China Says AI Can’t Fire You
Zhou was a quality assurance supervisor at a tech company in Hangzhou, earning 25,000 yuan a month. His employer
Zhou was a quality assurance supervisor at a tech company in Hangzhou, earning 25,000 yuan a month. His employer decided AI could do his job cheaper and offered him a demotion at 15,000 yuan. Zhou refused. The company fired him. Zhou took them to court, won at arbitration, won at the district level, and won again on appeal.
The Hangzhou Intermediate People’s Court published its ruling on April 30, 2026, the day before International Workers’ Day. AI job replacement, the court declared, is not legal grounds for dismissal in China.
Adopting AI is a business decision, not a catastrophe. Companies benefit from automation. They cannot pass the costs to their workforce.
What the Law Says
China’s Labour Contract Law permits termination when there is a “major change in objective circumstances” that makes a contract impossible to perform. The law was designed for genuine disruptions beyond anyone’s control: earthquakes, regulatory shifts, forced relocations.
The court ruled that AI adoption does not qualify. A company choosing to automate for competitive advantage is exercising judgment, not responding to catastrophe. The Hangzhou judges found that Zhou’s employer had failed to demonstrate that his contract had become impossible to perform. Offering him a 40 percent pay cut was not a reasonable reassignment.
A Beijing court had reached the same conclusion months earlier. Liu, a map data collector employed since 2009, lost his position when his company switched to AI-powered collection systems. The Beijing Municipal Bureau of Human Resources and Social Security ruled the dismissal unlawful in December 2025 and published it as a model case, signalling how future disputes should be handled.
The principle emerging from both rulings: workers acquire rights through years of labour that companies cannot extinguish by pointing at a new technology.
Racing to Build AI While Banning AI Layoffs
China’s core AI sector exceeded 1.2 trillion yuan in 2025, with more than 6,200 enterprises competing across the country. The government wants AI in manufacturing, services, and daily life. By 2030, next-generation AI terminals are projected to reach 90 percent penetration.
At the same time, youth unemployment hit a record 18.9 percent in August 2025 and remains nearly double levels from a decade ago. The property market is in distress. Consumer spending is weak. A wave of automation-driven layoffs would be politically catastrophic.
Beijing’s solution is to demand both: AI adoption without mass displacement. The 2026 government work report was the first to explicitly address AI’s impact on jobs. Liu Qingfeng, founder of AI firm iFlytek and a National People’s Congress delegate, proposed a dedicated AI-unemployment insurance fund. The State Council directed companies to pursue innovation with “greater job-creation potential.”
The message is that companies can automate, but they must absorb the transition costs themselves.

Nowhere Else Has This
Most countries treat technological redundancy as a legitimate reason to fire someone. In the UK, companies must follow consultation procedures, but “redundancy” due to automation is valid grounds. Germany’s strong worker councils can negotiate severance and retraining, but cannot block layoffs outright. France requires economic justification and offers generous unemployment benefits, but does not prohibit AI-driven terminations. Australia, Canada, Japan, South Korea, and most of Europe allow companies to restructure around new technology as long as they follow proper process.
The United States sits at the extreme end: employers can eliminate positions due to automation with no legal consequence beyond negotiated severance.
The idea that workers hold rights in their jobs that survive technological change briefly surfaced in America during the 1980s, when plant closures devastated the industrial Midwest. A grassroots campaign argued that a job is a right, that years of labour create a stake corporations cannot simply extinguish. The demand never became law anywhere in the West.
In China, the courts have made it law by interpretation. A business decision is not a natural disaster. Companies that choose to automate must find ways to do it without abandoning the people who built their operations.
What Companies Can Still Do
The rulings do not ban AI. They do not require companies to keep workers in roles that have genuinely ceased to exist. They require negotiation: reasonable reassignment, retraining, or proper compensation.
Zhou’s employer lost because offering a 40 percent pay cut is not reasonable reassignment. A company that provides comparable work at comparable pay, or offers fair severance, may still be able to restructure around AI. The rulings target pretextual firings disguised as technological necessity, not legitimate business evolution.
Whether this approach slows China’s AI ambitions is an open question. Protecting workers creates friction. But mass unemployment creates political instability, and Beijing has decided that social stability matters more than frictionless automation.
Zhou won his case. He got his compensation. Somewhere in Hangzhou, an AI system is doing quality assurance. And a principle has been established: the technology may be inevitable, but the layoffs are optional.
Sources
Fortune: Chinese court rules firms can’t lay off workers on AI grounds
Caixin Global: Chinese Courts Rule Companies Cannot Fire Workers Simply to Replace Them With AI
NPR: A tech worker in China is laid off and replaced by AI. Is it legal?
Lawfare: The Political Limits of China’s AI Diffusion Ambitions



