United Nations Secretary-General Antonio Guterres said on Wednesday that the world "cannot afford a race to the bottom" on artificial intelligence safety, calling for stronger international guardrails as AI development outpaces efforts to understand its risks.
Speaking at a news conference ahead of the 81st session of the U.N. General Assembly, held in Hamilton, Canada, Guterres said AI carries "enormous potential" to accelerate sustainable development, improve learning, strengthen health systems and boost climate resilience.
He said a growing number of people involved in building AI systems are warning that development is moving faster than understanding of the risks involved.
Guterres acknowledged geopolitical divisions among nations developing AI technology but said safety standards should not be compromised.
"We need guardrails to build trust and that make AI safe, transparent, accountable, with human dignity at the center," he said.
He added that AI will be a major topic in his discussions with world leaders in the coming week, and said the United Nations will continue to explore practical approaches in the days ahead.
Asked about the possibility of an AI summit during the General Assembly's high-level week, Guterres said such an event would make sense "at the right moment" but that next week would not allow sufficient preparation, given competition with other initiatives already scheduled.
He said AI requires dedicated attention as a subject "that needs to be seen by itself" rather than folded into other discussions.