---
title: Conversational authoring for announcements
description: Use Now Assist Chat to create Employee Slate announcements through natural language prompts and AI-assisted content generation.
locale: en-US
canonical_url: https://www.servicenow.com/docs/r/australia/employee-service-management/employee-experience-foundation/eslate-conversational-authoring-announcements.html
release: australia
product: Employee Experience Foundation
classification: employee-experience-foundation
topic_type: concept
last_updated: "2026-04-02"
reading_time_minutes: 1
keywords: [conversational authoring, ServiceNow Otto Chat, AI announcements, Employee Slate]
breadcrumb: [Employee communications, Working with EmployeeWorks capabilities, ServiceNow EmployeeWorks Web App, Unified Employee Experience, Employee Service Management]
---

# Conversational authoring for announcements

Use Now Assist Chat to create Employee Slate announcements through natural language prompts and AI-assisted content generation.

Conversational authoring with Now Assist Chat streamlines the announcement creation process by using natural language prompts and AI-generated content. Content managers can create announcements by describing their intent to the AI assistant, which then generates draft content based on existing knowledge articles or catalog items.

\[Omitted image "es-comm-authoring.png"\] Alt text: Integrated conversational authoring for announcements

## How conversational authoring works

The conversational authoring process follows a structured flow that guides content managers from initial prompt to published announcement:

-   The content manager provides a natural language prompt to Now Assist Chat, including an article ID, link, or context from an open knowledge article or catalog item.
-   The AI assistant processes the referenced content and generates an announcement draft using text and images from the source material.
-   The assistant requests additional details from the content manager, such as target audience and scheduling preferences.
-   The content manager enters the authoring edit mode to review, adjust, and finalize the announcement before publishing.

## AI content generation

When creating announcements from existing content, the AI assistant automatically extracts and transforms relevant information:

-   **Headlines**: Generated from the source article title or catalog item name
-   **Body text**: Summarized from the first paragraph or description of the source content
-   **Images**: Inherited from the source content when available
-   **Links**: Automatically configured to link to the source knowledge article or catalog item
-   **User criteria**: Inherited from the source content audience targeting

    **Note:** User criteria is only applicable to announcement flows targeting catalog/KBs.


## Human review and editing

All AI-generated announcements are drafts that require human review before publishing. Content managers can modify any aspect of the generated content, including:

-   Headline and body text for tone and accuracy
-   Image selection and focal point positioning
-   Content priority and publishing schedule
-   Audience targeting and chat promotion settings

This review process helps maintain content quality and alignment with organizational communication standards.

## Usage guidelines

When using conversational authoring for announcements:

-   Provide clear, specific prompts that include the article ID or link to the source content
-   Review all generated content for accuracy, tone, and alignment with organizational policy
-   Verify that inherited user criteria and audience targeting meet your communication goals
-   Test image focal points across different display contexts before publishing
-   Consider staggering chat promotions to avoid overwhelming communication channels