US China AI: The Great Unbundling of Global Power

How artificial intelligence is reshaping the geopolitical landscape and forcing humanity to confront the limits of bundled human capabilities

The rivalry between the United States and China over artificial intelligence represents more than a technological competition—it embodies the most profound unbundling of human capability bundles in history. As the world's superpowers entangle in what many have dubbed an "AI arms race," we're witnessing the systematic isolation and amplification of intelligence, creativity, and strategic thinking beyond human capacity. This competition isn't just about chips and algorithms; it's about which civilization will control the levers of the Great Unbundling.

Bottom Line Up Front

The US-China AI gap has dramatically narrowed to just 3-6 months in model performance as of 2025, with China's breakthrough DeepSeek R1 model matching OpenAI's capabilities at a fraction of the cost. While the US maintains advantages in private investment and talent, China's state-directed approach and efficiency innovations are challenging fundamental assumptions about American technological dominance. The competition is reshaping global power dynamics through the lens of J.Y. Sterling's Great Unbundling framework—where traditional human capability bundles are being systematically separated and automated.

The Unbundling Engine: How AI Competition Accelerates Human Obsolescence

In "The Great Unbundling," Sterling argues that capitalism serves as the primary engine driving the separation of human capabilities. The US-China AI race perfectly illustrates this dynamic. China's strategy is characterized by centralized planning, with direct state funding funneling into specific AI projects, while the US has primarily relied on private enterprises to drive AI advancements.

This competition accelerates unbundling in three critical ways:

Intelligence Separation: DeepSeek's R1 model demonstrates how reasoning capabilities—once bundled exclusively within human consciousness—can now be isolated, improved, and deployed at costs 27 times lower than competing US models. The model's success despite US export controls shows that intelligence is becoming increasingly divorced from the hardware and capital traditionally required.

Strategic Thinking Automation: The contest extends beyond military and economic advantages to four domains with world-altering significance: conflict norms, state power, emerging bioethics, and catastrophic risks. Both nations are unbundling strategic decision-making from human judgment, creating AI systems that can plan military operations, economic policies, and social control mechanisms.

Creative Problem-Solving Commoditization: DeepSeek managed to turn resource restrictions into innovation, prioritizing efficiency and resource-pooling over brute-force computational power. This demonstrates how creativity—traditionally a uniquely human bundle—is being systematically optimized and replicated.

Current State: The Narrowing Capability Gap

US Advantages Under Pressure

The US produced 40 AI models of note in 2024 compared to China's 15, and saw $109.1 billion invested compared to China's $9.3 billion. However, these traditional metrics obscure a more complex reality.

The Stargate Response: President Trump announced the $500 billion Stargate Project in January 2025, involving OpenAI, Oracle, SoftBank, and MGX, designed to secure American leadership in AI and create hundreds of thousands of jobs. This massive investment represents America's attempt to maintain its bundled advantage in capital, talent, and infrastructure.

Export Control Limitations: US export controls attempted to deny China an AI computing ecosystem, but DeepSeek's success demonstrates the limited effectiveness of these restrictions in slowing China's AI progress.

China's Efficiency Revolution

Chinese generative AI models likely lag behind US competitors by approximately three to six months as of early 2025, but this gap is rapidly closing. More significantly, China is pioneering a new approach to AI development that challenges fundamental assumptions about resource requirements.

Cost Innovation: DeepSeek claims its R1 model was developed using less than $6 million in computing power—a fraction of the multibillion-dollar AI budgets of US tech giants. This represents a fundamental unbundling of AI capability from massive capital requirements.

Open Source Strategy: DeepSeek-R1 is thrilling scientists as an affordable and open rival to reasoning models, making advanced AI capabilities accessible to researchers worldwide. China's embrace of open-source development unbundles AI progress from proprietary corporate control.

The Great Unbundling in Action: Four Critical Domains

1. Economic Value Unbundling

The AI competition is systematically separating economic value creation from human labor. Goldman Sachs research indicates that 300 million jobs are exposed to automation through AI advancement. Both nations are racing to capture the economic benefits while their populations face unprecedented displacement.

2. Military Capability Separation

AI progress in military applications is creating distinct challenges for each superpower's ambitions to lead internationally, with the US developing a coherent approach to AI in warfare while China aggressively pioneers techno-authoritarian systems. Traditional military advantage—once bundled with human judgment, courage, and tactical thinking—is being systematically automated.

3. Governance and Social Control

China leads in leveraging AI for state power, with Beijing pioneering techno-authoritarian systems that separate social control from human oversight. This represents the unbundling of governance from human moral judgment and democratic accountability.

4. Scientific Discovery Acceleration

Oracle's Larry Ellison contends that Stargate could lead to AI-facilitated production of mRNA vaccines against cancer, designed robotically in about 48 hours. Scientific breakthrough—traditionally requiring bundled human creativity, intuition, and expertise—is being systematically automated.

The Philosophical Challenge: Post-Human Competition

Sterling's framework reveals the deeper implications of US-China AI competition. We're not just witnessing technological rivalry; we're observing the death throes of human-centered civilization. The escalating AI competition poses significant threats to both nations and the entire world, potentially undermining global stability and leading toward dangerous technological brinkmanship.

The End of Human Exceptionalism

The sudden emergence of DeepSeek has challenged assumptions about US dominance in AI and raised fears that sky-high market valuations may be detached from reality. This revelation forces a fundamental question: if a small Chinese startup can match the capabilities of Silicon Valley giants at a fraction of the cost, what does this say about the value of human expertise, capital accumulation, and institutional advantage?

The Collapse of Traditional Competitive Advantage

Both nations built their strategies around bundled advantages—the US through capital markets, talent attraction, and technological infrastructure; China through state planning, manufacturing capacity, and population scale. The performance gap between leading US and Chinese AI models has shrunk from 9.3% in 2024 to 1.7% in early 2025. These traditional bundles are rapidly losing their competitive significance.

The Great Re-bundling Response: National Strategies

America's Re-bundling Through Capital

The Stargate Project represents America's attempt at re-bundling—combining massive capital investment, international partnerships, and technological integration. The venture plans to invest up to $500 billion in AI infrastructure by 2029, with initial construction of 10 data centers in Abilene, Texas.

However, this approach faces fundamental challenges:

• Dependency on Private Capital: Wall Street interpreted the Stargate project as encouraging AI sector growth, but questions remain about whether promised funding will materialize.
• Alliance Vulnerabilities: Asian nations must juggle research partnerships with China without jeopardizing collaboration with US institutions, highlighting the complexity of technological alliances.

China's Re-bundling Through Efficiency

China's response involves re-bundling through state coordination, resource efficiency, and open collaboration. Government policies, generous funding, and a pipeline of AI graduates have helped Chinese firms create advanced language models rivaling US capabilities.

This strategy includes:

• State-Directed Innovation: Direct government funding and coordination unbundled from market pressures
• Efficiency Focus: DeepSeek aimed for accurate answers rather than detailing every logical step, significantly reducing computing time while maintaining effectiveness.
• Global Talent Leverage: Almost half of all top AI researchers globally (47%) were born or educated in China.

Future Implications: The Path Forward

Scenario 1: Continued Competition and Fragmentation

We're witnessing the early stages of a digital iron curtain, where AI technologies will increasingly align with either the US or Chinese sphere of influence. This path leads to:

• Duplicated research efforts and resource waste
• Technological balkanization of the global internet
• Increased risk of AI-powered conflicts

Scenario 2: Collaborative Re-bundling

The world's superpowers need to work together to make sure AI benefits humanity, requiring joint identification and mitigation of threats. This would involve:

• Shared safety standards and governance frameworks
• Joint research on catastrophic risk mitigation
• Coordinated approaches to AI's economic disruption

Scenario 3: Leapfrog Innovation by Third Parties

DeepSeek's low-cost, open-source model could empower emerging economies' own AI innovation and entrepreneurship. Smaller nations and private actors might develop superior approaches by learning from both superpowers' successes and failures.

The Human Agency Question

Sterling's framework suggests that while the Great Unbundling appears inevitable, human agency remains in how we shape the response. The US-China AI competition presents three critical choices:

1. Acceleration vs. Governance: The US risks ceding ground to China by building high regulatory walls while not running fast enough to stay ahead. The challenge is balancing safety with competitive necessity.

2. Openness vs. Control: DeepSeek's open-weight model highlights shortcomings in America's AI strategy, particularly around open-source development. Nations must decide whether to share AI benefits or hoard them for competitive advantage.

3. Human Integration vs. Replacement: Both nations can choose to develop AI that augments human capabilities rather than simply replacing them. This requires conscious effort to maintain human agency in critical decisions.

Actionable Insights for the AI Age

For Policymakers

• Develop Adaptive Governance: Create regulatory frameworks that can evolve with rapidly changing AI capabilities
• Invest in Human Re-bundling: Focus resources on developing uniquely human skills that complement rather than compete with AI
• Build International Cooperation: Collaborate on developing global frameworks for regulating advanced AI models rather than erecting technological barriers.

For Organizations

• Embrace Efficiency Innovation: Learn from DeepSeek's approach to achieving more with less computational power
• Plan for Capability Shifts: Prepare for rapid changes in competitive advantage as AI unbundles traditional skill requirements
• Develop AI Literacy: Understand both opportunities and limitations of AI systems in your domain

For Individuals

• Focus on Re-bundling Skills: Develop capabilities that combine multiple domains in ways that are difficult to automate
• Stay Informed: Monitor developments in AI policy and capability to make informed career and life decisions
• Engage in AI Governance: Participate in discussions about how AI should be developed and deployed

Conclusion: Navigating the Unbundled Future

The US-China AI competition represents humanity's first confrontation with the limits of our bundled capabilities. Neither side likely will gain a decisive edge, with leading US and Chinese labs pushing forward in parallel toward artificial general intelligence. The question isn't who will win this race, but whether humanity can navigate the transition with purpose and dignity intact.

Sterling's Great Unbundling framework reveals that this competition is simultaneously destroying old forms of human value while creating opportunities for new kinds of re-bundling. The nations and individuals who succeed will be those who consciously choose how to integrate AI capabilities with essentially human qualities—consciousness, purpose, ethical judgment, and collaborative wisdom.

The age of bundled human dominance is ending. The age of conscious human choice in our technological future is just beginning.

---

Ready to explore the deeper implications of AI's impact on human civilization? Discover how "The Great Unbundling" framework explains not just the US-China AI competition, but the fundamental transformation of human value in the age of artificial intelligence. Learn more about J.Y. Sterling's comprehensive analysis of how we can navigate this unprecedented transition.

For more insights on AI geopolitics and the future of human capability, explore our AI Arms Race analysis and join our newsletter for regular updates on the Great Unbundling's progress.