Chinese courts have ruled that companies cannot terminate employees simply to replace them with AI systems, establishing a legal principle that puts the cost of technological transformation on employers rather than workers.
The most recent ruling came from the Hangzhou Intermediate People's Court, which upheld a finding that a tech firm had illegally fired a quality assurance worker identified only as Zhou after he refused to accept a demotion triggered by AI automation. At 25,000 yuan ($3,640) a month, his job centered on verifying whether large language models were producing accurate results. Once the company concluded that its AI systems had made Zhou's position redundant, it moved to slot him into a lower-level role paying 15,000 yuan — stripping away 40 percent of his income. Zhou rejected the offer, and his employer responded by ending his contract. Every venue that heard the dispute — arbitration, the district court, and the appellate court — ruled against the company and awarded Zhou compensation.
The court found that integrating AI into a business is a deliberate strategic decision, not an unforeseeable change in objective circumstances. That distinction matters because China's Labour Contract Law allows employers to end a contract when a fundamental shift in circumstances renders it unworkable — language that judges have traditionally reserved for disruptions no employer could have anticipated or prevented, like natural catastrophes or relocations ordered by the government. Because the company chose to adopt AI, the court found it could not invoke that provision to justify ending Zhou's contract.
A second case, highlighted by the Beijing Municipal Human Resources and Social Security Bureau in December 2025 as a top labor arbitration decision, used the same logic. This case was about Liu, who had worked for fifteen years doing manual map data entry at a tech company. In early 2024, the company replaced all manual work with AI tools, closed Liu's division, and ended his job, again claiming it was due to unexpected changes. Every authority—the arbitration panel, trial court, and appeals court—disagreed. They said that choosing to automate is a normal business decision, and any job losses from that choice are risks the employer should handle, not the workers.
The Hangzhou court suggested that if companies need to restructure because of AI, they should focus on retraining employees for jobs that need human judgment. Any new role offered should also come with fair pay.
These decisions come at a time when tech workers around the world are under a lot of pressure. By April 2026, more than 78,000 tech jobs had been cut, with about half of those linked directly to AI. In March, China's urban youth unemployment rate was 15.3%, so the government has political reasons to see mass layoffs as a threat to stability as well as a legal issue.
No equivalent protection exists in the U.S. or the E.U. In the United States, the at-will employment doctrine — applicable in all but one state — gives employers broad latitude to end a job for virtually any reason, and displacement by AI falls well outside the narrow list of terminations the law actually prohibits. Under the E.U.'s AI Act, systems deployed in hiring, performance review, and similar employment contexts carry a high-risk designation that triggers transparency and oversight obligations — but the regulation governs how those tools are used, not whether a company may restructure its workforce around them.
Questions about whether companies are accurately attributing layoffs to AI have grown alongside the broader wave of cuts. Some management experts argue that executives are using AI as cover for restructuring decisions driven by over-hiring, margin pressure, or strategic missteps — a phenomenon described as "AI-washing." The Chinese rulings create a framework that would make that framing legally consequential: if a company claims AI drove a termination, it is invoking the same rationale courts have now twice rejected as grounds for dismissal.