China's cabinet has put job protection at the centre of its AI strategy. On October 23 the General Office of the State Council published a policy opinion, numbered Guobanfa [2026] No. 29 and signed on October 16, that tells provincial governments and ministries to expand the jobs AI can create while actively slowing the displacement it causes. Reuters reported that the guidelines urge agencies to mitigate AI-driven job losses effectively.
The document lists 18 tasks under six headings: creating AI jobs, steering AI upgrades in an employment-friendly direction, raising workers' digital skills, using AI in public employment services, updating labour rights and social insurance, and building a system to monitor and respond to AI-related employment risk. Xinhua, China Daily and the Global Times carried summaries of the same text.
Several provisions go beyond general encouragement. Officials are told to favour AI deployment in jobs that face labour shortages or physical danger, and to handle its use in labour-intensive industries cautiously, managing the scope, intensity and pace of adoption so that workers are hit as little as possible. Companies are encouraged to try an unemployment buffer period, warning staff in advance before large-scale AI rollouts and helping them move to new roles. For AI application projects in market-access areas that could cause a large employment shock, an employment-risk assessment is to be carried out as part of project review.
On the creation side, the plan asks eastern provinces to host core AI industries while central and western regions take on supporting work such as data labelling and smart-device manufacturing. It calls for new occupations and standards, naming AI trainers and AI engineering technicians among them, more AI programmes at universities and vocational schools, and targeted training for graduates, migrant workers and the unemployed. State investment funds, including the national AI industry fund, are to back AI companies on market terms.
The labour-rights section is notable for its attention to algorithms. It calls for rules on AI use in human-resources management and for enforcement against hiring discrimination, fake recruitment and privacy violations caused by algorithm abuse. Gig and platform workers are to be encouraged into the urban employees' pension scheme, and a pilot of occupational-injury cover for new forms of employment is to be widened. AI-related job risk will also be folded into China's existing mechanism for preventing large-scale unemployment, backed by special surveys and new research centres.
The Global Times linked the document to a five-year employment-first plan the State Council released in June, which already promoted the 'AI Plus' initiative and human-AI collaboration. The new opinion turns that emphasis into a more specific checklist for local officials, with human-resources and development-and-reform departments given the job of coordination.
How much it changes in practice remains to be seen. The opinion is a policy directive rather than a law, many measures are phrased as encouragement, and it sets no budget, numerical targets or penalties. Key details, such as which projects count as likely to cause a major employment shock and how the risk reviews will be run, are left to later implementation by ministries and local governments.