By Jia Fei-mao | 2025-04-03

China's artificial intelligence (AI) sector has advanced rapidly, raising concerns that it could be used as a tool for censorship.

The launch of Chinese AI models DeepSeek and Manus not only boosted the Chinese and Hong Kong stock markets, but also became a symbol of Beijing's push for AI competitiveness.

An editorial published in the Chinese state-run Global Times on February 19 argues that China's rapid AI breakthroughs and adoption -- what it calls "Chinese acceleration" -- are challenging the narrative that China's economic growth has peaked.

The article further asserts that through initiatives like Belt and Road, digital cooperation and international open-source projects, China's AI influence will continue to expand globally.

However, the flip side of this AI boom is strict content regulation.

Observers warn that the Chinese government may be using AI to systematically rewrite history, erase records of human rights violations and censor criticism of the Chinese Communist Party (CCP).

OpenAI on March 13 sent a letter to the White House stating that "because DeepSeek is simultaneously state-subsidized, state-controlled, and freely available, the cost to its users is their privacy and security."

The letter suggested that the US government consider banning models developed by DeepSeek and other institutions backed by the Chinese government.

Global information freedom

Former US Representatives Loretta Sanchez and Greg Walden also warned of the risks posed by China's AI dominance in an opinion piece in The Hill published March 12.

"The winner of this race will steer the global digital ecosystem and decide what's amplified, what's buried and which set of values become the norm," they wrote.

When asked about the Tiananmen Square massacre, Tencent's AI model Hunyuan claimed that "no one was killed, and there was no massacre." Meanwhile, Alibaba's Qwen deleted the question without responding, wrote the authors, citing a study from The American Edge Project.

China's AI avoids discussing Tiananmen Square and refuses to comment on Chinese leader Xi Jinping, yet freely criticizes US politics and the US president.

This is not merely a technical issue but a built-in "double standard" designed to align with the CCP narrative, they added.

Sanchez and Walden further warned: "If China's AI tools become the default, the CCP will wield unprecedented geopolitical influence, capture trillions in economic value, undermine free expression across the globe, and establish a culture molded by control, censorship and propaganda, while rewriting the past."

China has been tightening AI regulations, requiring AI-generated content to "adhere to socialist core values and refrain from producing content that incites subversion of state power, overthrows the socialist system or endangers national security and interests," Tzeng Yi-suo, an associate researcher at Taiwan’s Institute for National Defense and Security Research, told Focus.

The rise of China's AI is not just a technological issue -- it's about the future of global information freedom, he said.

If China's AI sets the global standard, censorship and data surveillance could extend beyond national borders, reaching every corner of the world, said Tzeng.