Salesforce has announced the public beta availability of Einstein Copilot, a new customizable, conversational, and generative AI assistant for CRM at Mobile World Congress in Barcelona.
Unlike other AI assistants or copilots that lack adequate company data to generate useful responses, Einstein Copilot enables Salesforce customers to generate responses using their own private and trusted data, while maintaining strict data governance and without requiring expensive AI model training. As a result, Einstein Copilot can answer questions, summarise content, create new content, interpret complex conversations, and dynamically automate tasks on behalf of a user, all from a single, consistent user experience embedded directly within Salesforce’s #1 AI CRM applications.
This is made possible by combining a conversational UI, a foundational large language model, and trusted company data that enables Salesforce users to tap into the power of generative AI and interact with their applications in entirely new ways.
Grounding prompts in Data Cloud: Einstein Copilot grounds its responses with trusted business data from Data Cloud to provide the necessary context for the highest quality outputs. This allows Einstein Copilot to generate more precise and tailored responses based on trusted company data.
Out-of-the-box “actions”: Einstein Copilot comes with a library of out-of-the-box actions – pre-programmed capabilities, automated responses, or business tasks performed by Einstein Copilot – that the AI can perform for the user when prompted. Actions can be combined to execute dynamic multi-step plans. For example, a customer service agent can ask Einstein Copilot to close a case and open a sales opportunity or sell an add-on, and Einstein Copilot will understand the user’s intent and can execute the task — or tasks — in the flow of the service experience. With Einstein Copilot there are no silos between applications or data.
Customise Einstein Copilot for specific business needs: Einstein Copilot can be customized to accomplish specific sales, service, marketing, commerce, and IT tasks, ensuring company and industry policies are applied. Copilot Builder can create custom actions for Einstein Copilot, Prompt Builder activates custom prompts in the flow of work, and Model Builder uses proprietary AI models to power custom Einstein Copilot functionality.
Reasoning engine to interpret intent and choose the best action: Einstein Copilot’s reasoning engine – the process that interprets and processes information to make informed decisions, solve problems, or generate insights – interacts with an LLM by analyzing the full context of the user’s prompt, determining the actions or series of actions to take, and generating the output. For example, if a sales rep asks Einstein Copilot, “Help me recommend a new product tier to my customer,” Einstein Copilot can determine which products the customer currently has, understand the upgrade options, and transition the customer into a new high-value marketing segment. Finally, Einstein Copilot can update the information across multiple systems with MuleSoft and Salesforce Flow.
Generating trusted responses and actions: Einstein Copilot serves up trusted AI interactions with privacy and security measures provided by the Einstein Trust Layer. The Einstein Trust Layer is part of the Einstein 1 Platform and performs functions like masking personally identifiable information (PII), scoring outputs for toxicity, and helping to protect information from unauthorized access and data breaches through zero-data retention from Salesforce’s LLM partners.