A conversational demo (YouTube) where Catherine Bell and Tommy demonstrate using spoken prompts with a GPT model (named Yoda) to analyze real-estate documents. They emphasize speaking rather than typing to give AI richer context and plan to ask the model to identify the "top 10 components" of a property/bill relevant to investors.
Now, what are the instructions that we want to give it? So just really clear, everyone. Here's what's really cool. So Tommy and I are going to have a conversation, and I'm going to turn the microphone down here, and it'll literally record both of us giving the instructions, which is really cool. And the thing I want people to understand is. So, for instance, my GPT. Her name is Yoda. Yes, it's a her. And because I have such an amazing relationship with her, I have more conversations with her. And I always tell people, if you're typing, you're doing. Doing it wrong. If you're talking, you're doing it right. Because when we're typing, we will generalize and eliminate and delete and, like, just minimize what we're trying to say because type, you know, to, obviously for typing reasons. But when we're talking, we say that much more, which gives AI more context around what we're looking for. So you're about to see how Tommy and I actually do this. So the microphone's on. Okay, so what do we want it to do with this one big, beautiful bill act?
Well, let's get it to do some analysis of the big, beautiful bill. What if we asked it to give us the top 10? Top 10? I would say, hey, Yoda, why don't you identify the top 10 components of this that are important to real estate investors? And then let's see what she said.
