BEIJING: China is paying greater attention to the cybersecurity risks posed by artificial intelligence as advanced AI models gain stronger hacking abilities.
US researchers have warned about future superintelligence and its possible threat to humanity. Chinese policymakers have focused more on immediate risks, such as deepfakes and child safety.
Recent AI hacking incidents, however, are changing that approach. China’s spy agency chief warned this week that AI could threaten national security.
Concerns grew after programmers in California said they used AI to create a cyberattack “worm”. They said the tool could potentially compromise millions of WeChat accounts if hackers deployed it.
WeChat ranks among China’s most widely used digital platforms.
“There are more similarities than differences when it comes to AI safety issues” in China and the United States, said Kevin Xu.
Xu founded the US-based technology hedge fund Interconnected Capital. He said AI security concerns rarely receive the same public attention in China.
Xu said AI agents could create major cybersecurity risks when they operate without human oversight.
China and US face shared AI risks
The growing concerns come as Chinese President Xi Jinping and US President Donald Trump prepare for talks on Thursday.
The leaders are expected to discuss several issues. AI safety could also feature as both countries compete for technological leadership.
Sun Chenghao of Tsinghua University’s Centre for International Security and Strategy said both countries are gaining a clearer understanding of AI risks.
“Both powers are developing a more realistic and shared understanding of these risks,” Sun said.
He said that growing awareness could create pressure for greater cooperation.
The two countries could cooperate on AI safety tests and systems that warn governments about emerging risks.
However, their fierce technological competition could limit such cooperation.
“We’re leading China in AI… and frankly I want to keep it that way,” Trump said.
He made the comment after reporters asked about calls from leading US AI laboratories for a coordinated slowdown in advanced AI development.
Sun said China also worries that the US could use security concerns to restrict its technological progress.
China focuses on immediate AI threats
China has already introduced rules covering several uses of artificial intelligence.
For example, regulations prohibit people from creating AI-generated avatars using another person’s likeness without consent.
Brian Tse, CEO of Beijing-based AI safety consultancy Concordia AI, said China has focused mainly on current AI risks.
“Chinese AI policy focuses more on real, present-day harms to society and people,” Tse said.
He said Beijing views AI as a general-purpose technology. The government aims to manage its risks through regulation.
However, recent improvements in AI capabilities have changed the debate.
Tse pointed to several warning signs. These include major AI incidents and the rapid growth of AI-powered automation.
He said these developments have pushed AI safety higher on Beijing’s policy agenda.
Public concerns grow over AI security
People in Shanghai expressed mixed views about the growing risks.
Xie Zongbo, 42, said the reported WeChat cyberattack experiment could ultimately help cybersecurity.
He said researchers could identify weaknesses, fix them and improve digital security.
Many people also supported stronger government regulation of AI.
A 50-year-old pharmaceutical worker identified as Jenny said businesses were unlikely to impose strict limits on themselves.
Jia Zijian, founder of Singapore-based Inspired AI, said he worries about AI models cracking complex passwords.
He warned that highly capable AI could potentially break critical passwords or national security codes.
Such capabilities could create serious risks for both China and the United States.
“If our capabilities are strong enough, they would feel exposed,” Jia said.
He added that China could face the same danger if US AI systems become more capable while Chinese defenses remain weak.
The debate highlights a growing challenge for governments. AI can strengthen cybersecurity defenses, but it can also give attackers powerful new tools.
As AI systems gain greater autonomy, China and the United States face increasing pressure to address these risks.
