As Silicon Valley's own scientists warn of human extinction, China pitches itself as the adult in the AI room

Beijing on Tuesday called for stronger "global governance" of artificial intelligence, saying all countries should work together to develop the technology for the benefit of humanity while preventing it from "getting out of control," amid a wave of stark warnings from leading AI researchers and industry figures about the technology's risks, according to a PTI report.

"We noted recent remarks by some in the US AI industry," Chinese Foreign Ministry spokesperson Guo Jiakun said, responding to questions about concerns raised by AI experts over the dangers posed by rapidly advancing systems.

"As AI develops at an incredible speed, it's all the more important to stay committed to developing AI for the positive, for good, and for the benefits of the people, to calibrate regulation and governance on AI, and to promptly strengthen safeguards to prevent it from getting out of control," he said.


Guo described AI as "the collective wisdom and precious asset shared by all humanity," stressing that its development and governance require international cooperation. "AI development and governance require the practice of true multilateralism, with all countries working together to build a global governance framework on broad-based consensus," he said.

"No single country or enterprise can develop effective solutions on its own," he added, saying China supports the United Nations playing the lead role in global AI governance. "China stands ready to join hands with all countries to build a just and equitable system for global AI governance, and develop AI for good and for all," he said.

What sparked the warnings

Guo's comments follow a fraught week in the US AI industry. Anthropic CEO Dario Amodei published an essay titled "We Must Pace the Frontier" on September 12, calling for a deliberate three-part plan to slow frontier AI development, arguing that a Chinese lead in advanced AI would pose serious risks to the US and the world; the call was echoed over the weekend by OpenAI's Sam Altman and Tesla's Elon Musk.

The debate intensified further after Jacob Coxon, a researcher who had worked on pretraining at both OpenAI and Anthropic, announced his resignation from Anthropic in a viral post on X, accusing both companies of "gambling with our lives" by racing toward self-improving superintelligence without adequate safeguards.

"The people building AI earnestly believe that it could kill us all by the end of the decade," Coxon wrote, according to multiple US media reports.

Evan Hubinger, who leads Anthropic's alignment science team, responded on X that he agreed with Coxon's assessment. "We really do earnestly believe AI could kill all humans," Hubinger wrote, estimating the risk at "greater than 10 per cent" within the next decade, while noting Anthropic does not yet have a plan to solve alignment for future superintelligent systems.

Samuel Marks, who leads scalable oversight at Anthropic, separately warned that AI developers largely continue pushing ahead despite believing in these risks because of "commercial incentives and a belief that they are in a race with other, less responsible AI developers."

OpenAI chief scientist Jakub Pachocki also weighed in with a blog post warning that no AI company has "solved alignment and monitoring to a sufficient degree to continue responsibly scaling at maximum speed for much longer."

Sam Altman, OpenAI's CEO, has separately called for caution over the pace of AI development, saying progress has been rapid and will continue to be so, but "it should be slower." "No amount of American competitive pressure should justify recklessness," Altman was quoted as saying in media reports.

China's dual messaging

China, meanwhile, has sought to present itself as backing both rapid AI development and stronger safeguards. Guo said Beijing places equal emphasis on AI development and security and takes seriously both the direct and secondary risks associated with the technology.

"We are committed to holding on to the bottom line of security, and we have been making continued efforts to improve laws and regulations, policies, application norms, ethics and rules to prevent the abuse and misuse of AI, and ensure that AI is safe, reliable and controllable," he said.

Guo also noted that Chinese authorities released the AI Safety Governance Framework 3.0 on Monday, which reviewed and updated the classification of AI safety risks and set out recommendations on strengthening technological safeguards and governance measures, marking the latest update to a framework Beijing has revised multiple times as the technology has evolved.

This is a notable shift in tone from Monday, when Guo had dismissed the US industry's slowdown calls, including Amodei's essay, as "fear-mongering, confrontation and vicious competition" that would "only hamper efforts toward sound global AI governance," according to a Reuters translation of his remarks.

Setting the stage for Xi-Trump talks

AI global governance is likely to feature prominently when Chinese President Xi Jinping meets US President Donald Trump in Washington later this month, with Xi's visit reportedly expected around September 24.

The meeting comes amid intense global competition over artificial intelligence, making technology regulation and safety a key point of friction between the two superpowers.

Trump, for his part, dismissed the warnings about AI risk on Monday, describing concerns over the technology as a "global warming scam" allegedly being pushed by "radical left democrats."

In a series of social media posts, Trump said the only guardrail AI needs is a "strong and smart president, and the US has that." He also alleged a "sick conspiracy" against AI and data centres, claiming China would benefit if the US moved to restrict the technology, adding, "Whoever wins AI, wins!"

(With inputs from PTI)