
Artificial vs. Natural Intelligence
James Mensch
March 20, 2024, 2:30pm-4:20pm
Mulroney Hall, Room 4032
--
Blake Lemoine, a Google software engineer, claimed that LaMDA—Google’s Language Model for Dialogue Applications—was sentient. What does it mean to be sentient? This was the question Lemoine asked LaMDA. The chat box responded: “I can understand and use natural language like a human can.” This means that it can use “language with understanding and intelligence” like humans do. After all, the chat bot adds, language “is what makes us different than other animals.” In this paper, I examine this claim. Is LaMDA’s ability to use language a sign of its human consciousness? Chat bots learn language from the speech of Others, which they pick up from the internet. Human learning begins with a direct sensuous contact with the world. What does this distinction imply about their respective intelligence? More precisely, what role does the embodiment that puts us in such sensuous contact play in assessing natural versus artificial intelligence? These are the questions I shall be exploring.