The rapid and relentless advancement of artificial intelligence is not merely a technological revolution; it is fundamentally reshaping the global geopolitical order. According to a stark and compelling analysis by Alan Kohler for ABC News Australia, the AI race has effectively returned the world to a bipolar structure reminiscent of the Cold War, but with a new, arguably more pervasive technological battleground. In this new paradigm, the United States and China have emerged as the undisputed, dueling superpowers of the AI age, leaving the rest of the world—including Europe, Japan, and Australia, effectively “inside the cage,” forced to navigate a landscape defined and dominated by these two titans.
Kohler’s analysis highlights a critical reality: the scale of investment, the concentration of talent, and the sheer volume of data required to compete at the frontier of AI development have created an insurmountable barrier to entry for almost all other nations. The competition is no longer multipolar; it is a direct, head-to-head contest between Silicon Valley’s tech ecosystem and Beijing’s state-backed initiatives. This bipolarity has profound implications for global governance, economic competitiveness, and national security, as the decisions made in Washington and Beijing increasingly dictate the technological trajectory of the entire planet.
The Concentration of AI Power
The dominance of the US and China in the AI sector is evident across almost every metric. The United States currently leads in developing the most advanced frontier models, driven by massive investments from companies like OpenAI, Anthropic, Google, and Microsoft. These firms benefit from deep capital markets, a culture of aggressive innovation, and access to the world’s most sophisticated semiconductor technology. However, China is rapidly closing the gap, leveraging its vast domestic market, immense data resources, and a coordinated national strategy to accelerate its AI capabilities.
As highlighted in the ABC News report, recent data from OpenRouter underscores the speed of China’s ascent. In the week ending July 6, 2026, eight of the top ten AI models by output token volume were Chinese, with the top five Chinese models producing nearly three times the token volume of their US counterparts. This surge is driven by a combination of rapid technical progress by Chinese startups like DeepSeek and Moonshot, and aggressive pricing strategies that make their models highly attractive to developers globally. The sheer scale of China’s AI deployment demonstrates that the competition is not just about creating the smartest model, but also about achieving the widest adoption and integration.
(Related: China’s AI Distillation Campaigns Exposed: A Growing Threat to Frontier Models)
The “Cage” of Bipolarity: Implications for the Rest of the World
For nations outside the US-China duopoly, the current AI landscape presents a daunting challenge. Kohler’s metaphor of being “inside the cage” aptly describes the feeling of constraint and dependency experienced by other countries. They are increasingly reliant on AI infrastructure and foundational models developed by either American or Chinese firms, raising concerns about data sovereignty, technological lock-in, and the potential for geopolitical coercion.
Europe, for instance, has attempted to assert its influence through regulation, most notably with the comprehensive EU AI Act. However, while regulation can shape how AI is deployed within its borders, it does not create the indigenous technological capabilities necessary to compete at the frontier. Similarly, countries like Australia and Japan are investing in AI research and development, but their efforts are dwarfed by the scale of the US and Chinese ecosystems. The reality is that for the foreseeable future, the foundational technologies driving the global economy will be controlled by a handful of companies based in just two countries.
The Failure of Global Governance
The bipolar nature of the AI race also severely complicates efforts to establish effective global governance. The UN Independent International Scientific Panel on AI recently warned that the current governance landscape is “fragmented and ineffective.” This fragmentation is largely a result of the deep strategic mistrust between the US and China. Both nations view AI leadership as a critical component of national security and economic supremacy, making them reluctant to agree to binding international regulations that might constrain their development efforts.
The lack of a unified global approach to AI governance increases the risk of an unconstrained arms race, with potentially destabilizing consequences. Issues such as the militarization of AI, the proliferation of deepfakes and disinformation, and the potential for catastrophic accidents require coordinated international action. However, as long as the US and China remain locked in a zero-sum competition, achieving meaningful consensus on these critical issues will remain elusive. The world is effectively relying on the self-restraint of the two superpowers to manage the profound risks associated with advanced AI.
(Related: China Intensifies AI Export Controls Amidst Global Competition)
Navigating the New Bipolarity
The ABC News analysis provides a sobering assessment of the current geopolitical reality. The AI revolution has not democratized technological power; instead, it has concentrated it in the hands of two dueling superpowers. For the rest of the world, the challenge is no longer about winning the AI race but about navigating the complex and often perilous landscape defined by the US-China rivalry. This requires a delicate balancing act: leveraging the benefits of advanced AI technologies while mitigating the risks of dependency and ensuring that national interests are protected in a world increasingly shaped by the decisions of Washington and Beijing. The era of the AI superpowers has arrived, and the cage is firmly locked.