I’ve been working with Joe Truss on a project called ‘Book of Codes’, where we are examining the building blocks of the universe. The main premise is that we live in a Tetraverse, a universe where the smallest possible length in the universe (the Planck length) must be the edge length of a tetrahedron.
We put our ‘We Live in a Tetraverse‘ video into Google’s NotebookLM and had it create this Deep Dive Conversation.
I think this is quite insightful. I’ve been thinking about how to use this tool since I shared it by putting my blog posts into it and having it do a Deep Dive on the content. Since then, I’ve read a few things that have questioned just how useful this kind of podcast really is?
Alan Levine says in ‘Wow Us with your AI Generated Podcast…, “In one sample listen, you might be wowed. But over a series, Biff and Buffy sound like a bunch of gushing sycophants, those office butt kissers you want to kick in the pants.”
And in a comment response Aaron Davis mentions my blog’s Deep Dive, “I agree with you Alan about the initial amazement about what is possible, I am not sure how purposeful it is. I listened to David Truss’ podcast he posted and was left thinking about my experience with David Truss’ writing. I imagine that such tools may provide a possible entry way into new content, but I am not sure what is really gained by putting this into an audio format?”
It’s true, I’ve done this a few times and while it can be impressive, I do see that this can get a bit old pretty fast. Except for one thing… I think that if you are asking it for a general summary of light content, you are going to get a light and fluffy Deep Dive response. However, if you want to understand something really challenging or different or dense, this could be a really good way to get a general understanding of tough to understand content. The Deep Dive into the Tetraverse video actually did a really good job of describing new content in a clear way. I found the kaleidoscope metaphor it mentioned an insightful analogy and I think that listening to the audio first would help someone appreciate the video even more.
Like any new and shiny tool, this Deep Dive podcast on Google’s Notebook LM will get a lot of play and then dwindle in use… but that doesn’t make it useless. I think it will find it’s rightful place as a way to take dense material and make it digestible. It will be a great content introduction, an insightful entry into new learning. It won’t become something you go to listen to where you also listen to your favourite podcast episodes. Still, it will have a purpose and you might find yourself going to it, or to a similar tool, when you have too much content to summarize, or if the content is significantly challenging to parse.
Hey David, glad to “hear” this from your experience, it’s helpful to know from someone close to the complex source that the podcast was effective. Yes, not listening, the pair does condense it in a friendly format.
We will just see what happens. If people listen to the summary, will they go to the source for more? We’d like to think so. Or maybe they just leave with the summary. I guess it’s not terrible. Complex stuff you are into!
Can’t help being a nerd, it’s in my DNA 😆.
I’ve learned that you can give this AI duo instructions regarding what you what them to focus on and again found some insights that I might not have recognized myself.
I realize that this might be something Chat GPT or other AI tools can do equally or more effectively via text, but as an avid audio book and podcast listener, there is something about learning from conversations that appeals to me more than text.
Is this something I see myself getting bored with? Maybe, but as these LLM’s get better, so will the conversation. Imagine if this AI conversation became one that you could participate in… wouldn’t that be something really facilitating to participate in?