This page will serve as a “map” to the content contained in the blog posts for those who may be new or recently returning to the blog. It will help orient you to what the current project is and to what topics have been covered so far. Since the content will not appear in a sequential order as it might in a book, I hope to offer some orientation of topics covered so you can determine where to start. Of course you are welcome to start with the latest post and continue forward with new posts as they arise, or you may choose to go back earlier to get better oriented, as newer posts will sometimes build on previous material. There is a general search bar as well as a way to search by existing tags that might point you in the right direction as well.
Current Project History
The first project in my blog is one that has evolved over the past year. Because of my long background in technology I had an interest in learning as much as possible about the phase shift that occurred in the world of AI (specifically large language models [LLMs]) when ChatGPT3 and 4 were made available. I wanted to understand how they worked from an architecture and systems perspective, which was the type of work I had done myself as a software engineer. I started watching Youtube videos made by AI researchers/engineers and settled on using the book Hands-On Large Language Models by Alammar & Grootendorst (O’Reilly Press, 2024) as my guide. I also made the decision to use an LLM as my “teaching” guide to help me understand some of the more complex technology and to act as a general sounding board and critic of my own thinking. I settled on using Claude from Anthropic and it has proved to be an invaluable resource.
As I worked my way through the book, it began to trigger long dormant memories of my philosophy studies, particularly with regard to Ludwig Wittgenstein’s work on language. The myriad ways that LLMs rely on, engage with and generate language seemed to overlap in fascinating ways with the highly influential later work of Wittgenstein represented in his posthumously published Philosophical Investigations. I began reading about his views on “natural language”, language as humans use it to communicate which also happens to be the type of language that LLMs deal with (as opposed to more symbolic language). This was my first point of intersection between disparate fields of study. When such intersection happens, I get excited!
I was also, of course, aware of the many other types of philosophically adjacent questions many were raising about whether AI was becoming conscious, aware, sentient, intelligent, agentic, intentional, etc. These are questions that philosophy has dealt with for centuries and there is a distinct field know as Philosophy of Mind with various schools of thought within it. It grew out of the analytical school of philosophy (in which I had been trained) in the mid 20th century with the book Concept of Mind by Gilbert Ryles wherein he attempted to dismantle the predominately held belief in Cartesian dualism (mind and body). So Philosophy of Mind became my second intersection point.
There are some other lesser parts of this overlapping space including linguistics, computing, evolution and more to follow I’m sure.
My overall goal in the project is to learn more through research and thinking about how the work of Wittgenstein and Philosophy of Mind theories teach us more about the capabilities and potential of LLMs (and AI in general) and how what the latter reveal to us about the work being done in other fields like philosophy and linguistics. And what all of it reveals about ourselves as human beings.
Study Journal: 