【By Liu Bai, Observers Network】Schmidt, the former CEO of Google who has been closely watching China's artificial intelligence (AI) development, is anxious again. Recently, in a podcast program, he expressed concern that most governments may eventually use Chinese AI models due to cost considerations.

"This creates a strange situation: the most advanced model in the United States is closed-source, while the most advanced model in China is open-source," Schmidt said in a podcast titled "Moonshots" released on November 11. "The geopolitical issue involved is that the open-source model can be freely accessed, whereas the closed-source model requires payment for use."

"Therefore, governments and countries without the financial resources of the West will ultimately adopt Chinese models, not because they are better, but because they are free," Schmidt said.

Open-source AI models allow anyone to use, modify, and share the software for any purpose. Supporters believe that open-source can accelerate technological development, as anyone can modify and redistribute the code. Advocates of closed-source models argue that closed-source is safer because its code remains confidential.

"Business Insider" reported that with Chinese AI models such as DeepSeek and Alibaba's Tongyi Qianwen 3 becoming popular, concerns about data privacy, national security, and U.S. competitive advantages have become widespread.

On November 6, Schmidt attended the U.S. Business Forum held at the Kaseya Center in Miami, Florida. IC Photo

Schmidt served as CEO of Google from 2001 to 2015, leading the company through its initial public offering in 2004. He is currently a founding partner of the venture capital firm Innovation Endeavours and runs a space startup called Relativity Space, with an estimated net worth of nearly $5 billion.

In recent times, Schmidt has repeatedly spoken publicly about China's technology industry, especially the development of AI.

On October 30, Schmidt and his AI team member Selina Xu published an article in The Atlantic, stating that the West needs to think about a key question: How did China, within just a few decades, rise from a technologically backward country to a superpower?

The article mentioned that in 1896, Li Hongzhang first arrived in the United States. He was amazed by the skyscrapers in New York, saying, "I have never seen such buildings before." Today, foreigners visiting China often feel a "future shock," marveling at the skyscrapers, high-speed railways, giant bridges, electric vehicles, super apps, and triple-fold smartphones in major Chinese cities.

Schmidt's feelings are not unique.

"How many Silicon Valley technologies are built on top of Chinese AI?" A Bloomberg article published on November 9 asked this question, citing multiple examples to illustrate that low-cost open-source AI models from China are not only attracting global users but also quietly gaining favor in Silicon Valley. The author believes that if the U.S. government wants to maintain its lead in the long term, it should first consider why Silicon Valley is turning to Chinese AI.

Analysts point out that "Sovereign AI" is increasingly receiving attention from the tech and political communities. It refers to a country's control and governance over artificial intelligence technology, data, and infrastructure.

Several industry executives, including NVIDIA CEO Jensen Huang and CEO of French AI startup Mistral Arthur Mensch, have stated that it is crucial for countries to build their own AI systems. Like Schmidt, Huang and Mensch are also supporters of open-source models.

In a podcast in March this year, Mensch compared AI to electricity in the early 20th century and said that countries that do not establish their own AI systems will see their funds flow to other countries.

"A hundred years ago, if you didn't build a power plant, you could only buy electricity from neighboring countries - this eventually leads to dependence, and such dependence is not good."

In February, Huang told officials at the World Government Summit in Dubai that countries need to strive to build their own Sovereign AI.

He said, "If I were a leader of a developing country, the first thing I would do is to incorporate your language and cultural data into your own large language model."

This article is exclusive to Observers Network. Reproduction without permission is prohibited.

Original: https://www.toutiao.com/article/7572107917738918441/

Statement: This article represents the views of the author and is not necessarily the official position of Observers Network. Welcome to express your opinion by clicking on the [Up/Down] buttons below.