ExamFinch.com My Study

Study guide

6.2 Generating a Document from an Existing Document

Practice this exam

Microsoft AI Business Professional AB-730

Track your progress as you read

Finished sections get a check mark.

1 Introduction & Exam Overview

0.1 About the AB-730 Exam 0.2 How to Use This Study Guide

2 Chapter 1: Understand Generative AI Capabilities Across Microsoft 365

1.1 How Copilot Works: LLMs, Microsoft Graph & the Security Boundary 1.2 Context & Grounding: Work Files, Web Data and App Context 1.3 Chat Experiences vs. Agent Experiences 1.4 When to Create Your Own Agent 1.5 Copilot Capabilities Across Microsoft 365 Apps

3 Chapter 2: Responsible AI & Data Protection Practices

2.1 Generative AI Risks: Fabrications, Prompt Injection & Over-Reliance 2.2 Verification Steps: Citation Checks & Human Review 2.3 Recognizing & Mitigating Risks to Sensitive Data 2.4 How Data Protection Restricts Prompt Results

4 Chapter 3: Creating and Managing Prompts

3.1 Creating an Effective Prompt: The GCSE Framework 3.2 Selecting Resources to Reference in a Prompt 3.3 Saving and Sharing Prompts in Copilot Prompt Gallery 3.4 Scheduling Prompts

5 Chapter 4: Managing Conversations and Notebooks

4.1 Finding Previous Conversations 4.2 Renaming and Deleting Chats 4.3 Adding a Conversation to a Copilot Notebook

6 Chapter 5: Creating and Managing Microsoft 365 Copilot Agents

5.1 Agent Store vs. Creating a New Agent 5.2 Creating an Agent from a Template 5.3 Configuring an Agent That Has Knowledge 5.4 Configuring Instructions, Capabilities & Suggested Prompts 5.5 Sharing an Agent with Team Members

7 Chapter 6: Drafting and Analyzing Business Content

6.1 Creating a New Document from a Prompt 6.2 Generating a Document from an Existing Document 6.3 Generating a Management Summary 6.4 Moving Data & Insights Between Microsoft 365 Apps 6.5 Drafting Communications with Copilot in Outlook

8 Chapter 7: Meetings, Collaboration and Personalization

7.1 Using Copilot for Meetings: Setup & Live Intelligence 7.2 Meeting Recap, Transcripts & Catch-Up 7.3 Copilot Pages for Collaboration 7.4 How Copilot Uses Memory & Instructions

6.2 Generating a Document from an Existing Document

Key Takeaways Explain This Topic AI

Reference an existing file when drafting so Copilot rewrites verified content instead of generating unverified prose from the model's own knowledge.

In Word, referencing a source file turns a blank-page task into a transformation task, which materially lowers fabrication risk.

Rewrite, tone adjustment, expand, shorten, and Visualize as a Table act on text you have selected, not on the whole document.

Reusing an approved source document preserves the numbers and commitments that a from-scratch draft would invent.

Always review generated transformations against the source: Copilot preserves meaning well but can drop qualifiers that carry legal weight.

Last updated: August 2026

6.2 Generating a Document from an Existing Document

Quick Answer: The blueprint separates "create a new document from a prompt" from "generate a document from an existing document" because the risk profiles differ. Drafting from a prompt asks the model to supply both structure and substance. Drafting from a source document asks it only to re-express substance that already exists — which is why source-grounded generation is the default choice for anything containing figures, dates, or commitments.

Source-Grounded Generation (File-Referenced Drafting)

Source-grounded drafting anchors Copilot's generative process in verified organizational assets. By referencing existing files, Copilot extracts specific domain context, project milestones, budget figures, and stakeholder responsibilities, synthesizing them into a new document structure.

Ideal Use Cases: Converting a raw project charter into a formal client proposal, synthesizing multiple meeting transcripts into an executive briefing memo, or turning a technical specification into a customer-facing product overview.

The / (Forward Slash) Referencing Mechanism: Typing / in the "Draft with Copilot" dialog opens a dynamic file picker displaying recently accessed documents from OneDrive for Business and SharePoint Online. Users can select up to three reference files to ground the generation.

Operational Dimension Blank-Canvas Generation Source-Grounded Generation ( / )

Information Source Base LLM training knowledge + User prompt instructions Explicit enterprise files retrieved via Microsoft Graph

Prompt Complexity Required High; must supply context, domain background, and parameters Medium; prompt defines desired structure and transformation goal

Hallucination Risk Moderate to High if asked for specific corporate facts Low; model strictly synthesizes grounded source text

  • Citation Provenance None; content is newly synthesized Generates references grounded in source documents
  • Reference Limit N/A (No files attached) Up to 3 cloud-stored files ( .docx , .pptx , .pdf , transcripts)

Exam Warning: Copilot in Word can only reference files stored in OneDrive for Business or SharePoint Online . Files stored on a local computer drive (e.g., C:\Users\Desktop\file.docx ) that have not synchronized with the Microsoft 365 cloud tenant cannot be resolved by Microsoft Graph and will not appear in the / file picker.

1. In-Line Transformation & Iterative Refinement

Beyond drafting complete sections, Copilot in Word serves as an intelligent in-line editor. By highlighting any section of existing text, users unlock contextual transformation tools. ┌─────────────────────────────────────────────────────────────────────────────┐

  • │ IN-LINE TRANSFORMATION TOOLS │
  • ├──────────────────────────┬──────────────────────────┬───────────────────────┤
  • │ REWRITE WITH COPILOT │ TONE ADJUSTMENT │ VISUALIZE AS A TABLE │
  • │ │ │ │
  • │ [Selected Paragraph] │ [Selected Text] │ [Selected Prose/List] │
  • │ │ │ │ │ │ │
  • │ ▼ │ ▼ │ ▼ │
  • │ Alternative phrasing │ • Professional │ Formatted Word Table │
  • │ with side-by-side │ • Casual │ with auto-generated │
  • │ comparison view │ • Concise │ column headers │
  • └──────────────────────────┴──────────────────────────┴───────────────────────┘
  • 1. Rewrite with Copilot

Highlighting a sentence or paragraph and selecting the Copilot icon > Rewrite triggers an automated linguistic refactoring. Copilot generates multiple candidate revisions that enhance clarity, eliminate passive voice, and improve syntactic flow while preserving the core message.

2. Tone Shifting & Register Modification

Copilot provides pre-configured tone settings that instantly adapt selected text to different business contexts:

Professional: Enhances formal vocabulary, enforces objective phrasing, and removes conversational idioms. Ideal for client proposals, executive summaries, and legal disclosures.

Casual: Softens rigid structures, introduces conversational transitions, and enhances warmth. Suitable for internal team updates, newsletters, and informal project retrospectives.

Concise: Aggressively removes filler words, eliminates redundancy, and condenses compound sentences into direct, punchy assertions. Designed for busy executives and high-impact briefing memos.

3. Modifying Length (Expand vs. Shorten)

Expand: Instructs Copilot to elaborate on core concepts, add supporting justifications, or provide illustrative enterprise examples for high-level bullet points.

Shorten: Compresses multi-paragraph explanations into executive summaries or single-sentence takeaways while preserving essential data points.

4. Visualize as a Table

Dense narrative text containing comparative metrics, product features, or step-by-step responsibilities can be overwhelming to read. Selecting unstructured text and choosing Visualize as a Table prompts Copilot to:

Analyze the underlying categorical dimensions in the text.

Create a formatted Word table with structured column headers.

Populate the cells with extracted facts, preserving numerical figures and owner attributions.

Reviewing a Derived Document

Transformation is not lossless. Copilot reliably preserves the main assertions of a source but can quietly drop the qualifiers that lawyers and auditors care about — "up to", "subject to board approval", "in the pilot region only". Before a derived document leaves your hands:

Diff the numbers. Every figure in the output should appear in the source. Any figure that does not is either inferred or invented.

Check the hedges. Search the source for conditional language and confirm it survived.

Confirm the scope. A summary that silently promotes a regional pilot into a company-wide programme is the most common and most damaging transformation error.

  • Task Better approach Why
  • Board paper from an approved strategy memo Reference the memo Figures and commitments are already agreed
  • Client-facing summary of a signed SOW Reference the SOW Scope language must match the contract exactly

Brainstorm of possible campaign themes Prompt-only drafting No source of truth exists yet, and invention is the point

On the Exam: When a scenario involves an existing approved artifact — a contract, a policy, a signed proposal — the expected answer references that file rather than describing its contents in the prompt.