The Exchanges

Q Well, so you just touched upon it, uh, but maybe to, to wrap it up. So, um, Moody's has had a history with machine learning and AI. So you mentioned some of it, but like, when did it start? Who has been doing this?

A So, I, I would say Traditionally, it was mostly our, our quant groups, right? So we have a lot of quantitative models, and of course, they, they've been leveraging, you know, since I've been here, a lot of different quant models, I wouldn't be able to talk about them. Ah, but, you know, it was always those, the, the enhancements on our predictive modeling were always based on machine learning. Ah, then, you know, the fintech, the whole fintech disruption comes, we start looking at technology and saying, what does this mean for us? What does it mean for our customers? Our banks and insurance companies are going to be disrupted as, as, you know, as a result, and what should we do? And that's when we create the accelerator. And, um, and then we, one of the first priorities was, okay, what we can do with machine learning. So that was, I would say, kind of a re-underwriting our focus on machine learning that happened in 2016. Since then, I would say we're doing several things. You may can talk a little bit more about the applications. One of the ones I'm more particularly excited, because we just launched it on Friday, is, ah, one, ah, leveraging Gen AI, which is called Research Assistant. I can talk a little bit more, but I don't want to steal the thunder from, from Jimmy.

AI assessment note: “re-underwriting our focus on machine learning that happened in 2016”

Q Great. Could you maybe both, ah, double click on some of the things you mentioned? So you chose GPT-IV, but you mentioned you looked at others. So, Uh, why? What was the decision process? How did you evaluate? And then going into RAG, maybe, uh, what tools do you use? What database? Uh, how it works from a technical standpoint?

A Okay, so I'm not the technical person here, so I'm gonna pass it to you, but first I'm gonna maybe preface of how we look. We're always looking at new things, right? I'm, uh, we're, I think we're very passionate about tech and making sure we are at the cutting edge of it, of it, and, and the other thing I'm, I'm kind I'm kind of obsessed, and I think my team is obsessed on making sure that we're not dependent on one technology, but we have the flexibility to switch as time goes by, right? Ah, so we did look at, we're constantly looking at different LLMs and things that are coming. Ah, we did go with, you know, in our testing, and, and also we have an alliance with Microsoft that it's, it's very public. Ah, ChatGPT was the one, OpenAI was the one that was performing better in a, in a, in over Different parameters, right? That said, I do want to emphasize that we, we want to develop, and we keep, we want to keep developing in a very flexible way, so we're always able to take advantage of, of what's the best technology out there.

AI assessment note: “OpenAI was the one that was performing better in a, in a, in over Different parameters”