Both Chinese President Xi and US President Trump dismiss the call from the top executives of American AI companies to slow down the development of the sector as scaremongering.
The leaders behind major Western AI companies stated on Saturday that the sector is growing too fast to guarantee safety. For example, AI agents could independently carry out cyberattacks, wrote Anthropic CEO Dario Amodei, the developer of Claude, in an essay.
The leaders of ChatGPT, Grok, and Gemini supported the call.
Amodei also stated that a "Chinese lead in AI would pose a great danger to the United States and the world" and advocated for restrictions on the sale of advanced AI chips and equipment from the US to China.
President Trump responded today that there are already "enormous powers in criminal law and regulation." "There is a sick conspiracy going on against AI and data centers, and the only one happy about it is China. Whoever wins the AI race wins everything!" he wrote on his platform Truth Social.
A spokesperson for Chinese President Xi told reporters that "stories about threat, confrontation, and malicious competition only serve to disrupt the process of global AI regulation and are in no one's interest."
China does warn for balance
Chinese authorities do state that there must be "a balance between development and safety." But instead of delay, Chen Yixin, the Chinese Minister of State Security, advocated for national legislation and ethical guidelines to "safeguard the interests of the population and national security."
The minister also called for international cooperation.
The US Secretary of the Treasury and Chinese Vice Premier He Lifeng are trying to meet ahead of the summit between the two countries next week. He leads the Chinese agency that oversees the country's economic planning, which also includes AI development.
Trump and Xi are expected to discuss AI among other topics during the planned meeting on September 24.
Security threat or competition killing?
The discussion about AI safety has flared recently after several former employees of OpenAI (ChatGPT) and Anthropic (Claude) warned of possible threats. An Anthropic employee even stated that AI could exterminate humanity.
Not everyone believes in the genuine concern of the AI leaders. Some think that Anthropic and OpenAI are sowing fear so that more regulation comes from the US government. In practice, this could mean that smaller companies can no longer compete with the AI giants because they cannot meet the requirements.
For example, Jensen Huang, CEO of AI chip manufacturer Nvidia, said that the companies exaggerated safety concerns because they planned to introduce new cybersecurity products.
"What better way to create demand than by creating a problem?" Huang said during a technology conference in San Francisco last week. "Who wouldn't want the market to be in a state of hysteria over their product and people lining up around the corner for it?"