It Feels Like Squid Game: China's Workers Fight to Survive the AI Wave

Earlier this month, nearly a thousand people queued outside Tencent's headquarters in Shenzhen. Students. Retirees. Mid-career office managers. All of them waiting to get a piece of software installed on their devices.

The software was OpenClaw, an open-source AI agent built by Austrian programmer Peter Steinberger - dubbed "raising a lobster" in Chinese internet slang, a reference to the tool's red logo. The frenzy was not the kind of excitement you see at an iPhone launch. It was the kind you see when people are running from something.

"It feels like playing Squid Game," Shanghai-based software developer Lambert Li, one of the early OpenClaw users, told Rest of World. "You can get eliminated anytime. How can you not be anxious?" His employer had cut 30 percent of its workforce in 2025 - pruning the people who could not adapt fast enough to AI-assisted workflows. [Rest of World, March 2026]

Li's words sit in a broader landscape of numbers that do not lie. On RedNote, China's wildly popular social platform, the hashtag #AIAnxiety has accumulated around 2.6 million views. Posts read like dispatches from a workforce under siege: "Trying to keep up with AI is more exhausting than the job itself." "My boss asked me to write AI code to replace several staff members. When will it be my turn?"

This is what the global race for AI dominance looks like from the ground level - not from boardrooms, not from conference keynotes, but from the lives of the people it is eating.

China has one of the world's largest AI user bases, and the pressure to adapt is felt across every sector of the workforce. (Photo: Pexels)

The Fear Behind the Queue

To understand why a thousand people would queue to have software installed on their phones, you need to understand what is happening to China's job market at scale.

A study by Peking University analyzed more than one million online job postings between 2018 and 2024. The findings were stark: functions that AI could perform had seen a significant and sustained decline in hiring. Computer programming. Accounting. Editorial work. Sales. The roles that built China's middle class are the same roles AI is now eating. [Peking University study, cited Rest of World]

The unemployment rate for Chinese youth aged 16 to 24 already hovered between 15 and 19 percent through 2025 - significantly higher than the global average. In the United States, the same cohort sits at 9 to 11 percent. The AI acceleration is landing on a job market that was already fragile.

A survey conducted in May 2025 by Cheung Kong Graduate School of Business found that 85.5 percent of 11,814 Chinese respondents said they were worried about how AI could affect their employment. That is not a minority concern or a niche anxiety. That is a near-consensus fear cutting across an entire society. [CKGSB Survey, May 2025]

In an August 2025 survey of 38,000 working adults across 34 countries, nearly a third of respondents said they "strongly believe" AI could replace them and were actively seeking new employment. China's numbers in the same study skewed higher in AI optimism - 69 percent of Chinese respondents said AI's benefits outweighed its risks, versus only 35 percent of Americans - but that institutional positivity coexists awkwardly with the street-level panic visible in Shenzhen queues and RedNote threads. [ADP Research / KPMG Global AI Survey, 2025]

Lambert Li's Dilemma: Ride the Wave or Get Wiped Out

Lambert Li did not stay impressed by OpenClaw for long. After a few days of exploration, the 35-year-old software developer realized the agent was not particularly useful for his actual work. Unlike conversational tools like ChatGPT or Google Gemini, OpenClaw operates directly on a user's device, autonomously executing tasks across files and applications. Li worried it would make mistakes if given too much access to his work systems.

But he could not simply ignore it either. The social pressure to demonstrate AI literacy is now embedded in performance reviews, hiring conversations, and the informal currency of workplace status. Since 2025, Li has jumped from one AI tool to another, testing every major model update and productivity agent he encounters. Not because it makes him happier. Because he is terrified of being the person in the room who does not know how to use the tools.

His is not a solitary experience. Betty Lai, a product marketing manager, learned that her company's annual performance appraisals would now include an assessment of employees' knowledge and use of AI tools. A colleague immediately organized a voluntary training workshop. People fought for front-row seats.

"The pressure sometimes comes from the company expecting us to become more efficient with these tools. But that's not always true yet. It can take time to figure out how to actually incorporate them into your job. There's no point in being anxious. We're already in this wave. Either you ride it, or you get wiped out." - Betty Lai, product marketing manager, China, Rest of World March 2026

That phrase - "either you ride it, or you get wiped out" - captures something important. This is not a slow-moving cultural transition workers can plan around. The pace of change is the central source of distress. People are being asked to master tools that did not exist two years ago, tools that update monthly, on threat of elimination. The timeline is set by corporate adoption cycles, not human learning curves.

The image of a worker racing to master AI tools while quietly dreading obsolescence has become one of the defining portraits of 2026's China. (Photo: Pexels)

Lying Flat as Protest - or Surrender

Not everyone is running. Some people are stopping entirely.

Frank Wang is a 28-year-old programmer in Chengdu. He spent most of 2025 deeply anxious about AI replacing him - watching every new model release, calculating every new capability against his own job description. Then, sometime around the end of last year, something in him shifted.

"I am now just 'lying flat,'" Wang told Rest of World, using the viral Chinese phrase - tang ping - that means doing the bare minimum required to keep your position. "If they fire me, they fire me. I will wait for some welfare handouts."

Tang ping emerged in 2021 as a cultural rebellion against China's grinding work culture - the "996" grind of working 9am to 9pm, six days a week, that was long celebrated by tech executives as a sign of commitment. Young workers, exhausted and priced out of housing and family formation, started quietly refusing. They did their jobs and went home. They stopped chasing promotions. They disengaged.

AI anxiety has given tang ping a new dimension. Wang's version is not just a rejection of corporate hustle. It is a philosophical surrender to a force he cannot control. What is the point of upskilling for a tool that will be obsolete in eighteen months anyway? What is the point of fighting for job security when the definition of "valuable worker" rewrites itself every quarter?

The macroeconomic implications of mass tang ping are significant. Li Chen, a researcher at Beijing-based think tank Anbound, told Rest of World that when large numbers of middle-class workers fear AI-driven job loss, they stop spending and start hoarding savings. Consumer spending contraction is one of the main obstacles to China's economic recovery - and AI anxiety is feeding directly into it. The government is simultaneously pushing AI as the engine of future growth while creating the conditions for the spending paralysis that undermines current growth.

"Lying flat" - doing the bare minimum - has become both protest and coping strategy for young Chinese workers who see no clear path through the AI disruption. (Photo: Pexels)

The Education Trap - and Why Young People Are Most Exposed

Jack Linzhou Xing is a postdoctoral fellow at the Fairbank Center for Chinese Studies at Harvard University. He researches the sociology of technology in China, and when he looks at the AI anxiety data, he sees a structural problem that cannot be solved by individual resilience or corporate training workshops.

"As AI reshapes the job market, the challenges China faces in terms of changing the education system structurally - combined with social pressure on individuals to position themselves for the future - could make the anxiety facing Chinese youth even more acute than in the West," Xing told Rest of World.

The problem is the gap between what China's education system produces and what the AI economy needs. The gaokao - China's famously grueling college entrance exam - has historically funneled students toward credentials, toward safe, respectable, middle-class careers in programming, accounting, law, finance, and sales. Those are precisely the sectors the Peking University study found declining in hiring as AI capabilities expand.

Changing an education system is a generational project. You cannot pivot a national curriculum in two years. The young people entering the workforce now are carrying degrees optimized for a job market that is being rewritten faster than any institution can track. And the social stigma of job loss in China - where employment is still deeply tied to family reputation, marriage prospects, and social standing - amplifies the psychological toll.

Youth unemployment at 15 to 19 percent is not just an economic statistic. It is the background radiation of an entire generation trying to find solid ground in shifting terrain.

The View from Beijing - AI Triumphalism vs. Ground Reality

China's government is not conflicted about AI. The official position is unambiguous: AI is the future, AI will make China stronger, AI adoption is a national priority. State media runs stories about AI-powered factories, AI-driven healthcare, AI administrative services. The narrative of technological progress is a core part of how the Chinese Communist Party presents itself as competent custodians of the nation's destiny.

That triumphalism coexists uneasily with what Xing calls "the growing gap between China's narrative of technological progress and the reality many workers experience on the ground, where competition is intensifying even as the country races ahead in global tech development."

The government's response to AI-driven job anxiety has been largely rhetorical. Officials invoke "new quality productive forces" - a Xi Jinping phrase - to argue that AI creates more jobs than it destroys. Training initiatives have been launched. Digital skills programs have been funded. But the pace of workforce transformation and the pace of institutional support are not moving in sync.

Meanwhile, corporate China is accelerating. Companies that want to remain competitive are adopting AI tools rapidly, restructuring workflows, and quietly reducing headcount in roles that can be automated. They are not waiting for government transition programs to catch up.

The Human Cost That Doesn't Show Up in GDP

Economic anxiety has a body. It shows up in sleep disorders and cardiovascular stress. It shows up in deferred marriages and declining birth rates - already a national crisis in China. It shows up in the mental health data that lags years behind the conditions that create it.

Lambert Li does not sleep well. He spends hours on evenings and weekends testing AI tools he did not ask to need. Betty Lai sits in AI training workshops she was not given work hours to attend. Frank Wang has stopped caring whether he gets fired because the anxiety of caring became unsustainable. These are not dramatic stories. They are the texture of an ordinary professional life being quietly deformed by forces no one in the room controls.

There is something philosophically unnerving about the Shenzhen queue. The people waiting outside Tencent were not protesters. They were not asking for anything to change. They were asking to be granted access to the tool that might let them keep what they already had. That is a posture of compliance dressed up as agency. Survival behavior disguised as ambition.

"You can get eliminated anytime. How can you not be anxious?" - Lambert Li, software developer, Shanghai, to Rest of World, March 2026

The Squid Game comparison is not accidental. The Netflix series - which became a global phenomenon precisely because it dramatized economic precarity through lethal competition - resonates in China for a reason. The show is about what happens when the rules of survival change, when the old pathways to stability stop working, and when the system that governs your fate is indifferent to your humanity. The people in those queues in Shenzhen understand the metaphor instinctively. They are not watching a drama. They are in one.

The queue outside Tencent's Shenzhen headquarters - nearly 1,000 people waiting to install an AI tool - became an accidental symbol of an entire generation's fear. (Photo: Pexels)

What Comes Next - and What Nobody Is Saying

The uncomfortable truth at the center of China's AI anxiety story is that no one has a satisfying answer to the structural problem. Not the government, not the corporations, not the economists.

Retraining programs are the standard political answer, and they are not wrong, but they are insufficient at the scale and speed required. The skills gap in AI is not a training deficit that can be closed by enrolling in a night class. It involves fundamental shifts in how value is created and who creates it. The workers most at risk - accountants, editors, entry-level coders, sales staff, customer service workers - are not going to become AI engineers. The jobs they are being replaced by are not going to be created in equivalent numbers anytime soon.

Li Chen's warning about precautionary saving and consumer spending contraction points to a macro feedback loop that the official AI optimism narrative struggles to address. If AI anxiety suppresses consumption, the productivity gains from AI adoption may be partially offset by demand-side stagnation. China's economic planners are trying to thread a needle between accelerating the technology that causes the anxiety and managing the anxiety that undermines the economy.

Meanwhile, the young workers making individual calculations are not waiting for policy solutions. They are making rational choices under uncertainty. Some are upskilling aggressively. Some are lying flat. Some are leaving the workforce altogether, relying on family support, retreating from the competition they see as unwinnable. These are not failures of character. They are rational responses to structural conditions.

Jack Linzhou Xing at Harvard has watched this pattern across his research and sees the cultural stakes clearly. "AI anxiety is fueled by a growing gap between China's narrative of technological progress and the reality many workers experience on the ground," he said. That gap - between the triumphant story a society tells about itself and the lived experience of its people - is historically unstable. It generates resentment. It generates cynicism. It generates political energy that goes somewhere, eventually, whether authorities want it to or not.

The queue in Shenzhen was orderly. Everyone waited their turn. Everyone wanted the same thing. That is not a given. A generation that has "lying flat" as a cultural touchstone is one that has already lost faith in effort as a reliable path to stability. What replaces that faith when the AI tools fail to deliver the job security people queued for? Nobody in Beijing or Silicon Valley or Shenzhen is willing to answer that question out loud.

The Squid Game does not have a good ending for most players. Lambert Li knows that. He keeps testing the tools anyway. Because what else do you do when you are already in the game?