We argue that the notion of "mammalian value systems" points to a potential avenue for fundamental research in AI safety and AI ethics.
確定! 回上一頁