Conversational experience

  • Release version: Washingtondc
  • Updated April 25, 2024
  • 2 minutes to read
  • Summarize
    Summarized using AI
    This content was generated using new OpenAI-powered functionality. Results are provided on an as is basis and are not guaranteed to be accurate or complete.

    Summary of Conversational experience Enable AI experiences

    Conversational experience generative AI skills enhance user self-service capabilities within ServiceNow. Utilizing Now Assist, these skills produce search summaries, facilitate catalog item orders via chat, and enable the creation of Virtual Agent topics through large language model (LLM) topic discovery. Setting up these features can be done through the Conversational Interfaces Console or the Now Assist Admin console.

    Show full answer Show less

    Key Features

    • Now Assist Q&A Genius Results: This feature employs the Now LLM to extract pertinent answers from knowledge articles across various platforms, improving user search experiences by providing actionable answer cards linked to the source articles.
    • Now Assist Multi-Turn Catalog Ordering: Users can interactively request items from the Service Catalog, refining their requests in a conversational manner. For instance, they can specify details like color and storage capacity within the same dialogue, with AI generating natural language responses.
    • Now Assist Topics: Virtual Agent administrators can create topics that utilize LLM discovery, bypassing the need for extensive training on traditional NLU models. A simple, clear topic description allows the LLM to identify the best match for user queries.

    Key Outcomes

    By implementing these features, ServiceNow customers can expect improved self-service interactions, faster resolution times, and a more intuitive user experience. The generative AI capabilities streamline processes and enhance the overall effectiveness of the Virtual Agent, making it easier for users to find information and services they need.

    Conversational experience generative AI skills can improve the user self-service experience. Platform Now Assist skills generate search summaries, order catalog items from the chat window, and create Virtual Agent topics that use large language model (LLM) topic discovery.

    The following skills require Now Assist in AI Search and Now Assist in Virtual Agent. Both are available in some Now Assist products.

    You can set up Now Assist in Virtual Agent from the Conversational Interfaces Console or from the Now Assist Admin console. For more information, see Configuring Now Assist in Virtual Agent.

    Now Assist Q&A Genius Results

    Now Assist in AI Search uses the Now LLM to extract actionable Q&A Genius Result answers from knowledge article results found in Service Portal, Virtual Agent, Employee Center, and global searches. When a user submits a question in search, AI Search retrieves the top knowledge article result and passes it to the Now LLM for answer generation. Answers augment the user's search results, displaying as actionable Now Assist Q&A Genius Result answer cards. For reference, each answer card includes a link to its source knowledge article.

    Now Assist Q&A Genius Results in the Virtual Agent chat window.

    For more information, see Now Assist in AI Search.

    Now Assist Multi-Turn Catalog Ordering

    The Now Assist Multi-Turn Catalog Ordering skill gives users access to available options in the Service Catalog. Users can request an item, such as a mobile phone. The user can then provide more information to refine the search. For example, they may refine their request to a blue 256-GB iPhone. They can even request a new item instead, all in the same conversation, and the generative AI creates its responses using natural language.

    Figure 1. New email catalog request from the chat window
    User asks for a new email. Virtual Agent matches this request with a catalog item and presents it to the user.

    Now Assist Topics

    The Now Assist Topics skill lets Virtual Agent admins create topics that use large language model (LLM) discovery. Unlike NLU topics, LLMs don't require models, intents, or keywords to be linked to the topic. LLMs can discover topics and perform language-related tasks, such as text generation for case summaries and resolution notes, without months of training on NLU models. The LLM does all of the heavy lifting for you. The only requirement is a robust, plain language topic description. The LLM uses this description to find the best topic match for the user utterance. If there are multiple potential matches, the user will see a list of topics to choose from.