Definition
A Large Language Model (LLM) is a type of deep learning model trained on massive amounts of text data that can understand and generate human language at scale. In the Salesforce ecosystem, LLMs are the underlying AI technology powering features like Agentforce agents, Einstein Copilot, Einstein GPT, and prompt-based interactions across Sales, Service, Marketing, and Commerce clouds. Salesforce provides access to LLMs through the Einstein Trust Layer, which supports both Salesforce-hosted models and bring-your-own-model integrations from providers like OpenAI, Azure OpenAI, Google Vertex AI, and Amazon Bedrock. The Trust Layer adds data masking, toxicity detection, and audit logging around LLM interactions.
Real-World Example
an AI specialist at Nexus Innovations recently implemented Large Language Model to bring intelligent automation to a process that previously required significant manual effort. Large Language Model analyzes patterns in the data and surfaces insights that would take a human analyst hours to uncover, enabling the team to act proactively rather than reactively.
Why Large Language Model Matters
A Large Language Model (LLM) is a type of deep learning model trained on massive amounts of text data that can understand and generate human language at scale. In the Salesforce ecosystem, LLMs are the underlying AI technology powering features like Agentforce agents, Einstein Copilot, Einstein GPT, and prompt-based interactions across Sales, Service, Marketing, and Commerce clouds. They're the engine behind generative AI features that Salesforce offers.
Salesforce provides access to LLMs through the Einstein Trust Layer, which supports both Salesforce-hosted models and bring-your-own-model integrations from providers like OpenAI, Azure OpenAI, Google Vertex AI, and Amazon Bedrock. The Trust Layer adds critical safeguards around LLM interactions: data masking (sensitive data is hidden before reaching the model), grounding (responses are anchored in real CRM data), toxicity detection (harmful outputs are filtered), and audit logging (all interactions are tracked). These safeguards make LLM use viable in enterprise CRM contexts where unprotected LLM access would be a regulatory and security nightmare.
How Organizations Use Large Language Model
- •Vertex Global — Uses Einstein Copilot powered by LLMs through the Einstein Trust Layer, with sensitive data automatically masked before reaching the underlying model.
- •NovaScale — Integrates a third-party LLM via the Trust Layer's bring-your-own-model support, gaining access to the model's capabilities with Salesforce's safety controls.
- •Coastal Health — Trusts the Einstein Trust Layer to handle LLM interactions safely for clinical use cases where data leakage would be unacceptable.