Now Assist in Document Intelligence release notes

  • Release version: Zurich
  • Updated July 31, 2025
  • 9 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 Now Assist in Document Intelligence Release Notes - Zurich Release

    Now Assist in Document Intelligence leverages generative AI to extract information from documents, summarize content, and answer predefined questions. The Zurich release introduces significant enhancements, enabling ServiceNow customers to improve document processing efficiency, accuracy, and usability across various workflows.

    Show full answer Show less

    Key Features

    • Document and Visual Insights AI Agent Enhancements: Supports recognizing citations in multiple attachments, uploading files without predefined use cases, and displaying results in a dedicated document view. Users can now select large language models (LLMs) for processing, including Google Gemini and Anthropic Claude on AWS, alongside Now LLM Service and Azure OpenAI.
    • Unified Skill for Extraction and Q&A: The Extract information from documents skill combines document data extraction and question answering into one platform skill, simplifying setup and review processes.
    • Language and Image Processing Improvements: OCR models now support multiple languages, including Simplified Chinese and Japanese. Image mode can be enabled for multimodal AI models to process images more efficiently.
    • Operational Sustainability Risk Management: Extraction capabilities now include ESG-related data from invoices to support compliance and sustainability workflows.
    • Role-Based Security and Access Control: Additional role configurations and ACLs enhance security for agentic workflows and AI agents, allowing differentiation between AI user privileges and human user privileges.
    • Usability Enhancements: Features such as opening test documents in new tabs, updated UI banners, and a consolidated review panel improve the user experience.
    • New Default Behaviors and Theme Updates: Now Assist in Document Intelligence skills are enabled by default for new installations, and the new Coral theme—including a dark mode option—is available for portal, web, and mobile experiences.
    • Support for Larger and More Complex Documents: Increased file size limits (up to 20MB) and page count limits (up to 200 pages depending on mode), including support for DOCX files.

    Activation and Integration

    Now Assist in Document Intelligence requires installation of the Document Intelligence application (sndocintel) and at least one Now Assist product from the ServiceNow Store. Multiple plugins are available tailored to various domains such as ITSM, HRSD, FSM, and more, enabling broad applicability across enterprise workflows.

    Practical Benefits for ServiceNow Customers

    • Streamlined document processing through AI-powered extraction and summarization, reducing manual effort.
    • Improved accuracy and flexibility by supporting multiple languages, file types, and AI model providers.
    • Enhanced security controls to meet enterprise compliance requirements for AI workflows.
    • Unified and simplified user experience for managing document intelligence skills and reviewing AI outputs.
    • Scalability to handle larger documents and complex extraction needs within operational and sustainability contexts.
    • Easy activation and integration with various Now Assist applications to extend AI capabilities organization-wide.

    Additional Notes

    • Usage measurement for Now Assist transitioned to a burn-down model resetting at contract anniversaries starting Zurich Patch 5.
    • Some previous standalone skills such as Document extraction and Document Q&A have been consolidated and removed from the platform skills list for clarity.
    • Role configurations are mandatory for agentic workflows and AI agents to function correctly post-installation.

    The ServiceNow® Now Assist in Document Intelligence application uses generative AI to extract information from documents, summarize document content, and provide answers to predefined questions. Now Assist in Document Intelligence was enhanced and updated in the Zurich release.

    Now Assist in Document Intelligence highlights for the Zurich release

    Zurich Patch 7
    • View citations for multiple attachments using the document and visual insights AI agent.
    • Create skills with data extraction, question answering, and summarization capabilities using document and visual intelligence in Now Assist Skill Kit.
    • Now Assist in Document Intelligence skills are now turned on by default.
    Zurich Patch 5
    • Review changes to Now Assist usage measurement.
    Zurich Patch 4
    • View a test document in a larger workspace on a separate browser tab during use case setup.
    • Use the document and visual insights AI agent to upload files, extract information without a predefined use case using a selected LLM, and display the results in a dedicated document view.
    • Additional role configuration is required for agentic workflows and AI agents included with Now Assist applications.
    Zurich Patch 1
    • Leverage document data extraction and document Q&A capabilities in a single Extract information from documents skill.
    • Create skills with data extraction, question answering, and summarization capabilities by using document and visual intelligence in Now Assist Skill Kit.
    • Choose the language for a use case to help the optical character recognition (OCR) model better detect the text to extract from your files.
    • Extract information from files with text written in Simplified Chinese or Japanese.
    • Extract information from documents for Operational Sustainability Management (Operational Sustainability Risk Management) workflows.
    Early Availability
    • Use Google Gemini and Anthropic Claude on AWS as AI model providers for Now Assist skills and AI agents in addition to Now LLM Service and Azure OpenAI.

    See Now Assist in Document Intelligence for more information.

    Important:
    Now Assist in Document Intelligence is available in the ServiceNow Store. For details, see the "Activation information" section of these release notes.

    New in the Zurich release

    Zurich Patch 7
    Document and visual insights AI agent enhancements
    The document and visual insights AI agent can recognize and provide citations for multiple attachments.
    Zurich Patch 4
    Open document in a new tab
    View a test document in a larger workspace on a separate browser tab during use case setup.
    Image captioning in Knowledge Center
    Extract information from images to support accessibility captioning in knowledge base articles.
    Document and visual insights AI agent enhancements
    Leverage features that enable you to use the document and visual insights AI agent with selected LLMs to upload files, extract information without a predefined use case, and display results in a dedicated document view.
    • Select an LLM for the AI agent to use when processing documents and images.
    • Now LLM Service is not an available option for LLM selection.
    • Prompt the AI agent to upload a file to process.
    • View the results of the AI agent actions in a document view screen.
    Zurich Patch 1
    Document Intelligence for Now Assist Skill Kit
    Use document and visual intelligence to leverage extraction, question answering, and summarization capabilities for a skill created with Now Assist Skill Kit.
    Language model selection
    Select the language for a use case to help the optical character recognition (OCR) model to detect text in the document and image files.
    Image mode selection for use cases
    Turn on image mode to process images more efficiently for use cases that use a multimodal AI model.
    Model provider flexibility at the use case level
    Choose a model provider at the use case level, including a Now Assist default option that uses the instance-level model selected in AI Control Tower.
    Information extraction for Operational Sustainability Risk Management
    Extract ESG-related information from invoices to support Operational Sustainability Risk Management initiatives and regulatory compliance.
    Improved security for the document and visual insights AI agent
    Enable access control lists (ACLs) to improve the security for the document and visual insights AI agent. ACLs provide you with the capability to run AI agents and agentic workflow executions either as a dynamic user or an AI user.

    The document and visual insights AI agent and its skills are classified as a worker or helper agent. A worker can act as an AI user with different privileges than the human user. A helper only has the privileges of a human user.

    Early Availability
    New third-party AI model provider options available for all Now Assist applications
    Google Gemini and AWS Claude are available for Now Assist skills and AI agents in addition to Now LLM Service and Azure OpenAI.

    UI changes

    Zurich Patch 8
    Updated banner in the extraction review panel
    Changed the label in the extraction review panel banner from Document Q&A to Extracted data.
    Zurich Patch 1
    Usability changes for the Extract information from documents skill
    Various features for the Extract information from documents skill are updated to unify the experience for data extraction and document Q&A.
    • The Extract information from documents skill has been added to the list of Platform skills in the Now Assist Admin console to incorporate both the Document extraction and the Document Q&A skills.
    • In the use case setup, the panel for defining a field displays the options for Field, Question, and Table.
    • Table columns defined for a use case are displayed in a group in the fields list on the use case page.
    • The Document Intelligence workspace displays the fields, tables, and questions in the same review panel.
    Early Availability
    Coral theme
    Coral is now the default theme for new portal, web, and mobile experiences with Next Experience or Core UI enabled. This theme provides a fresh look and feel, featuring brand-neutral illustrations to enhance your user experience. A dark theme option is available for web and mobile experiences.

    Changed in this release

    Zurich Patch 7
    Document and visual insights AI agent enhancements
    Tools used by the document and visual insights AI agent are consolidated to improve performance.
    Changes to limitations
    The file size limit for uploading a file using the attachment summarization feature is changed from 10MB to 20MB.
    Document Intelligence for Now Assist Skill Kit
    The document and visual intelligence capabilities used to leverage extraction, question answering, and summarization capabilities for skills created with Now Assist Skill Kit are available to users with the appropriate role(s).
    Now Assist in Document Intelligence skills are now turned on by default
    The new default behavior works as follows:
    • New customers: When you install a Now Assist product, designated skills are turned on automatically.
    • Existing customers who are upgrading: Any previously unconfigured skill is turned on automatically (the skill was never configured and turned on, then turned off again). Previously configured skills that were turned on, then off, remain inactive.
    Zurich Patch 5
    Changes to Now Assist usage measurement
    Starting with Zurich Patch 5, Now Assist usage measurement is transitioning from a 365-day look-back model to a 365-day burn-down model, with usage resetting at the contract anniversary date. For more information, refer to KB KB2704710: Now Assist Usage - Overview & New Measurement Logic.
    Zurich Patch 4
    Changes to limitations
    The page count limit decreases to 20 pages per file for an extraction based on a use case with a table defined.
    Role configuration required for agentic workflows and AI agents
    Agentic workflows and AI agents included with Now Assist applications require additional security configuration. If you select Users with selected roles for your user access security controls for an agentic workflow or AI agent, you must add the installed roles, or they won't execute. Data access settings must also include these roles. See the documentation for the agentic workflow or AI agent for the specific roles you must add. After the roles are configured, users must have the specified role to invoke the agentic workflow or AI agent.
    Zurich Patch 1
    Merged skills
    Document data extraction and document Q&A capabilities are available in a single Extract information from documents skill.
    • The Extract information from documents skill is available in the list of Platform skills in the Now Assist Admin console.
    • Data extraction and document Q&A capabilities can be set up for the same use case.
    • Agents can review the AI predictions for the fields, tables, and questions in the same task.
    Document and visual insights AI agent enhancements
    Leverage added tools that enable the AI agent to display the extracted data and to look up existing use cases or create use cases based on descriptions and document attachments.
    Changes to limitations
    The limitations set for Now Assist in Document Intelligence properties are updated to support processing larger documents.
    • DOCX files are supported for information extraction.
    • The page count limit is 200 pages per file If image mode is turned off for the use case. If image mode is turned on, the page count limit is 10 pages per file.
    • The file size limit is 20 MB.

    Removed in this release

    • The Document extraction skill has been removed from the list of Platform skills in the Now Assist Admin console.
    • The Document Q&A skill has been removed from the list of Platform skills in the Now Assist Admin console.

    Activation information

    Now Assist features are available with activation of any Now Assist plugin from ServiceNow Store. The following plugins are available:

    For more information, see Configuring Now Assist in Document Intelligence.

    Additional requirements

    Now Assist in Document Intelligence requires the installation of the Document Intelligence application (sn_docintel) and at least one Now Assist product.

    Accessibility information

    Dark theme
    The new Coral theme includes a dark theme option for web and mobile experiences. This option is commonly used to alleviate eye strain and improve readability.

    Localization information

    Zurich Patch 1
    Support for files in multiple languages
    Process files with text in multiple languages, including Simplified Chinese and Japanese.

    Related ServiceNow applications and features

    Document Intelligence
    Use ServiceNow® Document Intelligence to streamline the extraction and classification of data from a variety of documents. Document Intelligence is designed to work with existing automated systems, making operations more efficient and saving effort.
    Now Assist
    Improve the productivity and efficiency in your organization, deliver better self-service, recommend actions, provide answers, and empower your users to search more effectively.
    Now Assist Skill Kit
    Use the generative AI skills in ServiceNow® Now Assist products to meet the needs of users in different workflows, including case or incident summarization, chat summarization, resolution notes generation, and code generation.
    Now Assist in Contract Management
    Use ServiceNow® Now Assist in Contract Management to analyze contracts for non-standard and missing clauses, and extract information from signed contracts to automatically add in the contract repository.