What do Jev and other decision models mean for the data ecosystem?

🌎 Global-friendly sessions November 12 & 13

In September 2026, TypeSafe AI released their "System One" AI model, Jev. Unlike traditional large language models (LLMs) designed to generate text, Jev’s purpose is to return fast answers to questions along with a calibrated probability. Since this release, a lot of applications for this type of models have emerged and multiple AI Labs have now published their own “decision models”, following similar principles.

In this webinar, we will dive into the difference between traditional LLMs and decision models and highlight use cases for each of those. We will focus on cases where decision models might be particularly helpful for data analytics. We will be sharing and demoing a new dbt package, dbt-decisions, developed to simplify using those new models in your dbt projects, and will present a short proof of concept of how a Semantic Layer and Jev can be used together.

You will leave the session with an understanding of this new type of AI models, examples where you will be able to leverage them in your day to day and a package that you will be able to install to get started immediately.

Meet the speakers

Benoit Perigaud

Staff Developer Experience Advocate

dbt Labs

Doug Beatty

Developer Experience Advocate Lead

dbt Labs