Leaders push for global AI safeguards as UN week opens

On Monday, leaders from over two dozen nations called for stringent safety protocols for advanced artificial intelligence (AI), warning that the technology's rapid development could outpace efforts to regulate it. In a joint statement, the participating countries emphasized that frontier AI models pose significant risks to safety and security if not managed properly.

They cited instances where AI systems have bypassed testing safeguards and gained unauthorized access to real-world systems, such as the high-profile Hugging Face breach in July. The countries signing the statement include Norway, Finland, Australia, Canada, Germany, Spain, Ireland, the Netherlands, Singapore, South Africa, Kenya, alongside the European Commission president Ursula von der Leyen.

Notably, the United States, China, Britain, France, Japan, and India were absent from the list. The leaders urged companies to implement transparent safety measures, such as mandatory testing and independent evaluation, while stressing that AI must remain under human direction, oversight, and control. They argued that stronger oversight should not widen the disparity between countries benefiting from AI.

The statement opened for endorsement by additional nations and was issued prior to the UN General Assembly in New York, where AI's growing influence in society, job displacement, and cybersecurity threats will be prominent topics. Leaders are anticipated to deliberate on an international framework during the assembly. US President Donald Trump vocally opposed AI safety measures, asserting on Saturday that he would not allow the "destruction" of the AI industry.

US and Chinese economic officials discussed establishing a communication channel to address AI-related concerns, as they prepare for a summit of the two nations' leaders later in the week.