Fixing per content author

Author: Randall-Knapp

learn-pr/wwl/analyze-requirements-for-ai-powered-business-solutions/includes/2-assess-use-agents-task-automation-data-analytics-decision-making.md

@@ -2,13 +2,10 @@
 
 By the end of this unit, learners will be able to:
 
-Explain how AI agents support **task automation**, **data analytics**, and **decision-making**.
-
-Describe how **Microsoft Copilot experiences** enhance productivity across workflows.
-
-Identify how **generative AI** unlocks productivity and supports responsible decision-making.
-
-Assess scenarios where agents add measurable value for enterprise environments.
+- Explain how AI agents support **task automation**, **data analytics**, and **decision-making**.
+- Describe how **Microsoft Copilot experiences** enhance productivity across workflows.
+- Identify how **generative AI** unlocks productivity and supports responsible decision-making.
+- Assess scenarios where agents add measurable value for enterprise environments.
 
 ## 1. Introduction
 
@@ -34,7 +31,7 @@ Triggering multi-step processes (approvals, notifications, content generation).
 
 These capabilities reduce cognitive load and help teams focus on strategic, not repetitive, work.<br>
 
-**Chart: Examples of Agent-Driven Task Automation**
+**Examples of Agent-Driven Task Automation**
 
 | **Task Area** | **How Agents Help** | **Tools** |
 |---|---|---|
@@ -61,7 +58,7 @@ Copilot experiences help employees make sense of data without requiring advanced
 
 **Visual Diagram: AI Agents in the Analytics Workflow**
 
-User Question → AI Agent → Data Retrieval → Insight Generation → Recommended Actions
+:::image type="content" source="../media/ai-agents-analytics-workflow.png" alt-text="Diagram showing how AI agents integrate into the analytics workflow, from data ingestion through insight generation and decision support.":::
 
 ## 4. Agents in Decision-Making
 
@@ -99,16 +96,4 @@ Use these links as the primary sources for this unit:
 
 **Explore Copilot Experiences**<br>[Explore Copilot Experiences](/training/modules/business-value-microsoft-copilot-solutions/3-explore-copilot-experiences) 
 
-**Unlock Productivity with Generative AI - Microsoft Learn**<br>[Unlock Productivity with Generative AI](/training/modules/generative-ai-productivity/)
-
-**Knowledge Check**
-
-**1. Multiple Choice**
-
-**Which of the following is a key benefit of using AI agents for data analytics?**<br>A. They replace the need for all BI tools<br>B. They translate natural language questions into insights<br>C. They eliminate the need for data governance<br>D. They require manual configuration for every dataset
-
-**Correct Answer: B**
-
-**2. Open Reflection / Discussion**
-
-**How could AI agents improve decision-making in your team's current processes? Provide an example.**
+**Unlock Productivity with Generative AI - Microsoft Learn**<br>[Unlock Productivity with Generative AI](/training/modules/generative-ai-productivity/)

learn-pr/wwl/analyze-requirements-for-ai-powered-business-solutions/includes/3-review-data-grounding-accuracy-relevance-timeliness-cleanliness-availability.md

@@ -2,13 +2,10 @@
 
 After completing this unit, learners will be able to:
 
-Evaluate the quality of grounding data across five dimensions: **accuracy, relevance, timeliness, cleanliness, availability**.
-
-Explain why grounding is essential for reliable AI agent behavior.
-
-Identify data sources appropriate for grounding AI using Microsoft 365 and Copilot Studio.
-
-Apply best practices when preparing and validating grounding data.
+- Evaluate the quality of grounding data across five dimensions: **accuracy, relevance, timeliness, cleanliness, availability**.
+- Explain why grounding is essential for reliable AI agent behavior.
+- Identify data sources appropriate for grounding AI using Microsoft 365 and Copilot Studio.
+- Apply best practices when preparing and validating grounding data.
 
 ## 1. Understanding Grounding in AI Agents
 
@@ -20,67 +17,73 @@ To support advanced grounding, the **Copilot Retrieval API** retrieves relevant
 
 ## 2. The Five Dimensions of Grounding Data Quality
 
-## 2.1 Accuracy
+### 2.1 Accuracy
 
 Data should reflect real, verifiable facts. Inaccurate content leads to incorrect or harmful agent outputs.<br>AI agents must use **trusted and validated datasets** during retrieval to avoid generating incorrect responses. 
 
 **Indicators of accuracy:**
 
-Verified by SMEs
+- Verified by SMEs
+
+- Matched to authoritative sources
 
-Matched to authoritative sources
+- Free from errors or outdated assumptions
 
-Free from errors or outdated assumptions
+### 2.2 Relevance
 
-## 2.2 Relevance
+Grounding data must match the **intended use case** of the agent.
 
-Grounding data must match the **intended use case** of the agent.<br>When data is irrelevant, semantic search may retrieve conceptually similar—but contextually wrong—content.<br>
+When data is irrelevant, semantic search may retrieve conceptually similar—but contextually wrong—content.
 
 Relevance ensures the model surfaces information aligned with the user's scenario, workflow, or business domain.
 
-## 2.3 Timeliness
+### 2.3 Timeliness
 
-AI outputs degrade when data is stale.<br>The **semantic index** in Microsoft 365 continuously updates as content changes, ensuring the grounding layer reflects the latest documents, conversations, and knowledge.<br>
+AI outputs degrade when data is stale.
+
+The **semantic index** in Microsoft 365 continuously updates as content changes, ensuring the grounding layer reflects the latest documents, conversations, and knowledge.
 
 Timeliness includes:
 
-Modified dates
+- Modified dates
 
-Seasonal or compliance-related updates
+- Seasonal or compliance-related updates
 
-Data refresh schedules
+- Data refresh schedules
 
-## 2.4 Cleanliness
+### 2.4 Cleanliness
 
-Clean data reduces noise and increases retrieval precision. 
+Clean data reduces noise and increases retrieval precision.
 
-Data Pollution in AI refers to the degradation of data quality that negatively impacts the performance and reliability of AI systems. 
+Data Pollution in AI refers to the degradation of data quality that negatively impacts the performance and reliability of AI systems.
 
-<br>Cleaner data improves embedding quality and helps the agent retrieve the most appropriate content.
+Cleaner data improves embedding quality and helps the agent retrieve the most appropriate content.
 
 Clean data characteristics:
 
-Clear structure
+- Clear structure
+
+- No duplicates
 
-No duplicates
+- Minimal irrelevant metadata
 
-Minimal irrelevant metadata
+- Stable formatting and predictable layout
 
-Stable formatting and predictable layout
+### 2.5 Availability
 
-## 2.5 Availability
+Agents can only ground responses from **data the user has access to**.
 
-Agents can only ground responses from **data the user has access to**.<br>The Retrieval API respects permissions and will not return content beyond a user's access scope.<br>
+The Retrieval API respects permissions and will not return content beyond a user's access scope.
 
 Availability depends on:
 
-Storage in SharePoint/OneDrive or connected systems
+- Storage in SharePoint/OneDrive or connected systems
 
-Proper indexing in Microsoft Graph
+- Proper indexing in Microsoft Graph
 
-Clear access controls
+- Clear access controls
 
-## 3. Chart: Dimensions of Grounding Data Quality
+## 3. Dimensions of Grounding Data Quality
 
 | **Dimension** | **Definition** | **Impact on Agent Performance** |
 |---|---|---|
@@ -92,7 +95,7 @@ Clear access controls
 
 ## 4. Diagram: How Microsoft Copilot Grounds AI Responses
 
-  _Diagram grounded in documentation describing semantic index + Retrieval API pipeline._<br>
+  _Diagram grounded in documentation describing semantic index + Retrieval API pipeline.
 
 ## 5. Best Practices for Reviewing Grounding Data
 
@@ -116,20 +119,4 @@ _Semantic indexing for Microsoft 365 Copilot_<br>[Semantic indexing for Microsof
 
 _Microsoft 365 Copilot Retrieval API_<br>[Microsoft 365 Copilot Retrieval API](/microsoft-365-copilot/extensibility/api/ai-services/retrieval/copilotroot-retrieval)
 
-_Manage Grounding With Bing in Microsoft Foundry and Azure_<br>[Manage Grounding With Bing](/azure/ai-foundry/agents/how-to/manage-grounding-with-bing)
-
-**Knowledge Check**
-
-**1. Multiple Choice**
-
-Which grounding data dimension ensures the model retrieves information aligned with the intended business scenario?
-
-A. Cleanliness<br>B. Availability<br>C. Relevance<br>D. Timeliness
-
-**Correct Answer: C. Relevance**
-
-**2. Discussion Question**
-
-What would happen if you do not take grounding into consideration in your AI strategy?
-
-How might poor data timeliness impact an AI agent that relies on policy or compliance documents?<br>Provide examples from your organization's workflows
+_Manage Grounding With Bing in Microsoft Foundry and Azure_<br>[Manage Grounding With Bing](/azure/ai-foundry/agents/how-to/manage-grounding-with-bing)

learn-pr/wwl/analyze-requirements-for-ai-powered-business-solutions/includes/4-organize-business-solution-data-available-other-ai-systems.md

@@ -2,29 +2,26 @@
 
 By the end of this unit, learners will be able to:
 
-Explain why wellorganized business solution data is essential for AI readiness.
-
-Identify architectural components that enable AI agents and AI systems to consume organizational data.
-
-Apply best practices for structuring, storing, indexing, and exposing data so AI systems can use it reliably.
-
-Understand how the Azure platform, Microsoft databases, and data architecture patterns support enterprise AI scenarios.
+- Explain why wellorganized business solution data is essential for AI readiness.
+- Identify architectural components that enable AI agents and AI systems to consume organizational data.
+- Apply best practices for structuring, storing, indexing, and exposing data so AI systems can use it reliably.
+- Understand how the Azure platform, Microsoft databases, and data architecture patterns support enterprise AI scenarios.
 
 ## 1. Introduction
 
 AI systems—whether Copilot, autonomous agents, or custom-built AI applications—require **highquality, structured, and accessible data**. Poorly organized data leads to weak grounding, hallucinations, data quality issues, and unreliable decisionmaking.
 
 Organizing business solution data is not only a technical requirement but a business imperative. When data is structured correctly, it becomes available for:
 
-Copilot for Microsoft 365
+- Copilot for Microsoft 365
 
-AI agents built in Copilot Studio
+- AI agents built in Copilot Studio
 
-Custom AI apps built with Azure AI
+- Custom AI apps built with Azure AI
 
-Retrievalaugmented generation (RAG) pipelines
+- Retrievalaugmented generation (RAG) pipelines
 
-Analytics and automation solutions
+- Analytics and automation solutions
 
 Retrieval-Augmented Generation (REG) is an architecture that separates prototypes from trustworthy systems. A RAG pipeline is the system that performs all the steps required to make RAG work in a production environment, handling the data ingestion, streaming, cleaning, chunking, embedding, indexing, retrieval, prompt assembly, orchestration, and monitoring that allow an LLM to use retrieved context when generating an answer There are several advantages of leveraging RAG pipelines:
 
@@ -36,7 +33,7 @@ This unit explains how to organize your business data to become **usable, discov
 
 ## 2. Key Concepts for Organizing AIReady Data
 
-## 2.1 Drive Transformation with Azure Data & AI Platforms
+### 2.1 Drive Transformation with Azure Data & AI Platforms
 
 _(From: Leverage AI tools - Drive Transformation on Azure)_
 
@@ -50,7 +47,7 @@ Azure provides the foundational components necessary to **centralize, transform,
 
 **Interoperability**: APIs, event hubs, and data streaming allow multiple AI systems to use the same data.
 
-**Chart - Azure Data Estate for AI**
+**Azure Data Estate for AI**
 
 | **Layer** | **Purpose** |
 |---|---|
@@ -59,7 +56,7 @@ Azure provides the foundational components necessary to **centralize, transform,
 | **Intelligence Layer (Search + Vector Index)** | Enable grounding, retrieval, semantic search |
 | **AI Apps + Agents** | Copilot, custom RAG apps, autonomous agents |
 
-## 2.2 Data Architecture for AI Agents Across the Organization
+### 2.2 Data Architecture for AI Agents Across the Organization
 
 _(From: Data architecture for AI agents across your organization - Cloud Adoption Framework)_
 
@@ -77,7 +74,7 @@ Microsoft's Cloud Adoption Framework defines how to design a **data architecture
 
 **Visual Diagram - AI Agent Data Architecture**
 
-## 2.3 Make Databases AIReady
+### 2.3 Make Databases AIReady
 
 _(From: Building intelligent AI apps with Microsoft databases)_
 
@@ -93,13 +90,13 @@ Microsoft database technologies provide builtin intelligence for AI:
 
 Use cases:
 
-Storing embeddings for RAG apps
+- Storing embeddings for RAG apps
 
-Managing structured and unstructured content
+- Managing structured and unstructured content
 
-Supporting real-time AI agent decisions
+- Supporting real-time AI agent decisions
 
-Performing high-volume transactions required by autonomous agents
+- Performing high-volume transactions required by autonomous agents
 
 ## 3. Best Practices for Organizing Business Data for AI Systems
 
@@ -119,27 +116,27 @@ Microsoft 365 Copilot requires semantic indexing for grounding content correctly
 
 Expose data through:
 
-APIs
+- APIs
 
-Search indexes
+- Search indexes
 
-RAG pipelines
+- RAG pipelines
 
-Graph connectors
+- Graph connectors
 
-SQL endpoints
+- SQL endpoints
 
 **✔ 5. Implement governance early**
 
 Use Purview for:
 
-Access policies
+- Access policies
 
-Sensitivity labels
+- Sensitivity labels
 
-Lineage
+- Lineage
 
-Data quality rules
+- Data quality rules
 
 **✔ 6. Keep data authoritative & updated**
 
@@ -153,18 +150,4 @@ Use these links for this unit:
 
 [Building intelligent AI apps with Microsoft databases](https://techcommunity.microsoft.com/blog/azuredatablog/building-intelligent-ai-apps-with-microsoft-databases/4413833)
 
-[Data architecture for AI agents](/azure/cloud-adoption-framework/ai-agents/data-architecture-plan) 
-
-**5. Knowledge Check**
-
-**Question 1 — Multiple Choice**
-
-Which component ensures AI agents can retrieve and interpret enterprise data accurately?
-
-A. User interface customization<br>B. Semantic indexing<br>C. Microsoft Teams channels<br>D. Email distribution lists
-
-**Correct Answer: B. Semantic indexing**
-
-**Question 2 — Open Discussion**
-
-Your team wants to deploy an AI agent that answers policy and process questions.<br>What types of data must be collected and categorized first? What risks exist if the data is not governed?
+[Data architecture for AI agents](/azure/cloud-adoption-framework/ai-agents/data-architecture-plan)