Can AI Shape a Post-Capitalist Economy?

Overview

The debate over artificial intelligence’s economic future is no longer confined to Silicon Valley boardrooms or academic philosophy seminars. It has reached a fever pitch as competing national strategies emerge, each promising a different trajectory for global power and human welfare.

While some analysts warn that open-source AI models from China could lead to a form of “full AI communism,” others argue this narrative overlooks the deeper structural shifts at play. The rise of large language models (LLMs) challenges not just labor markets, but the very foundations of capitalist property relations and innovation incentives.

This analysis explores whether AI will reinforce existing hierarchies or provide the tools to construct a post-capitalist economy, where human need—not profit—drives technological development.

The Ghosts of Aristotle and Marx

The idea that autonomous machines could dismantle property, hierarchy, and inequality is one of the oldest dreams in Western political thought. In his Politics, Aristotle referenced a fantasy from the Iliad: if shuttles could weave and quills could play harps on their own, “master-craftsmen would have no need of assistants and masters no need of slaves.”

This vision was later revived by Karl Marx, who imagined that automation would replace intellectual property with “socialised knowledge.” He termed this merger of collective human understanding the “general intellect,” a force capable of blowing the foundations of capitalism sky high.

Today, these philosophical specters haunt modern tech leaders. In July, Dean Ball, head of strategic futures at OpenAI, expressed discomfort with China’s advocacy for open-source AI. He characterized Beijing’s approach—treating AI as a public good provided by the state—as a dystopian step toward “full AI communism.”

“AI is a public good which will ultimately be provided by the state as a kind of ‘digital public infrastructure’,” Ball noted. “This strikes me as a dystopian hellscape.”

However, interpreting China’s AI strategy solely through a Cold War lens misses its economic logic. Beijing is not necessarily seeking ideological purity; it is countering American digital hegemony with its own form of technological sovereignty.

A Contest of Digital Hegemonies

The current AI landscape is defined by two rival strategies for global domination, rather than a clash between capitalism and communism.

The US Model: Proprietary Cloud Capitalism The United States strategy centers on hoarding computing power and granting access only via internet-connected cloud services. This approach ensures that the rest of the world remains dependent on American software, cloud infrastructure, and regulatory frameworks. It is a system of info-capitalism where value is extracted through subscription and data exploitation.

The Chinese Model: Open Source as Strategy China’s counterplay involves releasing open-source models like Kimi, Qwen, and Deepseek. This allows users to run AI locally on their own servers, effectively firewaliling their data from foreign providers. While Ball views this openness with alarm, it serves a pragmatic purpose: preventing data colonialism.

Beijing gambles that if AI becomes a “general purpose technology,” the People’s Republic of China can win supremacy in downstream innovations. These include brain-machine interfaces, bioengineering, and the “low altitude economy”—a sector encompassing delivery drones and autonomous logistics.

Who wins this contest will determine the trajectory of information technology for the rest of the century. But what if there is a third way?

The End of Capitalism as We Know It

Regardless of which nation dominates AI development, the technology itself poses an existential threat to capitalism’s core mechanisms. Large language models already exhibit the features of Marx’s “general intellect,” embodying and learning from nearly every item of human knowledge ever produced.

Frontier models now surpass the output of trained knowledge workers in law, coding, and design. Yet, unlike previous industrial revolutions, AI does not merely displace manual labor—it threatens the roles of entrepreneurs, innovators, and capitalists themselves.

Economist Philippe Aghion has argued that AI may not only automate entrepreneurial skill but also sever the link between innovation and profit. If a machine can emulate and improve upon a filed patent blueprint instantly, the incentive to innovate under current intellectual property regimes vanishes.

The job-destroying quality of AI is unprecedented. While vaudeville theaters were replaced by Hollywood, creating new industries, few can confidently say what will replace the roles AI eliminates. Capitalism relies on a class of human workers, innovators, and entrepreneurs; if AI renders them obsolete, the system cannot function as it currently does.

Marx described the advance of industry as moving humans “to the side” of the production process, turning operatives into supervisors. AI is doing this for brain work, achieving high standards of automation in less than five years.

Designing a Post-Capitalist Future

The rational question is no longer how Britain or America builds a sovereign AI to rival its competitors. The question is: How does humanity shape AI toward the fulfilment of human need?

To answer this, we must look beyond the binary of US proprietary control and Chinese state-led openness. Both systems replicate specific social realities: one mercantilist and secretive, the other focused on global hegemony.

A third path requires deliberate intervention to divert technological development toward social-democratic, cooperative, or even anarchist goals. This involves several key steps:

• Decoupling Work from Wages: Implement universal basic services and incomes to sustain populations displaced by automation.
• Promoting Cooperative Models: Encourage non-profit and cooperative business structures that prioritize social utility over shareholder value.
• Replacing Market Allocation: Use AI’s computational power to replace inefficient market mechanisms with rational planning for resource distribution—a feat Soviet-style planning never achieved due to lack of data.

Critics may argue that pursuing such a vision is naive in a world dominated by competitive nation-states. However, technological determinism is not inevitable. If society is willing to trail six months behind models like Claude or Kimi, it can use those very tools to design an un-ownable AI model with humanism and universal values built into its core.

The choice remains stark: delay and obstruct AI adoption in the name of preserving social justice, or embrace it in a radically altered form. The latter offers a chance to realize the ancient dream of Aristotle and Marx—a society where machines serve human flourishing, not capital accumulation.