Emotions and Reasoning
Eliezer discusses the inherent complexities of human emotions and contrasts them with the capabilities of language models. He suggests that while humans possess innate emotional responses, understanding the inner workings of AI like GPT remains a challenge. He also posits that with dedicated research, we could eventually uncover the reasoning capabilities of large language models, as evidenced by their ability to play chess.In this clip
From this podcast

Lex Fridman Podcast
Eliezer Yudkowsky: Dangers of AI and the End of Human Civilization | Lex Fridman Podcast #368
Related Questions