightly. The study explicitly does not allow for any direct conclusions about overall employment and wages.

The results suggest that company projects should be evaluated not only based on potential staff savings, but also on additional benefits, quality, and training effects. However, they do not justify a general productivity assumption of 15 percent or a blanket job guarantee.

Different industries require different transitions

The Chinese strategy promotes human-machine collaboration in industry, the digital modernization of agriculture, and new service offerings. It prioritizes AI applications in areas with labor shortages and hazardous working conditions. In labor-intensive sectors, the scope, intensity, and pace of AI implementation should be carefully considered.

For manufacturing, a combination of specialist knowledge, process understanding, and digital competence appears particularly relevant. In logistics, a distinction should be made between physical relief and more comprehensive automation of multiple process stages. In agriculture, integration opportunities depend on whether maintenance, processing, and consulting services are provided regionally. Services are not automatically a safe haven, because standardized information processing can also be automated there.

The aforementioned company examples are analytical use cases and not claims about actual employment results achieved by individual Chinese companies. Their function is to illustrate different mechanisms of action.

Small and medium-sized enterprises and qualifications determine the breadth

The guidelines support cost-effective and easily deployable digital solutions, open development ecosystems, and public support programs for small and medium-sized enterprises. They link this to the goal of stabilizing and expanding employment. At the same time, they aim to develop new AI-related professions and professional standards.

A new job title is not proof of increased demand. What is economically relevant are skills that companies actually need and can afford. Therefore, mixed profiles combining industry knowledge, data understanding, and quality control appear to be a particularly important subject of analysis.

Universities and vocational training institutions should adapt their programs and collaborate more closely with businesses. Specialized training programs should reach, among others, young people, migrant workers, and the unemployed. The impact of such measures should be assessed not only by participation numbers but also by subsequent employment, income, and the stability of the career transition.

Entering the workforce presents an additional conflict of objectives: if simple tasks are eliminated, learning opportunities can also be lost. A sustainable long-term personnel policy must therefore explain how new employees can acquire practical experience and judgment despite automation.

Early warning systems need reliable root cause analyses

China plans to increase data sharing between government agencies and continuously monitor AI adoption, job creation, job displacement, and skills needs. In addition, research activities and institutions are planned to investigate the employment and distributional consequences.

For the evaluation, layoffs, unfilled vacancies, reduced working hours, and the development of entry-level positions should be examined separately. Job cuts should not be automatically attributed to AI if they coincide with order declines, relocations, or other organizational changes. Longer observation periods and comparisons with similar companies would be more meaningful.

Transitional periods offer economic benefits primarily when they allow for further training, internal transfers, and realistic subsequent employment opportunities. A mere delay without adjustments can increase costs without improving prospects. Therefore, risk assessments should compare different courses of action and not simply aim to avoid all personnel changes.

Employment quality is part of the balance sheet

The guidelines address algorithmic discrimination, false or fraudulent job offers, and infringements on privacy and the right to peace and quiet. They also aim to protect wages and occupational safety in the context of changing forms of human-machine collaboration. Public employment services should use AI to make suitable jobs and support services more accessible.

The evaluation should therefore encompass more than just the number of jobs preserved. Workload, decision quality, and actual opportunities for correction are also economically relevant. A placement system that rewards rapid transitions into precarious employment could produce worse long-term results than a somewhat longer but targeted training program.

Hedging also stabilizes demand

For affected employees, counselling, training, job recommendations, and start-up support are planned. Eligible insured persons will receive unemployment benefits; social security for flexible and new forms of employment will be improved. The document does not include a general income guarantee for all people affected by AI.

Analyses of the Chinese economy link uncertain income prospects and weak social security with cautious consumption and increased savings. Thus, social support measures also have a demand-side function. They can increase the willingness to undertake further education and career changes without excessively restricting consumer spending as a precautionary measure.

The guidelines also address the further development of remuneration and distribution mechanisms. However, they do not specify a binding general distribution of additional productivity gains. Therefore, it remains open for later evaluation whether gains will be passed on more broadly through wages, prices, and public services, or whether they will primarily benefit a limited circle of companies and skilled workers.

Implementation is the decisive test

Eastern regions should focus on developing core AI industries, while central and western regions, according to their respective capabilities, should take on areas such as data processing and the production of smart devices. However, for employees, not only the number of jobs in a given region is important, but also accessibility, skills matching, and social benefits. Access to and the transferability of social benefits remain significant challenges for migrant and informal workers.

The strategy refers to loans, funds, start-up support, and educational programs, but does not specify a uniform additional overall budget for all 18 tasks. At the same time, economic analyses describe financial constraints on local administrations, partly due to declining revenues from land leases. This results in a regionally varying level of support as an analytical risk.

A reliable impact assessment of the guidelines published at the end of October 2026 is not yet available. For a later evaluation, the transition period, income development, job quality, company productivity, and the reach of social support are more meaningful than the number of announced programs. Economic success should be measured by whether increased technological capacity is accompanied by productive employment and a broader income base.

Xpert.Digital is a data-driven B2B industry hub led by Konrad Wolfenstein . The company acts as an external, quasi-in-house solution for industrial partners, closing operational gaps in marketing, content, and sales – without requiring additional resources on the client side.

More information here:

• The quasi-in-house solution: How Xpert.Digital closes operational gaps in B2B marketing and sales – Smart Content-Driven Business

☑️ Our business language is English or German

☑️ NEW: Correspondence in your native language!

 

I and my team are happy to be available to you as your personal advisor.

You can contact me by filling out the contact form here wolfenstein@xpert.digital:or simply call me at +49 7348 4088 965. My email address is

I'm looking forward to our joint project.

 

 

☑️ SME support in strategy, consulting, planning and implementation

☑️ Creation or realignment of the digital strategy and digitization

☑️ Expansion and optimization of international sales processes

☑️ Global & Digital B2B trading platforms

☑️ Pioneer Business Development / Marketing / PR / Trade Fairs