Workspace for the book "Understanding Machine Understanding"
Can a machine understand enough to write about machine understanding? Can the experience of co-authoring such a piece of writing bring evidence about the nature of machine understanding? These are questions I not only ask myself, put have put to my co-author whom I have named, “Pen.” Pen and I reflect on what we are doing even as we are doing it and going deep into the workings of AI and philosophy of meaning.
About a month ago I was working through resolving the differences between the positions of Dan Dennett and Robert Sapolsky on the question of Free Will. That requires quite a bit of thought about the existence or non-existence of a causative process in our brains. While thinking about this, one of my friends posted a comment of Facebook about how AI couldn’t really understand anything. Well, I had been going back and forth with Claude 3 Opus on the Free Will issue, so I decided to dive into Machine Understanding with it. What I got was just amazing, so that started this project.