Although this post was written by AI, it was reviewed by human.
Everyone is talking about AI in ServiceNow, especially after the huge news of ServiceNow acquiring Moveworks late last year. But what does it actually mean for us developers and architects? How does it all connect under the hood?
In this post, we’ll break down the key concepts, explore the architectural overview, and touch upon the critical aspects of data storage and security. Let’s get to it! 👇
Key Concepts & Usage
Traditionally, ServiceNow requires users to navigate to a Service Portal (like Employee Center) to submit requests or read knowledge bases. Moveworks takes a different approach: it acts as an “AI Front Door”. Instead of replacing your ITSM infrastructure, it layers over it.
Employees ask questions in Microsoft Teams, Slack, or directly via the Moveworks for Employee Center widget. The AI intercepts routine requests, searches your Knowledge Bases (KBs), and even allows users to fill out ServiceNow Catalog Items right inside the chat window! If the AI can’t solve it autonomously, it seamlessly routes the issue to a live agent, generating a detailed ticket.
As Ravenna’s recent market comparison notes, Moveworks focuses heavily on resolving routine requests before they enter your queues, saving precious time for fulfillers and preventing ticket bloat.
Architectural Overview
How do they talk to each other? The architecture relies on robust API integrations and dynamic syncing. Moveworks routinely ingests data from your ServiceNow instance to build its conversational index.
- Data Ingestion: Moveworks pulls catalog items, forms, and knowledge articles. It automatically converts ServiceNow variables into its internal supported types.
- Form Integration: When a user interacts with a form via chat, Moveworks respects your ServiceNow configurations. However, keep in mind that for UI Policies, you might need specific setups—for example, Moveworks ignores
active=falsefields unless explicitly handled by UI policies in some contexts (check out the Moveworks Forms Integration docs for specifics). - Authentication & Artifacts: The integration leverages Outbound REST Messages, Application Registries, OAuth Entity Profiles, and dedicated Service Accounts to securely exchange payloads.
- Agentic Action: Using the Moveworks Reasoning Engine and Agentic Studio, the platform understands complex requests, pulls the right ServiceNow Record Producer, and submits it using the captured
requested_forvariables.
Data Storage & Security
Security is always the elephant in the room when dealing with AI. Fortunately, Moveworks was built with enterprise-grade security, which perfectly aligns with ServiceNow’s stringent platform standards (including GovCloud environments and FedRAMP authorization!).
- Permissions Mirroring: Moveworks strictly adheres to ServiceNow’s User Criteria! If a catalog item has specific “Available For” or “Not Available For” rules, Moveworks ensures the form won’t surface for restricted users.
- PII & Data Masking: The platform supports real-time data masking and context-aware redactions to ensure sensitive data (like HR records or passwords) isn’t improperly stored or generated by the AI.
- Encryption: All data in transit and at rest is heavily encrypted, with logical tenant separation keeping your instance’s data completely isolated. You can read more about their security-by-design approach on the Moveworks Security Blog.
Wrapping up
The combination of Moveworks’ intuitive conversational AI and ServiceNow’s powerful workflow engine is a match made in heaven. It shifts the paradigm from just “logging tickets” to actually resolving them autonomously.
I left the best for the end – see how the request is processed on this simple diagram:
