How China wants to dominate the world of artificial intelligence. Xi Jinping announces 'major technological revolution with epochal impact'

Beijing wants to be a world leader in developing cutting-edge artificial intelligence, but it also wants firms to follow an increasingly complex set of rules, The New York Times reports.

In late January, China's leader Xi Jinping told a group of officials across the country that China was on the brink of a “major technological revolution with epochal impact.”

Artificial intelligence, he said, is as revolutionary as the steam engine, electricity and the Internet. However, China must not let the new technology “get out of control,” Xi warned, according to state media. China must act early and decisively, anticipating and preventing problems with caution and prudence, he continued.

Fast, but respecting the rules

Xi's remarks highlight a tension shaping China's technology industry, notes the New York Times. The Chinese leadership has decided that artificial intelligence will drive the country's economic growth in the next decade. At the same time, it cannot allow new technology to disrupt the stability of Chinese society and the Communist Party's control over it.

The government is therefore pushing Chinese AI companies to do two things at once: move quickly for China to outpace its international rivals and be at the forefront of technological change, but also comply with an increasingly complex set of rules.

When Zhipu AI, one of China's most promising artificial intelligence start-ups, filed for an IPO in Hong Kong last month, valuing the company at more than $6 billion, it warned investors about the substantial burden of complying with AI regulations. Similar to OpenAI, Zhipu develops artificial intelligence models and applications that use them, such as the ChatGLM chatbot.

China is pushing its AI companies to move quickly to outpace international rivals, but also to comply with a complex set of rules.

The Censorship System of the Communist Party

Among specific compliance requirements, the rules required Zhipu to act as a watchdog to prevent the spread of information the Chinese government deems illegal. Zhipu could not be sure that regulators would always find the company compliant, which “may expose us to significant legal, financial and operational consequences,” the company said in a filing.

The Internet was once perceived as an existential threat to the ruling Communist Party, but Beijing has brought it under control through a system of censorship and tight control over China's largest Internet companies. Artificial intelligence presents a similar dilemma: a transformative force that promises economic gains but also has the potential to undermine the party's grip on power.

Although the Trump administration has eased trade restrictions on certain powerful AI chips made by Nvidia, the main concern of China's artificial intelligence industry remains access to the semiconductors needed to power advanced systems.

Information control and data protection

But China's regulatory barriers add another layer of constraint and put Chinese firms such as Alibaba, DeepSeek and Zhipu in a difficult position: racing to develop AI systems as powerful as those of their foreign rivals, such as OpenAI and Anthropic, while complying with more regulations than their global competitors.

“What OpenAI and Alibaba are legally required to do in terms of pre-release testing is quite different,” said Scott Singer, a researcher at the Carnegie Endowment for International Peace.

California's new AI law, for example, requires companies to plan for the possibility that an artificial intelligence system will overcome human control or cause a large number of deaths. China's rules, by contrast, focus more on information control and data protection, Singer said.

From 2022, Chinese tech companies are required to provide the government with details about the algorithms behind features such as endless scrolling videos in apps like RedNote and the Chinese version of TikTok, Douyin.

Companies must explain to the government how their apps work and keep authorities informed as technology evolves. The more the authorities perceive an app's influence on public opinion, the more attention they pay to it.