Do AI Models Know What They're Thinking About?

In this episode

After over a century of studying introspection in humans and animals, researchers are beginning to explore the phenomenon in unlikely new subjects: large language models. These systems, which are based on mathematical structures called neural networks, excel at cognitive tasks that until recently seemed exclusive to humans. Can they introspect as well? On this episode of The Quanta Podcast, host Samir Patel speaks with staff writer Ben Brubaker about “injecting” thoughts into LLM neural networks — and what researchers are doing to map out AI’s capacity for introspection. This topic was covered in a recent story for Quanta Magazine. Each week on The Quanta Podcast, Quanta Magazine editor-in-c…