China’s government has been unusually direct about the labour side of artificial intelligence. A guidance document issued by the General Office of the State Council sets out how the country intends to build an employment-friendly pattern of development as AI spreads through factories, offices and services.
The framing is notable. Employment is described in the document as the most basic foundation of people’s livelihoods, and the stated aim is to keep the rollout of AI aligned with the creation, rather than the subtraction, of jobs. In practice that means treating worker transition as something to be designed for in advance, not cleaned up afterwards.
That is easier to write than to deliver. AI adoption in China has moved fastest exactly where labour is most concentrated: customer service desks, content moderation, routine accounting, quality inspection on production lines, and the back offices of large platforms. Each of those functions employs large numbers of people with narrow, task-specific skills, and each is the kind of work that current models handle competently.
The document sits alongside the broader AI Plus initiative, which has pushed the technology into manufacturing, healthcare, education and public services. The combination matters because the same policy system is simultaneously encouraging adoption and trying to slow the displacement it causes, an internal tension that other governments are wrestling with in much the same way.
What the guidance signals to employers and local governments is that AI deployment will increasingly be judged on more than productivity gains. Retraining capacity, the treatment of displaced workers and the pace at which new roles appear are likely to become part of how such projects are assessed. Whether that translates into concrete obligations is the question foreign investors and domestic firms will now be watching.

Dgcity