State Council General Office opinion on employment-friendly AI development
国务院办公厅关于适应人工智能技术发展助力构建就业友好型发展方式的意见 (国办发〔2026〕29号)
A · Strong evidence Primary text of a national State Council General Office opinion that sets China's official line on AI and jobs; it carries no evidence on AI's labour impact, so it earns A as an authoritative, checkable primary source rather than S.
Summary based on the full text·Original: about 9 minutes to read
In short
A policy opinion from China's State Council General Office, signed 16 October 2026 and published 24 October, on adapting to AI while protecting employment (mee.gov.cn repost of the central government text). It asks for an "employment-friendly" path: grow AI-related jobs, deploy AI first where labour is short or work is hazardous, slow rollout in labour-intensive sectors, and add employment-risk assessment to the review of AI projects that could displace many workers. It encourages, without mandating, firms to create an "unemployment buffer period" with advance notice to affected workers.
Key numbers
- Guobanfa [2026] No. 29document number of the opinion (State Council General Office, 2026, No. 29)
- 18numbered tasks, (一) to (十八), grouped under six policy areas
- 16 October 2026date the opinion was signed; published on the central government site 2026-10-24
As reported by the source.
Chinese·Web page·opens mee.gov.cn
What it reports
- Six substantive areas with 18 tasks: job creation from AI; an employment-friendly upgrade path; digital skills; AI in public employment services; worker rights and social security; employment monitoring and risk response.
- Job creation: eastern regions host core AI industry; central and western regions take data labelling and smart-device manufacturing. Eligible AI firms can access special loans tied to keeping and adding jobs. New occupations named: digital manager, AI engineering technician, AI trainer.
- Small firms: an AI start-up support plan, open-source ecosystems, and "small, fast, light, precise" digital products to cut the cost of AI adoption.
- Task (六): deploy AI first in labour-short and hazardous jobs; set the scope, intensity and pace of AI in labour-intensive industries "steadily and prudently"; firms are encouraged to set an unemployment buffer period, notify affected workers in advance during large-scale rollouts, and provide redeployment support.
- Task (七): AI application projects in market-access areas that may cause large employment shocks get an employment-risk assessment during review; other projects get one "as appropriate".
- Skills: universities and vocational schools to add AI majors and "micro-majors"; targeted AI training for graduates, migrant workers and the unemployed; free public courses.
- Services: AI to push employment policies to eligible people ("policy finds the person"), upgrade national job platforms, and support labour-dispute mediation.
- Rights: regulate AI in HR management; act against algorithmic hiring discrimination, fake recruitment and violations of privacy and rest rights. Extend pension and occupational-injury cover to gig and platform workers.
- Monitoring: cross-department data sharing, regional and sector employment monitoring models, a special survey on AI's employment impact, AI employment research centres, and inclusion of AI job risk in the existing large-scale unemployment response mechanism.
How they got it
This is a policy instrument, not research. It cites no data, studies or forecasts on AI displacement. Implementation is assigned to local governments with human resources and development-reform departments coordinating.
Limits
- No numbers, targets, budgets or deadlines; most measures use "encourage", "guide" or "explore".
- The buffer period and advance notice are encouraged for firms, not required; no length or threshold is defined.
- "Large employment shock" and "large-scale rollout" are not defined, so the reach of the risk assessment is unclear until ministries issue detail.
- Library entry dates the item 2026-10-23; the page shows publication on 2026-10-24 and signature on 2026-10-16.
Why it matters for companies
Firms operating in China should expect AI projects with visible labour effects to face questions on redeployment, notice and retraining, especially in labour-intensive sectors and in anything that needs regulatory approval. The text points to the safer sequencing: automate hazardous or hard-to-staff roles first and document transition support. The HR clauses matter for anyone using AI in hiring or workforce management in China: discrimination, fake postings and privacy are named enforcement targets. Watch for ministry-level rules that set thresholds.
Key passages in translation
Chinese original
鼓励企业探索建立失业缓冲期制度，在大规模推广应用人工智能过程中提前告知受影响劳动者、提供转型转岗就业支持。
Firms are encouraged to explore setting up an unemployment buffer period system, notifying affected workers in advance during large-scale AI rollout and providing support for transition and redeployment.
Chinese original
对市场准入领域可能造成较大就业冲击的人工智能应用场景项目，在评审时同步开展就业风险评估
For AI application projects in market-access areas that may cause a large employment shock, an employment-risk assessment is to be carried out at the same time as the project review.
Chinese original
优先在劳动力紧缺、环境高危等岗位推广应用。
Prioritise deployment in jobs where labour is short or the environment is high-risk.
Summary based on the full text. Read the original for the full argument and data.Read the original on mee.gov.cn
- Tags
- Future of work, China, Labour market