- Post History
- Subscribe to RSS Feed
- Mark as New
- Mark as Read
- Bookmark
- Subscribe
- Printer Friendly Page
- Report Inappropriate Content
on 07-31-2025 08:17 PM - edited a month ago
Definition
AI Search in Now Assist for CSM enhances the way customers and agents find relevant content across the platform. It provides semantic, intent-aware results that go beyond keyword matching, enabling smarter self-service and faster case resolution.
For customer-facing use cases, AI Search powers the search bar in portals, Engagement Messenger, and Virtual Agent, helping users find the most relevant articles, catalog items, or community posts. It also underpins features like Suggested Answers and Q&A response generation. On the agent side, it supports contextual search in the workspace, enabling quick lookup of similar cases or knowledge in Now Assist Panel, even during live interactions.
When used in tandem with Agentic AI, AI Search helps determine the right next action or workflow to trigger based on what the user is asking or searching for. It plays a foundational role in retrieval-augmented generation (RAG) patterns within Now Assist capabilities and connects to do semantic indexing with Knowledge Graph inside and outside (via Workflow Data Fabric) of ServiceNow.
Key Capabilities
AI Search Admin Console: no-code management tool to configure, test, and enable AI Search solutions across different types of search applications.
Multi-content Q&A Genius Results: use an LLM to generate search and chat responses that synthesize information from knowledge articles, Service Catalog items, and other available content types.
External Content Connectors: enables AI Search applications to search content and metadata from supported external source systems, such as Atlassian Confluence Cloud and Microsoft SharePoint Online
Assist Actions for External Users: enable search for knowledge articles for external users
Implementation
2. Configure Now Assist in AI Search
3. Reference for Now Assist in AI Search
4. Get Started with External Content Connectors
5. Configuring External Content Connectors
6. Using External Content Connectors
7. Get Started with Knowledge Graph
8. Configuring Knowledge Graph
9. Using Knowledge Graph Designer
Key Best Practices
- Now Assist in AI Search FAQ
- Now Assist in AI Search - Introduction
- Knowledge management best practices for AI
- Make AI Search Work: Practical Lessons from the Field
Formal Learning
Now Assist in Search Implementation
Frequently Asked Questions (FAQs)
1. What is AI Search and how is it used in CSM?
AI Search replaces keyword-based lookups with semantic, vector-based matching. In CSM, it powers portals, Virtual Agent, case deflection, and agent workspace search. It allows customers and agents to find the most relevant KBs, cases, or catalog items based on intent rather than exact text.
2. How does AI Search improve self-service?
It returns smarter results, even when customers don’t phrase their issue exactly. For example, a search for "I can't log in" can surface a knowledge article titled "Troubleshooting Authentication Issues." It also integrates with Virtual Agent to answer questions or guide users to actions.
3. How does AI Search tie into Agentic AI and orchestration?
AI Search helps identify patterns across user input, such as recurring queries. It can also provide signal context to LLMs or agentic workflows. For example, it can detect that a user searching for refund policies has unresolved cases and offer a proactive case triage or follow-up action.
4. What sources can AI Search pull from in CSM?
It can index Knowledge Bases, Service Catalogs, Communities, Portals, and even custom tables. You can configure what each persona (e.g., customer, agent) sees using source filters and user criteria.
5. Can I control which results are returned?
Yes. You can tune search sources, apply relevance boosts, exclude outdated or irrelevant content, and apply audience-specific rules. You can also pin content or create dynamic filters to promote urgent updates.
6. What if AI Search is returning irrelevant results?
Use the AI Search Analytics dashboard to identify low-performing queries. Adjust metadata, refine search sources, or rewrite confusing articles. For agents, consider enabling result filtering by case context or role.
7. How do I configure AI Search for my CSM portal?
Ensure the AI Search plugin is enabled, and your search application is mapped to the correct portal. Review Knowledge Base visibility, enable result types like KB or catalog, and test result quality before launch.
8. Is AI Search required to use Now Assist features?
Yes. AI Search is a foundational capability that supports generative answers, semantic lookup, and contextual results across Now Assist Virtual Agent, Suggested Answers, and recommended actions in workspace.