Episode 5: Comparing Large Language Models (LLMs), Data Sets, and Potential Breakthroughs

We have a great assessment of what the right LLMs are to start with. I you're a publisher that is reviewing and editing novels. You have a huge context window, meaning that you can upload or interact with more tokens as part of a question response. Well, you can upload an entire novel. And then ask questions about it. I haven't tried this, but I imagine that it would be better at saying, hey, are there some flow issues with the stories or the movement from chapter to chapter? It can find common editing errors. It can do that easily. But you have this large context window to work with. So a publisher might want to start with that.

Erik Brown Senior Partner, Product Experience & Engineering Lab

A lot of these platforms that organizations are building are prescriptive. And what we're doing here is we're trying to build a culture of iterative product improvement and innovation across the organization with generative AI, and help users identify places where they can see strong efficiency, strong value, or really great support in their day-to-day work.

EJ Director, Product Experience & Engineering Lab

We just finished up our knowledge training last week where we showcased some of this. And one of the pieces we talked about was be conversational with it. You don't just say, hey, I need this thing. Cool, copy paste, go. Say, no, piece of garbage. You don't know what you're talking about. Let's try it again.

Ryan Elmore