China Issues AI Employment Guidelines Urging Companies to Assess Job Risks and Support Workers

China has issued new guidelines calling for a more employment-friendly approach to artificial intelligence, directing authorities and businesses to strengthen worker support while continuing to expand the use of AI across the economy.

The guidelines, issued by the General Office of the State Council and published on October 23, set out 18 tasks in six areas. These include increasing AI's potential to create jobs, making adoption more employment-friendly, improving workers' digital skills, strengthening employment services, protecting labour and social-security rights, and monitoring employment risks. The full policy is available through the Chinese government's official English-language website.

The policy reflects a growing concern among governments worldwide: AI can improve productivity and create new industries, but it can also reduce demand for certain tasks and occupations. China's approach seeks to encourage technological upgrading while mitigating the impact on workers whose jobs may be changed or displaced by automation.

The guidelines call for businesses to explore mechanisms such as employment buffer periods when large-scale AI deployment affects workers. Companies are encouraged to inform affected employees in advance and provide support for retraining, job transfers and other employment transitions. The document also calls for assessing the employment risks of AI applications, particularly in projects that could have a significant impact on jobs.

This approach is intended to make workforce planning part of technology adoption rather than treating job losses solely as a consequence to be addressed after a system has been introduced. For employers, it could mean that decisions about automation increasingly require consideration of staff training, reassignment options and communication with workers.

The guidelines also call for a measured approach to introducing AI in labour-intensive industries. Authorities are encouraged to prioritise applications in areas where workers are in short supply or face hazardous working conditions. In those settings, automation could reduce physical risks and help employers address labour shortages. By contrast, rapid deployment in sectors with large workforces may require more careful planning to limit disruption.

The policy does not call for China to stop using AI in workplaces. Instead, it emphasises combining technological development with employment support. The government wants AI to create opportunities in emerging industries, support entrepreneurship and improve the quality of existing jobs, while providing assistance to workers affected by changes in labour demand.

Training is a central component. The guidelines call for improving digital skills and employment services, including targeted career guidance, vocational training and job recommendations for people who need to move into new roles. These measures could be particularly relevant for workers whose current tasks are increasingly performed by automated systems, as well as people seeking to enter industries that rely on data, software and AI-related services.

The document also addresses risks associated with the use of AI in human-resource management. It calls for action against unlawful practices arising from the misuse of algorithms, including employment discrimination, fraudulent recruitment and violations of privacy or rest rights. Such safeguards matter because AI systems can influence recruitment, screening, scheduling and performance assessment. If those systems rely on biased data or opaque criteria, workers may have difficulty understanding or challenging decisions that affect them.

The policy also seeks to strengthen protections for flexible workers and people in new forms of employment. As digital platforms and automated systems change how work is organised, social-security arrangements may need to adapt to people whose income or working hours do not follow conventional employment patterns. The guidelines encourage improvements to social protection and related mechanisms for these workers.

For companies, the new framework could create additional responsibilities when adopting AI at scale. Employers may need to evaluate potential workforce effects, provide earlier communication about major changes and consider training or reassignment before implementing systems that significantly alter job requirements. The specific implications will depend on how the guidelines are translated into administrative practice and how existing labour laws apply to individual cases.

The measures also carry economic implications. AI investment is often justified by productivity gains, but abrupt employment disruption can affect household income and confidence. By promoting training and smoother job transitions, authorities hope to balance industrial upgrading with the goal of maintaining stable and higher-quality employment.

The guidelines do not establish that every AI deployment requires the same formal assessment or that all job reductions are prohibited. Rather, they set a policy direction focused on risk monitoring, worker protection and the development of new employment opportunities. The practical test will be whether employers, local authorities and training providers can translate those principles into timely and effective support.