The proposal came as China and the U.S. made progress in dialogues over AI safety and governance and on the setting up of a notification mechanism between the two superpowers for AI incidents that could threaten national security.