‘My biggest fear’: Google ex-CEO Eric Schmidt warns China’s open-source AI may overtake the West

Former Google CEO Eric Schmidt has sounded the alarm over China’s rapidly expanding influence in artificial intelligence, warning that open-source Chinese AI models could soon dominate globally. Speaking on the Moonshots podcast released this week, Schmidt described this trend as one of his “biggest fears” for the future of technology and global security.

He noted that while leading US AI models are closed and expensive, China’s are open-source and free, giving them a major accessibility advantage. “This produces a bizarre outcome where the biggest models in the United States are closed source and the biggest models in China are open-source,” he said. “The geopolitical issue there, of course, is that open source is free and the closed source models are not free.”

‘Free’ Chinese models could shape global AI standards

Schmidt warned that economic disparities between nations could lead most countries to adopt Chinese AI systems—not because of superior performance, but because they are available at no cost. “The vast majority of governments and countries who don’t have the kind of money that the West does will end up standardising on Chinese models, not because they’re better, but because they’re free,” he added.

Open-source AI allows anyone to access, modify, and distribute the code, fostering innovation but also increasing risks related to data privacy and misuse. Closed-source systems, such as those developed by OpenAI or Google DeepMind, are often more secure but limited to paying users or licensed organisations.

Rise of China’s AI giants: DeepSeek and Alibaba’s Qwen3

China’s leading tech firms, including Alibaba and DeepSeek, have launched powerful open-source AI systems that are being rapidly adopted by developers worldwide. Alibaba’s Qwen3 model, in particular, has drawn attention for its accessibility and performance, challenging the dominance of Western players in the AI field.

However, these advances have also sparked debates around cybersecurity, national data control, and the geopolitical balance of power in the AI sector. Analysts suggest that by offering free, open models, China could extend its technological influence across developing countries, giving it leverage in global digital infrastructure.

The global debate over ‘sovereign AI’

Schmidt’s remarks add to a growing chorus of voices calling for “sovereign AI” — a concept that urges nations to develop and control their own AI systems. Nvidia CEO Jensen Huang and Mistral CEO Arthur Mensch have both argued that countries need independent AI models to maintain cultural, economic, and political autonomy.

Earlier this year, Huang told the World Governments Summit in Dubai that countries must build their own large language models to safeguard national interests. Mensch compared AI to electricity in the early 20th century, warning that nations that fail to establish domestic AI infrastructure could become dependent on others — echoing Schmidt’s concerns.

A growing geopolitical divide in AI

Schmidt, who led Google between 2001 and 2011 and now advises the US government on AI policy, said that the divide between open and closed AI systems could become a defining geopolitical fault line.

While the US focuses on protecting its intellectual property and ensuring safety, China’s open approach could enable it to set global standards, particularly in regions like Africa, the Middle East, and South Asia.

Experts warn that this shift could weaken Western influence in emerging markets and raise new challenges for international AI governance.

Eric Schmidt’s warning underscores the growing tension between innovation, accessibility, and control in the AI era. As China’s open-source models gain global traction, the question now facing policymakers is whether the United States and its allies can strike a balance between protecting their commercial interests and preventing technological dependency on Beijing.

The outcome, analysts say, could shape the next decade of global power dynamics — where the race for AI dominance may hinge not just on who builds the best models, but who shares them most freely.