A court ruling in China has drawn a clear line around one of the biggest questions in workplace automation: AI adoption does not, by itself, make a layoff lawful. In a dispute involving a technology worker whose duties were partly taken over by large language models, the court sided with the employee and rejected the company’s attempt to justify the termination as a restructuring move.
The case has become a notable signal for employers racing to cut costs through automation. It suggests that efficiency gains from AI do not automatically erase labor protections, even when the technology can perform tasks that once belonged to a human employee.
A dismissal tied to automation
The dispute centered on a man surnamed Zhou and a technology company in Hangzhou. Zhou joined the company in 2022 as a quality assurance supervisor and earned 25,000 yuan per month.
His role was closely connected to large language models. He reviewed system outputs, matched user questions, and filtered illegal or problematic content.
Over time, the company began using AI to take over part of his work. It then proposed moving him into a lower position with pay reduced to 15,000 yuan.
Zhou refused, arguing that the change in role and the salary cut were unreasonable. The company later ended his contract, saying the move was part of an organizational restructuring and a reduced need for staff.
Why the court rejected the company’s argument
Zhou did not accept either the dismissal or the compensation package offered to him. He brought the dispute to arbitration, where the panel ruled in his favor and found the termination unlawful.
The company challenged that outcome in court and then appealed to the Hangzhou Intermediate People’s Court. The higher court still upheld the earlier decision.
In its view, the company failed to prove that Zhou’s position had truly become impossible to maintain. That finding became central to the ruling, because the company had relied on the claim that AI had changed the nature of the work enough to justify ending the employment relationship.
The court also concluded that replacing a worker with AI does not automatically satisfy the legal standard of a “major change in circumstances” under China’s Labour Contract Law. That clause can sometimes be used as a basis for ending a contract, but the judges said the existence of new technology alone is not enough.
What the ruling means for employers
The court also examined the company’s offer to reassign Zhou to a lower-paid role. It decided that a large reduction in salary could not be treated as a reasonable reassignment.
That point mattered because it showed the company had not simply adapted the job in a fair way. Instead, it had tried to move the employee into a less favorable position after AI absorbed part of his responsibilities.
Legal experts cited in the case materials said the ruling sets an important boundary for companies using AI. They noted that businesses may benefit from AI-driven efficiency, but they still remain responsible for the effect of those changes on workers.
The case was published as part of a collection of “typical examples of protection of AI company and worker rights” ahead of International Workers’ Day on May 1. That framing suggests the matter was viewed as more than a one-off employment dispute.
A broader pattern in China
The Zhou case is not the only one of its kind. A similar dispute arose last year in China involving a map data collector whose work was automated.
In that case as well, the court said AI replacement alone was not enough to justify a dismissal. Beijing Municipal Bureau of Human Resources and Social Security later included that ruling in a set of typical arbitration decisions for 2025 published in December.
Taken together, the cases point to a clearer direction in Chinese labor disputes involving automation. Companies can pursue AI-led efficiency, but they cannot assume that technology alone removes their obligations to employees.
