Executives of major artificial intelligence companies and prominent researchers warned the United Nations Security Council about the safety risks of increasingly capable AI systems and urged international cooperation to oversee them.
Sam Altman, chief executive officer of OpenAI, said AI development has reached a turning point and requires "extraordinary caution." He argued the technology could both spark a new renaissance of creativity and discovery and produce severe social disruption. Altman highlighted two main dangers: first, that system capabilities could advance so rapidly people may be unable to monitor or intervene; and second, that excessive power could concentrate in the hands of a few companies or countries.
Yoshua Bengio, a Canadian researcher often referred to as one of the "godfathers" of artificial intelligence, told the Council that the risks are real and immediate, and no single country can handle them alone.
Dario Amodei, chief executive officer of Anthropic, described AI as among the world's most important security issues and warned it could pose risks to all humanity if mismanaged. Amodei called for global rules for testing new AI models and for prohibiting the use of the technology in developing biological weapons.
During the discussion, representatives of major powers expressed differing views on global regulation. Michael Kratsios, the White House science and technology advisor, rejected the idea of global AI regulation, arguing that it is difficult to regulate effectively a technology that is not yet fully understood. Fu Cong, the Permanent Representative of the People's Republic of China to the UN, urged stronger regulatory frameworks, crisis management and international cooperation.
The intervention of industry leaders and researchers at the Security Council underscores that policy discussions are keeping pace with rapid technological change: both scientists and some industry figures are calling for international solutions that go beyond the authority of any single state or company.
Related event
Further debate on the topic is planned at the Portfolio AI & Digital Transformation conference on November 26, where the issue of AI safety and governance will also be addressed. Registration and details have been announced.
Note: the cover image is illustrative.
