The News: Salesforce announced Einstein Service Agent, a new autonomous artificial intelligence (AI) agent that is designed to improve the functionality of chatbots by resolving a greater range of tasks autonomously, while seamlessly handing off issues to human agents when required to ensure the best outcomes. According to Salesforce, Einstein Service Agent is currently in pilot and will be generally available later this year.
You can read the press release detailing the news at Salesforce’s website.
Salesforce Debuts Einstein Service Agent
Analyst Take: Salesforce announced the company’s first fully autonomous AI agent, named Einstein Service Agent. The chatbot is designed to resolve cases in natural language quickly, while also being able to detect when a handoff to human agents is required to ensure customer satisfaction and the best possible outcome to the interaction.
Einstein Service Agent is built on the Einstein 1 Platform and is designed to interact with large language models (LLMs) by analyzing the full context of the customer’s message and then autonomously determining the next actions to take. The chatbot then uses generative AI to create conversational responses that are tailored to adhere to a company’s brand voice, tone, and other communications guidelines.
The result is a service-focused bot that is available on a 24/7 basis to communicate with customers in natural language, grounding its responses to service queries in the organization’s vetted and trusted business data. The bot will respond to customers across self-service portals and messaging channels and perform tasks proactively, without customers having to ask, while adhering to operating guardrails that companies define through the Einstein 1 Platform. Salesforce says that Einstein Service Agent can be set up in minutes with user-friendly interfaces, pre-built templates, and low-code actions and workflows.
Einstein Service Agent’s Core Features and Capabilities
Einstein Service Agent is designed to address two core issues. Research has shown that while customers prefer to deal with human agents to handle complex tasks, and do not mind waiting to do so, they also prefer to resolve simpler issues with the help of chatbots. Einstein Service Agent, using the power of generative AI, and ground LLMs with trusted company data, enables organizations to understand and solve a greater number of customer service issues autonomously. This frees up human agents to handle more complex tasks. Moreover, using AI to detect when a customer would benefit from jumping straight to a human improves customer satisfaction and, in many cases, drives retention and loyalty.
Einstein Service Agent features the following capabilities:
- Human-like Reasoning and Responses: Einstein Service Agent’s reasoning engine interacts with LLMs by analyzing the full context of the customer’s input, and then determines the right set of actions to take. The bot then uses generative AI to create responses that are tailored to a company’s brand voice, tone, and guidelines.
- Grounding LLMs to Trusted Company Data: Einstein Service Agent grounds its responses in a company’s trusted business data, including Salesforce CRM data. Furthermore, customers that also use Data Cloud and Unified Knowledge can integrate data and knowledge from third-party systems, such as SharePoint, Confluence, Google Drive, and company websites, and files to help Einstein Service Agent generate accurate responses that are personalized to every customer’s specific needs and preferences.
- Einstein Trust Layer: Einstein Service Agent is secured by the Einstein Trust Layer, which is part of the Einstein 1 Platform, and handles functions like masking personally identifiable information (PII) and defining clear parameters and guardrails for Einstein Service Agent to follow.
- Fast Time to Value: Einstein Service Agent can be set up quickly, using its out-of-the-box templates, Salesforce components, and the LLM of the customer’s choosing. In addition, customers can reuse existing Salesforce objects, such as flows, Apex, and prompts, to provide the Einstein Service Agent bot with additional skills. Custom actions that are specific to their business needs can be created using a low-code builder and natural language instructions to reduce costs and speed implementation.
- Support via Multiple Channels and Modalities: Einstein Service Agent can assist customers anytime across self-service portals and messaging channels, such as WhatsApp, Apple Messages for Business, Facebook Messenger, and SMS. Because Einstein Service Agent understands text, images, video, and audio, customers can send photos when their issue is too difficult to explain in words.
Seamless handoffs to humans: If an inquiry is off topic or falls outside of Einstein Service Agent’s scope, the bot will seamlessly transfer the conversation to a human agent using Service Cloud, and the human agent will receive the full context of the conversation and can pick up where Einstein Service Agent left off without asking the customer to repeat themselves.
Supporting Service from All Angles
Organizations continue to face challenges providing excellent support experiences, largely due to the wide variety of potential support issues, the relatively short window afforded by customers in which to reach a resolution, and the challenge of immediately identifying which support interactions should be handled by a bot versus a human agent.
Salesforce’s Einstein Service Agent has been designed specifically around solving these challenges and is using a combination of different types of AI (analytics, predictive, and generative) to power a bot that is more capable, responsive, and able to interact with humans than the bots that are typically used solely as part of a deflection strategy.
To me, the most interesting capabilities of the Einstein Service Agent are the support for multimodal interactions, and the ability to quickly detect when a seamless handoff to a human should occur.
Often, customers are unable to fully describe in words what their issue is, either due to communication barriers (language, poor voice connection quality, accents, etc.) or because they don’t fully understand the true nature of the problem and may be describing it incorrectly. For some problems, this can be compounded by service agents who may not fully understand the problem and may misclassify the issue.
Enabling customers to record, photograph, and video an issue they’re having not only helps to ensure that the root cause of the problem is identified quickly, but also provides reassurance that the company is doing all it can to make it easy for customers to explain their problems.
Moreover, the data that can be collected via these modalities will be able to provide rich data that can be associated with text descriptions of issues, thereby creating a more comprehensive knowledge base.
Disclosure: The Futurum Group is a research and advisory firm that engages or has engaged in research, analysis, and advisory services with many technology companies, including those mentioned in this article. The author does not hold any equity positions with any company mentioned in this article.
Analysis and opinions expressed herein are specific to the analyst individually and data and other information that might have been provided for validation, not those of The Futurum Group as a whole.
Other Insights from The Futurum Group:
Salesforce $9.13B Q1 FY 2025 Revenue Narrowly Misses Estimates
Salesforce Connections Focuses on Eliminating Data Silos
Salesforce Data Management with Ridecell and Validity DemandTools
Author Information
Keith Kirkpatrick is Research Director, Enterprise Software & Digital Workflows for The Futurum Group. Keith has over 25 years of experience in research, marketing, and consulting-based fields.
He has authored in-depth reports and market forecast studies covering artificial intelligence, biometrics, data analytics, robotics, high performance computing, and quantum computing, with a specific focus on the use of these technologies within large enterprise organizations and SMBs. He has also established strong working relationships with the international technology vendor community and is a frequent speaker at industry conferences and events.
In his career as a financial and technology journalist he has written for national and trade publications, including BusinessWeek, CNBC.com, Investment Dealers’ Digest, The Red Herring, The Communications of the ACM, and Mobile Computing & Communications, among others.
He is a member of the Association of Independent Information Professionals (AIIP).
Keith holds dual Bachelor of Arts degrees in Magazine Journalism and Sociology from Syracuse University.
