Artificial intelligence is reshaping the world of work at breakneck speed, and almost everywhere the same question is being asked: who pays the price when a machine can do your job? Across much of the capitalist world the answer has been brutally simple – the worker does. In China, the courts are giving a very different answer.
In this detailed original analysis for Friends of Socialist China, İbrahim Can Eraslan examines a landmark ruling handed down by the Hangzhou Intermediate People’s Court on the eve of International Workers’ Day 2026, which found that a technology company had unlawfully dismissed an employee after replacing his role with AI. Far from being an isolated case, Eraslan shows how it builds on a consistent and growing body of jurisprudence in Beijing, Guangzhou and beyond, all resting on a single principle: that voluntary AI adoption is a business decision, and that companies which automate must therefore shoulder the corresponding social responsibilities rather than dumping the costs onto their workforce.
Crucially, Eraslan situates these rulings within China’s wider policy architecture – from the 15th Five-Year Plan’s commitment to assessing AI’s employment effects, to proposals for compulsory employment impact assessments before large-scale AI deployment – and contrasts this socialist, employment-first approach with the ‘employment at will’ doctrine of the United States and the patchwork protections of the European Union. The result, he argues, is one of the first coherent national legal frameworks anywhere in the world for managing AI-driven job displacement: not Luddism, but a principled insistence that the fruits of technological progress be shared and that its costs not be socialised onto working people.
He concludes:
China’s courts have drawn a line in the sand. AI is welcome. But AI cannot be an excuse to fire workers. The cost of progress must be shared, and the burden must not fall on those who can least afford to bear it. In a world where capital everywhere is racing to automate labor out of existence, China’s socialist legal system is saying: not so fast.
A worker’s victory in Hangzhou
On April 30, 2026 – the eve of International Workers’ Day – the Hangzhou Intermediate People’s Court delivered a ruling that has sent shockwaves through China’s booming tech sector and captured global attention. The court upheld the decision that a technology company had illegally dismissed a quality assurance supervisor after replacing his role with AI systems, and it designated the case as a model for future rulings on AI-driven dismissals.
The worker, identified by the court only as Zhou, had been employed since November 2022 at an AI-related tech company in Hangzhou, one of China’s premier AI industry hubs. His monthly salary was 25,000 yuan (approximately $3,640), and his duties involved matching user queries with large language models and filtering out illegal or privacy-violating content – the very kind of nuanced, human-in-the-loop work that companies across the world have been racing to automate. When the company deployed AI systems capable of handling those tasks, it did not offer Zhou severance or retraining. Instead, it attempted to reassign him to a lower-level position at a sharply reduced salary of 15,000 yuan per month, a 40 percent pay cut. When Zhou refused, the company terminated his contract, offering 311,695 yuan in compensation and citing organizational restructuring driven by AI adoption.
Zhou contested the dismissal. An arbitration panel ruled in his favor, finding the termination unlawful. The company appealed to a district court and then to the Hangzhou Intermediate People’s Court. On April 30, the intermediate court upheld the lower rulings, publishing the case as part of a set of “typical examples of protecting the rights of AI enterprises and workers” – a deliberate and highly symbolic timing, designed to send a clear message on the eve of May Day. The court found that AI-driven job displacement does not constitute a “major change in objective circumstances” under China’s Labor Contract Law, and that a 40 percent pay cut was not a reasonable reassignment proposal. The dismissal was unlawful on both counts, and the company was ordered to pay Zhou over 260,000 yuan in additional compensation.
As Xinhua reported, the court’s message was unambiguous: “The termination grounds cited by the company didn’t fall under negative circumstances such as business downsizing or operational difficulties, nor did they meet the legal condition that made it ‘impossible to continue the employment contract.’” This was not a mere technicality. It was a principled stand.
A pattern of protection: Beijing, Guangzhou, and beyond
The Hangzhou ruling was not an isolated decision. It built upon a growing body of jurisprudence across China’s court system that is consistently rejecting the argument that AI adoption alone can justify worker dismissal.
In December 2025, the Beijing Municipal Bureau of Human Resources and Social Security released its annual compilation of typical arbitration cases, including a dispute involving a map data collector whose entire department had been eliminated after the company adopted AI systems for data gathering. The arbitration panel ruled the dismissal unlawful, finding that the company’s adoption of AI was a “voluntary move to stay competitive” – not an external force majeure or unavoidable business catastrophe. By citing AI replacement as grounds for dismissal, the panel concluded, the company had “effectively shifted the risks of technological iteration onto its employees.” That, the panel ruled, is illegal.
A year earlier, in 2024, the Guangzhou Intermediate People’s Court issued a similar ruling in favor of a graphic designer at a tech company who had lost his job to AI programs including Midjourney and Pixso AI. The court ruled that using AI was a business decision to adapt to market changes and did not qualify as a “major change in objective circumstances” that would permit contract termination under Chinese labor law.
These three cases – spanning Hangzhou, Beijing, and Guangzhou – represent a coordinated judicial response to one of the defining labor questions of our era. As the Hangzhou court’s white paper noted, the city handled 12,359 labor dispute cases in 2025, up a staggering 61.68 percent from the prior year, with a growing number involving artificial intelligence and big data. The courts are not ducking the issue. They are engaging with it head-on, and they are ruling consistently on the side of workers.
The legal principle: social responsibility over corporate convenience
What makes these rulings so significant is not merely that they sided with individual workers, but the legal principle they establish. At the heart of every case is a narrow but crucial interpretation of Article 40(3) of China’s Labor Contract Law, which permits employers to unilaterally terminate a contract when a “major change in the objective circumstances” makes performance impossible. Chinese courts have historically interpreted this phrase narrowly, restricting it to external shocks such as force majeure, mergers, asset transfers, and regulatory closures – events outside the employer’s discretionary control.
The courts are now making clear that voluntary AI adoption does not qualify. When a company chooses to deploy AI systems to capture efficiency gains and cost advantages, that is a business decision – a “voluntary move to stay competitive,” as the Beijing arbitration panel put it. It is not an earthquake, a flood, or a government-mandated restructuring. It is a calculated corporate strategy, and the courts are insisting that companies who make that choice must also accept the corresponding social obligations.
As lawyer Wang Xuyang of Zhejiang Xingjing law firm summarized in commentary to Xinhua: “While companies may benefit from AI-driven efficiency gains, they must also bear corresponding social responsibilities. AI replacement does not automatically justify terminating a labor contract.” This formulation captures the essence of the judicial approach: the costs of technological transformation should not be borne by workers. Companies that automate cannot simply announce that AI has erased a job. They must deal with the worker – through negotiation, training, reasonable reassignment, or lawful compensation – instead of dumping the cost of technology onto the employee.
As CCTV noted in its commentary, the Hangzhou court’s ruling provides a positive guideline: “When job adjustments are indeed necessary, priority should be given to training employees, improving their skills, and internal transfers, rather than unreasonable job adjustments with significant salary reductions or direct dismissal.” The priority is clear: people before profits.
The policy architecture: from courtroom to Five-Year Plan
What distinguishes China’s approach from the scattershot, reactive labor policies seen in many Western countries is the systematic, whole-of-government architecture being constructed around AI and employment. The court rulings are not operating in a vacuum – they are the judicial expression of a broader policy direction emanating from the highest levels of China’s planning apparatus.
The 15th Five-Year Plan (2026–2030) explicitly calls for integrating AI’s employment effects into assessment mechanisms for major policies, projects, and productivity planning. This is not a footnote or an afterthought. It is a structural commitment to evaluating how technological deployment affects employment before decisions are made, not after workers have already been thrown out of their jobs.
In January 2026, the Ministry of Human Resources and Social Security (MOHRSS) announced at a press conference that it would accelerate the development of an AI employment impact monitoring and early warning system and issue a dedicated policy document addressing AI’s effects on the labor market. As China Daily reported, the ministry also pledged to foster an “employment-friendly development model” and expand skills training, with planned measures including a new five-year employment plan and broader social insurance coverage for gig and flexible workers.
At the 2026 Two Sessions – China’s annual parliamentary gathering – Ma Yide, an NPC deputy and dean of the School of Intellectual Property at the University of Chinese Academy of Sciences, proposed that companies be required to submit employment impact assessment reports to regulators before large-scale AI deployment, with pre-deployment evaluation and ongoing monitoring. As PolicyCN reported, Ma warned that the current policy gap, if not addressed, could evolve into major risks affecting social stability. He identified four critical employment impacts from AI: the systematic replacement of white-collar positions, intensified youth employment pressure, middle-aged workers facing “skill depreciation,” and the weakening of traditional service industry “reservoir” functions as AI penetrates customer service, delivery, and financial advisory roles.
Wang Xiaoping, China’s Minister of Human Resources and Social Security, reinforced this direction at a March 2026 press conference, stating that the ministry is studying measures to promote the use of AI to create new jobs while empowering traditional ones, with the goal of “promoting inclusive development that aligns technological progress with improvements in people’s livelihoods.” As Semiwiki reported, the minister emphasized that the employment-first strategy will be further implemented during the 15th Five-Year Plan period.
The scale of the challenge: context and stakes
The urgency of these legal and policy measures becomes clear when one considers the sheer scale of China’s AI sector and its penetration into the economy. China’s core AI industry exceeded 1.2 trillion yuan (approximately $174 billion) in 2025, with more than 6,200 enterprises operating in the field, according to Minister of Industry and Information Technology Li Lecheng. Over 30 percent of manufacturing enterprises had adopted AI technologies by the end of 2025, and Chinese companies released over 300 humanoid robot products – more than half the global total. Chinese open-source large models ranked first globally in downloads last year.
By 2030, the penetration rate of next-generation intelligent terminals and AI agents in China is expected to exceed 90 percent. The International Labour Organization’s 2025 study estimated that a quarter of all jobs worldwide – equivalent to roughly 838 million positions – could be affected by generative AI. High-income countries face the largest exposure, with 33.5 percent of total jobs falling into one of four exposure categories.
Against this backdrop, China’s court rulings take on historic significance. They represent one of the first coherent national legal frameworks anywhere in the world for addressing AI-driven job displacement – not with Luddite resistance to technology, but with a principled insistence that the fruits of technological progress be shared and that the costs not be socialized onto working people.
A socialist approach to the AI revolution
What China’s courts and policymakers are articulating is, at its core, a socialist response to the AI revolution. The principle they are establishing – that companies benefiting from AI-driven efficiency must bear corresponding social responsibilities – stands in sharp contrast to the approach taken in capitalist economies, where the default assumption is that employers have broad discretion to restructure, automate, and dispose of workers as they see fit, limited only by weak severance obligations and easily circumvented labor protections.
In the United States, for example, the doctrine of “employment at will” means that private-sector workers can be fired for virtually any reason – including being replaced by AI – with no legal recourse unless discrimination laws are implicated. In the European Union, the AI Act has made strides in regulating algorithmic management and automated decision-making, but employment protections vary widely by member state, and no unified legal principle comparable to China’s emerging jurisprudence exists at the EU level. No major Western jurisdiction has yet established the clear judicial precedent that voluntary AI adoption cannot justify dismissal.
China’s approach reflects the fundamental orientation of its socialist legal and economic system. As researcher Wang Tianyu of the Chinese Academy of Social Sciences told state media, “Technological progress may be irreversible, but it cannot exist outside of the legal framework.” Forward-thinking institutional design, he emphasized, is necessary to protect the dignity and rights of workers. This is not anti-technology sentiment. It is the insistence that technology serve people, not the other way around – a principle that lies at the heart of socialist development theory.
The court rulings also embody the “employment-first” strategy that has been a consistent priority of China’s governance model. Even as Beijing pushes aggressively for AI leadership – investing billions, nurturing over 6,200 AI enterprises, and racing toward 90 percent penetration of intelligent terminals – it is simultaneously building guardrails to ensure that this technological leap does not come at the expense of working people’s livelihoods. This is the dual-track approach that socialist systems are uniquely positioned to implement: unleashing productive forces while maintaining social protections that capitalist systems, driven by short-term profit maximization, consistently fail to deliver.
The Hangzhou decision is being hailed by legal scholars as a reassuring signal for labor rights protection at a time when the central Chinese leadership is pushing for industries to widely adopt AI technology. It is precisely this balance – between encouraging innovation and protecting workers – that China’s socialist system is demonstrating can be achieved.
Looking forward
The Hangzhou ruling, alongside the Beijing and Guangzhou precedents, marks the beginning of what will likely be a much larger body of AI labor jurisprudence in China. Multiple similar disputes are reportedly active across Chinese cities including Shanghai and Shenzhen. Judge Shi Guoqiang of the Hangzhou Intermediate People’s Court told broadcaster CCTV that the court does not believe “AI technology has reached the point where it can substantially replace human workers” – a judicial assessment that aligns with the practical reality that AI systems, for all their advances, still require human oversight, judgment, and management.
Zhou’s lawyer, Jiang Xiaotong, captured the human dimension of these cases. Zhou, she noted, “not only suffered a blow to their income but also experienced acute professional anxiety, becoming deeply apprehensive about their future career prospects.” He is middle-aged and faces significant family and financial pressures – precisely the kind of worker that market-driven capitalism discards without a second thought. The court’s decision to designate the ruling as one that other courts can follow, Jiang said, was significant: “Now that a precedent-setting case has been established, people are far more willing to take up the weapon of the law to defend their legitimate rights and interests.”
These developments offer both practical lessons and theoretical insights. They demonstrate that a socialist state can embrace cutting-edge technology without abandoning its commitment to working people. They show that legal frameworks can be developed proactively to address new challenges, rather than lagging behind corporate innovation by decades. And they affirm that the fundamental socialist principle – that the economy should serve the people, not the other way around – remains not only relevant but essential in the age of artificial intelligence.
China’s courts have drawn a line in the sand. AI is welcome. But AI cannot be an excuse to fire workers. The cost of progress must be shared, and the burden must not fall on those who can least afford to bear it. In a world where capital everywhere is racing to automate labor out of existence, China’s socialist legal system is saying: not so fast.