NYU philosopher David Chalmers explores how modern large language models challenge our traditional views of identity and ethics by generating short-lived, conversational "selves" rather than operating as a single, permanent mind.