Can we have real conversations with AI? How do illusions help us make sense of the world? - Highlights - KEITH FRANKISH

“Generative AI, particularly Large Language Models, they seem to be engaging in conversation with us. We ask questions, and they reply. It seems like they're talking to us. I don't think they are. I think they're playing a game very much like a game of chess. You make a move and your chess computer makes an appropriate response to that move. It doesn't have any other interest in the game whatsoever. That's what I think Large Language Models are doing. They're just making communicative moves in this game of language that they've learned through training on vast quantities of human-produced text.”

Keith Frankish is an Honorary Professor of Philosophy at the University of Sheffield, a Visiting Research Fellow with The Open University, and an Adjunct Professor with the Brain and Mind Programme in Neurosciences at the University of Crete. Frankish mainly works in the philosophy of mind and has published widely about topics such as human consciousness and cognition. Profoundly inspired by Daniel Dennett, Frankish is best known for defending an “illusionist” view of consciousness. He is also editor of Illusionism as a Theory of Consciousness and co-edits, in addition to others, The Cambridge Handbook of Cognitive Science.

www.keithfrankish.com
www.cambridge.org/core/books/cambridge-handbook-of-cognitive-science/F9996E61AF5E8C0B096EBFED57596B42
www.imprint.co.uk/product/illusionism

www.creativeprocess.info
www.oneplanetpodcast.org
IG www.instagram.com/creativeprocesspodcast