China has surpassed the United States in 69 out of 74 critical and advanced technologies, according to the latest data from the Australian Strategic Policy Institute (ASPI). Peter Hartcher’s analysis in the Sydney Morning Herald frames the findings in the context of the AI race. The tracker, updated in June 2026, provides a stark assessment of the shifting balance of technological power, a trend that is now becoming increasingly apparent in the race for artificial intelligence supremacy.
What 69 Out of 74 Actually Means
In an opinion piece published in the Sydney Morning Herald on August 4, political and international editor Peter Hartcher highlighted the ASPI findings as evidence of a “structural shift under way in global technological power.” Hartcher argues that while the US currently maintains a lead in frontier AI models, that advantage is rapidly eroding as Chinese firms demonstrate an ability to produce highly capable models at significantly lower costs.
Hartcher cites Dewardric McNeal, an American business adviser at Longview Global writing for CNBC, who argues that the pattern of Chinese AI breakthroughs represents something far more consequential than isolated successes. “Viewed collectively, they reveal something far more consequential than the emergence of several successful Chinese AI companies,” McNeal writes. “They demonstrate that China has cultivated a frontier AI ecosystem capable of repeatedly producing world-class capabilities across multiple firms.” In other words, it is a systemic challenge to US dominance, not an episodic one.
The analysis points to recent breakthroughs by Chinese AI startups as concrete evidence of that systemic shift. DeepSeek’s V3 model, for example, achieved performance comparable to OpenAI’s GPT-4o but was reportedly developed at a cost of just $5.58 million, compared to the estimated $100 million-plus required for GPT-4. Similarly, Moonshot AI’s Kimi K3 was recently ranked as the world’s third most intelligent AI product by Artificial Analysis, yet it operates at roughly one-third the cost of its US rivals.
This cost advantage is proving decisive in the enterprise market. As US companies grapple with the ballooning costs of deploying proprietary models from OpenAI, Anthropic, and Google, many are turning to cheaper, open-weight alternatives developed in China. The ability of Chinese firms to consistently produce world-class capabilities across multiple companies suggests a mature and highly efficient AI ecosystem, rather than isolated successes.
Outspent but Not Outpaced
Hartcher also contrasts the investment strategies of the two nations. The US AI sector is currently experiencing a massive influx of capital, with the four largest tech firms (Google, Amazon, Microsoft, and Meta) investing a combined $1.1 trillion over the past three years. The total market capitalization of AI-related companies in the US has swelled to $27 trillion, representing 40% of the entire share market.
In contrast, China’s AI-related market capitalization stands at just $4 trillion, and its total AI investment last year was a comparatively modest $12.4 billion, according to the Stanford AI Index. Despite this massive disparity in funding, Chinese firms are capturing a disproportionate share of global revenue. Goldman Sachs recently rated the Chinese AI sector a “buy,” noting that it accounts for 10% of global AI capitalization but generates 16% of global revenue.
The analysis draws parallels between the current US AI boom and historical investment manias, such as the railway boom of the 1840s and the dot-com bubble of the late 1990s. The Bank of International Settlements has warned that the current frenzy of investment in AI may exceed what commercial returns can justify, potentially leading to a painful market correction.
China, meanwhile, may have already absorbed its economic pain. Hartcher suggests that President Xi Jinping’s deliberate repression of the tech and real estate sectors five years ago was a preemptive move to avoid the kind of convulsive boom-and-bust cycle that currently threatens the US economy. As the AI race enters its next phase, China’s frugal, highly efficient approach to innovation may prove more sustainable than America’s capital-intensive strategy.
The Limits of the Export Control Strategy
The ASPI data also has implications for US policy. If China leads in 69 of 74 critical technologies, the US strategy of imposing export controls to slow Chinese progress may be achieving less than Washington hopes. Export controls on advanced chips have undoubtedly created friction for Chinese AI developers, but they have not prevented the emergence of world-class models from DeepSeek, Moonshot, Alibaba, and Zhipu AI. The gap between hardware constraints and software output suggests that Chinese engineers are finding ways to work around the restrictions that US policymakers did not anticipate.
The broader question raised by the ASPI tracker is whether the US can reverse a technology gap that has been building for years. The answer likely requires not just defensive measures like export controls, but a sustained, coordinated investment in basic research, engineering education, and domestic manufacturing capacity. Without that, the US risks winning individual battles in the AI race while losing the broader technological competition.
For China, the ASPI data is both a validation and a warning. EastFrontier has previously reported on China producing 279.8 billion chips in H1 2026 and how US export controls are part of the reason why. Leading in 69 of 74 critical technologies is a remarkable achievement, but it also raises the stakes for maintaining that lead. As US policymakers absorb the full implications of the tracker’s findings, the political pressure to respond with more aggressive countermeasures will intensify. The question is no longer whether China is a technological peer competitor, the data makes that case conclusively, but how Washington chooses to respond to that reality.