AI Foundry – Page 3 – GM Dynamics & Agentic AI Consulting

Based on the Microsoft Business Applications Launch Event blog post from October 9, 2025, here is the breakdown of the major AI innovations announced for the 2025 Release Wave 2.

Microsoft Business Applications Launch Event: 2025 Release Wave 2 Innovations

(Focusing on the transition from AI assistants to Agentic AI across Dynamics 365, Power Platform, and Copilot Studio).

Real-time User Journey: The Agentic Workflow

This journey showcases how new autonomous agents collaborate across the platform to resolve a complex business challenge:

Detection: A Service Operations Agent detects a spike in support tickets related to a specific product defect.
Collaboration: The Service Agent automatically alerts a Supply Chain Agent, providing a summary of the affected batches.
Autonomous Action: The Supply Chain Agent checks inventory levels and proactively contacts the supplier to pause orders and request replacements.
Sales Insight: Meanwhile, Dynamics 365 Sales agents receive a notification from Copilot informing them of the delay for their open deals, along with AI-generated talking points to manage customer expectations.
Supervisor Oversight: The Operations Manager monitors this entire “Agent-to-Agent” interaction through the Agent Activity Feed, ensuring the logic remains within company policy before the agents finalize the supplier communications.

Step-by-Step: How to Enable These Innovations

Because this event covers a “Wave” of features, enabling them follows the standard Release Wave process:

Step 1: Access the Release Planner

Navigate to the Microsoft Release Planner to identify which specific agents (e.g., Sales, Finance, or Service) are available for your region and environment type.

Step 2: Environment Validation

Log in to the Power Platform Admin Center. Create a Sandbox environment to test new features without impacting production.

Step 3: Opt-in to 2025 Wave 2

Go to Environments > [Your Env] > Settings > Updates. Click “Update now” under the 2025 Release Wave 2 header to enable early access features.

Step 4: Configure Copilot Studio Agents

Open Microsoft Copilot Studio. Use the new Agentic Templates to build autonomous agents that can trigger Power Automate flows and query Dataverse without manual intervention.

Step 5: Enable the Agent Hub

In the Customer Service/Contact Center Admin Center, navigate to the Agent Hub to toggle on specific autonomous agents like the Knowledge Management Agent or Case Management Agent.

Step 6: Deploy to Users

Update your Agent Experience Profiles to ensure the new Copilot sidecar features (like real-time web search and autonomous summaries) are visible to your staff.

Infographic: The Shift to Agentic AI

Feature	Copilot as Assistant (Current)	AI Agents (Wave 2)
Initiative	Reactive (Wait for user prompt).	Proactive (Starts work based on events).
Logic	Follows specific instructions.	Reasoning-based (Solves multi-step problems).
Integration	Limited to the current app.	Cross-platform (Talks to other agents/apps).
Human Role	Doing the work with AI help.	Supervising the AI doing the work.
Scale	1:1 interaction.	1:Many (Agents handle thousands of tasks).

References

Explore new AI innovation for Dynamics 365, Microsoft Power Platform, and Copilot Studio at the Business Applications Launch Event

AI Foundry, architecture, copilot-studio

What’s New in Copilot Studio: September 2025 – Expanding Reach, Automation, and Analytics

October 4, 2025May 3, 2026 Goloknath

What’s New in Copilot Studio: September 2025 – Expanding Reach, Automation, and Analytics

The September updates focused on three core areas: building richer agent experiences (specifically via Computer Use and WhatsApp), advanced developer tools like Code Interpreter, and deeper ROI/Analytics for measuring business impact.

Real-time User Journey: Computer Use (Public Preview)

The most innovative journey introduced is the ability for an agent to interact with UIs like a human:

Scenario Discovery: A user needs to enter data into an old legacy desktop application that doesn’t have an API.
Instruction: The user tells the agent, “Open the legacy ERP, look up invoice #123, and copy the total into this Excel sheet.”
Vision & Reasoning: The agent uses its “vision” to see the screen and its “reasoning” to identify buttons and text fields.
Execution: On a hosted Windows 365 browser or local device, the agent moves the virtual mouse, clicks, and types to complete the task.
Completion: The agent confirms the data has been moved and provides a screenshot or log of the completed action.

Step-by-Step: How to Enable WhatsApp (General Availability)

With WhatsApp integration now GA, you can deploy agents to customers globally:

Step 1: Prerequisites: Ensure you have a Meta Business Account and a verified phone number for WhatsApp.
Step 2: Access Channels: In Copilot Studio, open your agent and go to the Channels tab.
Step 3: Select WhatsApp: Choose WhatsApp from the list of available channels.
Step 4: Connect to Meta: Follow the guided wizard to link your Copilot Studio agent to your Meta Business Manager.
Step 5: Configure Interactions: Set up how the agent should handle attachments (images/files) and authentication (via phone number).
Step 6: Go Live: Once connected, your agent is accessible to anyone messaging that verified WhatsApp number.

Infographic: Scaling and Measuring AI Impact

The September update introduced a suite of “Enterprise-Ready” tools to manage agents at scale:

Capability	What it Does	Business Value
Code Interpreter (GA)	Executes Python code directly within agents.	Solves complex math, data visualizations, and CRUD operations on Dataverse.
ROI Analysis (GA)	Tracks savings in time or money per agent run.	Provides real-time data to justify AI investment and prioritize projects.
File Groups (GA)	Organizes up to 12,000 files into 25 groups.	Improves knowledge retrieval accuracy and reduces “context chaos.”
Agents Client SDK	Embeds agents in native Android, iOS, or Windows apps.	Keeps users in their flow of work without switching to Teams or Web.

References

What’s new in Copilot Studio: September 2025

AI Foundry, architecture, contact-centre, copilot-studio, Customer-service

Automate intent discovery and resolution with Customer Intent Agent in Dynamics 365 Contact Centre

August 24, 2025May 3, 2026 Goloknath

Automate intent discovery and resolution with Customer Intent Agent (A specialized AI agent that proactively identifies emerging customer needs and builds the necessary logic to resolve them).

Real-time User Journey

This journey highlights how the agent transitions from “Discovery” to “Resolution” without manual administrative work:

Passive Discovery: The Customer Intent Agent continuously scans thousands of historical transcripts and real-time interactions across voice and chat.
Pattern Identification: The AI identifies a new, rising trend: many customers are asking about “Loyalty Point Expiration” following a recent policy change—a category that didn’t exist in the current routing rules.
Intent Proposal: The agent alerts the supervisor: “I’ve discovered a new intent: ‘Loyalty Inquiry.’ Would you like me to create a resolution path for this?”
Autonomous Training: Upon approval, the agent automatically maps relevant phrases to this new intent and suggests the best knowledge articles or automated bot topics to address it.
Closing the Loop: The next customer who asks, “When do my points vanish?” is immediately recognized by the new intent and routed to the correct automated resolution, bypassing the general “Other” queue.

Step-by-Step: How to Enable This Feature

The Customer Intent Agent is part of the Agentic Contact Center suite. Follow these steps to activate it:

Step 1: Admin Center Access

Step 2: Access the Agent Hub

Navigate to Operations > Agent Hub. This is the centralized control center for all autonomous agents.

Step 3: Enable Intent Discovery

Locate the Customer Intent Agent card and select Manage. Toggle the Discovery setting to On. This allows the agent to begin analyzing historical and live data.

Step 4: Configure Data Sources

Specify which channels (Voice, Chat, Email) the agent should monitor. You can also point it toward specific historical timeframes (e.g., the last 30 days) to find recent trends.

Step 5: Set Approval Workflows

Choose between Semi-autonomous (the agent suggests new intents for your approval) or Fully autonomous (the agent creates and applies the intent logic automatically). Microsoft recommends Semi-autonomous for the first 30 days.

Step 6: Link to Copilot Studio

Ensure the agent is connected to your Copilot Studio environment so it can automatically suggest or create new bot topics based on the discovered intents.

Infographic: The Customer Intent Agent Lifecycle

Feature	Traditional Manual Discovery	Customer Intent Agent
Speed	Weekly or monthly manual audits.	Real-time continuous monitoring.
Accuracy	Subject to human bias/sampling.	Data-driven across 100% of transcripts.
Implementation	Manual rule-building & bot training.	Auto-suggests phrases and resolution paths.
Scalability	Hard to track multiple minor trends.	Captures long-tail and emerging issues instantly.
Outcome	Reactive updates to routing.	Proactive optimization of the service journey.

References

Automate intent discovery and resolution with Customer Intent Agent

AI Foundry, architecture, Azure Development, contact-centre, copilot-studio, Customer-service

Expanding SMS capabilities in Dynamics 365 Contact Center

August 23, 2025May 3, 2026 Goloknath

Expanding SMS capabilities in Dynamics 365 Contact Center (Introducing deeper integration with Azure Communication Services and enhanced two-way messaging features).

Real-time User Journey

This journey highlights how a retail customer uses the enhanced SMS features for an order issue:

Notification: A customer receives an automated SMS update: “Your order #789 has been delayed. Reply HELP to speak with an agent.”
Customer Reply: The customer replies “HELP.” The Unified Routing engine identifies the customer based on their phone number in the CRM.
Agent Interaction: The SMS conversation appears in the Agent Workspace. The agent sees the full history of automated notifications sent to this customer.
Rich Media Exchange: The customer sends a photo via MMS showing a damaged product they received previously. The agent views this directly in the chat window.
Proactive Resolution: The agent sends a link via SMS for the customer to pick a new delivery slot.
Contextual Handoff: If the customer asks to speak with someone, the agent can transition the SMS thread into a live voice call within the same interface, maintaining the transcript.

Step-by-Step: How to Enable This Feature

To set up expanded SMS using Azure Communication Services (ACS), follow these steps:

Step 1: Provision ACS Resource

Log into the Azure Portal and create an Azure Communication Services resource.

Step 2: Connect to Dynamics 365

In the Contact Center admin center, go to Channels > SMS. Select Add Channel and choose Azure Communication Services.

Step 3: Get an SMS Number

Follow the prompts within the ACS interface to purchase or port a toll-free or long-code phone number capable of SMS/MMS.

Step 4: Configure Workstream

Create a new Workstream for the SMS channel. Define the routing rules (e.g., route all SMS with the keyword “Order” to the Shipping Queue).

Step 5: Enable Two-Way Messaging

In the SMS channel settings, ensure the “Enable two-way SMS” toggle is set to On. This allows agents to reply to incoming messages rather than just sending outbound alerts.

Step 6: Set Up SMS Templates

Navigate to Workforce Management > Templates and create standardized SMS replies to ensure consistent branding and speed of response.

Infographic: SMS Channel Comparison

Feature	Basic SMS (Legacy)	Expanded SMS (New)
Communication Flow	One-way (Notifications only).	Two-way (Full conversation).
Media Support	Text only.	MMS (Images, PDFs, Links).
Routing Logic	Static/None.	Unified Routing (Intent-based).
Platform	Third-party gateways.	Native Azure Communication Services.
Context	Isolated messages.	Full CRM Context & Timeline.

References

Expanding SMS capabilities in Dynamics 365 Contact Center

AI Foundry, contact-centre, copilot-studio, Customer-service, Power Apps

Empowering frontier firms with the Case Management Agent in Dynamics 365 Contact Centre

August 16, 2025May 3, 2026 Goloknath

Empowering frontier firms with the Case Management Agent (Advancing autonomous and semi-autonomous case lifecycle management in Dynamics 365 Customer Service).

Real-time User Journey

This journey illustrates the Fully Autonomous workflow for an email-based customer inquiry:

Inquiry Receipt: A customer sends an email regarding a malfunctioning device.
Autonomous Creation: The Case Management Agent (CMA) detects the intent, creates a new case, and automatically populates fields (Priority, Product Category, etc.) using AI-powered prediction.
Intelligent Triage: The CMA realizes it needs more details. It automatically sends a follow-up email to the customer with specific clarifying questions.
Resolution Logic: Once the customer replies, the CMA searches the knowledge base and triggers a custom agent (configured with the firm’s specific troubleshooting logic).
Proposal & Follow-up: The CMA drafts and sends a resolution email. It then waits for a predefined interval (SLA-based).
Autonomous Closure: After sending a satisfaction confirmation and receiving no further objections within the set timeframe, the CMA closes the case and triggers the Knowledge Management Agent to document the fix.

Step-by-Step: How to Enable This Feature

To enable and configure the Case Management Agent, follow these administrative steps:

Step 1: Admin Center Access

Step 2: Navigate to Autonomous Agents

Go to Operations > Insights > Autonomous service agents (preview).

Step 3: Configure Case Creation Rules

Under the CMA settings, define the field mappings and business context (e.g., which channels like email or chat should trigger the agent).

Step 4: Set Resolution Parameters

Select the Knowledge Bases the agent should access and link any Custom Agents or Power Automate flows required for specialized resolution logic.

Step 5: Define Communication Templates

Choose the email templates for “Clarifying Questions” and “Proposed Resolution.” Set the Timing Intervals for follow-ups and the conditions for automatic closure.

Step 6: Toggle Autonomy Level

Decide if the agent should run in Semi-autonomous mode (drafting for human review) or Fully autonomous mode (direct interaction with customers).

Infographic: CMA Lifecycle Automation

Phase	Capability	Benefit
Creation	Auto-detection of intent across voice, chat, and email.	Eliminates manual data entry and triage.
Resolution	Drafts context-aware emails; triggers custom logic.	Accelerates time-to-resolution with expert accuracy.
Escalation	Proactively identifies when a human or supervisor is needed.	Ensures complex issues never fall through the cracks.
Follow-up	SLA-based check-ins to ensure customer satisfaction.	Improves customer sentiment without agent effort.
Closure	Fully automated case resolution and record update.	Maintains clean data and closes the “Knowledge Loop.”

References

Empowering frontier firms with the Case Management Agent