Anthropic co-founder tells Tokyo that AI safety must come quickly
Clark told the audience that Anthropic did not make a system to monitor AI.
Jack Clark, a co-founder of Anthropic, gave a lecture at the University of Tokyo on the 13th and held a press conference on the 14th. Clark told the audience that AI can start Recursive Self-Improvement, where AI increases its functions without help from persons. This can make control by persons hard. Clark told them that AI rules must be more strong. AI development must be more slow. Safety measures must come quickly.
Sources
Posted
