Einstein Summary Setup in Automotive Cloud

Set Up Data Cloud

Data Cloud helps you ingest, harmonize, and analyze your data at scale. Some Einstein generative AI features, such as Einstein Trust Layer, work only when Data Cloud is provisioned and enabled in your org. Set up Data Cloud to use the Einstein Summary component.

Some prompt templates, such as Summarize Vehicle Telematics and Services Data, use data that's stored in some data model objects (DMOs) in Data Cloud. Make sure that you ingest data from external sources in to the DMOs, such as Vehicle Telematics Event, Vehicle Fault Code Mapping, and Vehicle Performance Summary.

See Set Up Data Cloud for Automotive Cloud.

Set Up Einstein Generative AI

Einstein Generative AI is a suite of features that includes Prompt Builder and the Einstein Summary component. To use generative AI features, you must first turn on Einstein in Setup.

Einstein Trust Layer is a secure data architecture that's built into the Salesforce platform to help you safely explore and use generative AI features. It includes a set of data security and privacy controls that helps Einstein to deliver AI that’s securely grounded in your data, without introducing potential security risks. Based on your privacy and security requirements, configure the Einstein Trust Layer settings in Setup.

To collect Einstein generative AI audit and feedback data, configure Einstein Feedback settings in Setup.

See Set Up Einstein Generative AI.

Enable Automotive Generative AI

To configure and use the prompt templates for automotive, turn on Automotive Generative AI in Setup.

Set Up Context Service

Some predefined prompt templates use context definitions to dynamically and safely ground prompts with CRM data. Context definitions consist of the canonical structure of data that's required for summarization. Context definitions retrieve data from Salesforce objects and Data Cloud data model objects and ground prompts. Grounding provides large language models (LLMs) the context to create accurate, up-to-date summaries.

Set up Context Service to use context definitions for AI-based summaries. See Context Service.

Assign Permission Sets

Grant users the permissions to set up and use Prompt Builder and Context Service features.

Grant users the right level of access to the records that they want to summarize and their related records. See Prompt Templates for Automotive.

Assign permission sets to use the required features in Data Cloud. See Data Cloud Standard Permission Sets.

Explore Prompt Templates for Automotive

Prompt Builder helps you build, test, and manage prompt templates to get consistent, information-rich, and high-quality AI-generated outputs. Prompt templates are reusable, structured prompt formats that include data placeholders, specific instructions, and contextual information to shape an LLM's response.

The Einstein Summary component works with prompt templates to generate predictable, relevant record summaries based on a large number of data sources, such as Salesforce objects and data model objects.

Automotive Cloud comes with predefined prompt templates that you can use with the Einstein Summary component on key record pages such as Vehicle and Asset.

See Prompt Templates for Automotive.

Add the Einstein Summary Component to a Page

Add the Einstein Summary component to a record page by using Lightning App Builder. In the component properties pane, enter a component label and a button label. Select the prompt template types and the prompt templates that can be used with the component. For example, to help service managers easily analyze vehicle services, you can add the component to a vehicle record page, select Record Summary as the prompt template type, and select Summarize Vehicle Services as the prompt template.