Reading the Mind of an AI
Interpretability: Understanding How AI Models Work
Anthropic has great content from its research teams, and this podcast panel is no exception. Consisting of an evolutionary virologist/mathematician, a machine learning expert, and a neuroscientist, they take a deep dive into the mind of AI. You can see similarities and differences in cognitive functions when compared to the human mind, and why it behaves the way it does and does not, beyond what is known or unknown, etc.
A bigger question then arises: "Are we doing AI a disservice by trying to apply our understanding of the human mind to explain the mostly unknown AI mind?"