(Bloomberg) -- China will promote the sharing of artificial intelligence advances while fighting attempts to monopolize the technology, an influential state newspaper wrote, setting the tone for President Xi Jinping's keynote remarks at the country's premier tech summit.
Most Read from Bloomberg
US Hits Iran With Strikes, Blockade as Trump Plans Hormuz Charge - Thailand Scraps Plan to End Visa-Free Entry for Indian Tourists
Beijing is opposed to using AI as a tool to pursue hegemony, dominate resources, and create tech barriers, People's Daily said in a commentary Tuesday. While it stopped short of specifics, the column emerged as Washington pressures prominent American labs such as Anthropic PBC to curtail foreign access to cutting-edge models, citing national security.
The state paper's commentary comes after Reuters reported that Beijing was considering similar measures to limit overseas access to China's most advanced AI models. It echoed remarks from Zhipu founder Tang Jie, in which the Tsinghua University professor argued that frontier AI should remain broadly accessible rather than controlled by select individuals, weighing in on a debate about the risks posed by ever more powerful models.
The People's Daily described an "oil mindset" to AI development in which countries treat data and computing as scarce resources. That erects barriers and spurs needless competition. The reverse "water" approach focuses on shared resources for a greater good, it added. And the paper warned of an "AI Iron Curtain" developing — borrowing a phrase often used to refer to the divide between Communist and democratic Western European nations.
"The development of artificial intelligence must never devolve into a monopolistic practice of self-imposed isolation, but should instead remain focused on the fundamental goal of serving humanity," the paper wrote in its editorial.
Chinese developers such as DeepSeek have largely embraced the open-source framework, which allows anyone to examine their codes and download to use for free, turbo-charging their global adoption and boosting usage. That has helped propel the global adoption of models such as Alibaba Group Holding Ltd.'s Qwen series and Zhipu's GLM series.
At the same time, AI labs including Anthropic and OpenAI and governments around the world are grappling with the growing capabilities of models that some experts warn will arm cyber-attackers and compromise national security.