Incorporating LLMs for analysts with SQL AI Functions

There are many use cases where you can integrate an LLM, such as DBRX or OpenAI, for insights. With the Databricks Data Intelligence Platform, it’s also possible for analysts who are most comfortable in SQL to take advantage of advances in machine learning and artificial intelligence.

Within Databricks, you can use AI Functions, which are built-in SQL functions to access LLMs directly. AI Functions are available for use in the DBSQL interface, SQL warehouse JDBC connection, or via the Spark SQL API. In Figure 8.16, we are leveraging the Databricks SQL editor.

Foundational Models API

The storage and processing of data for Databricks-hosted foundation models occur entirely within the Databricks Platform. Importantly, this data is not shared with any third-party model providers. This is not necessarily true when using the External Models API, which connects you to services such as OpenAI that have their own data privacy...