Here’s how you can now create a Canvas App using a Vibe coded experience. This is Node based and the type is Code.
Let’s see how we can get to a working app quickly.
This article is contributed. See the original author and article here.
A Dynamics 365 Customer Service admin has just configured the Case Management Agent to auto‑populate case fields from incoming emails and added field descriptions to help the AI make better predictions. The configuration looks right, but how will the admin ensure that the AI is able to predict case field values accurately? The new case enrichment simulation in Dynamics 365 Customer Service—now generally available—answers exactly that question.
Admins can measure AI‑powered field prediction accuracy against real organizational data before enabling case enrichment in production: test against historical records, review the results, refine field descriptions, and enable the feature when accuracy meets your bar.
Empower admins to validate AI quality
Case Management Agent simulations include:
Flexible test data — run predictions against historical case records or sample email and chat data uploaded via Excel
Field‑level match reports — see prediction accuracy across every field
Case‑level comparison view — review actual versus predicted values side by side
Excel export — share results offline with stakeholders
Re‑run capability — iterate after refining lookup field descriptions or prediction rules
Shadow runs — Validate predictions by activating case enrichment with live production data without writing to records.
Run your first simulation today
Don’t wait until production to find out how well the AI performs. Open the Copilot Service admin center, navigate to Case settings > Case Management Agent > Case creation and update > Simulation, and run your first test against historical data. Within minutes you will know exactly where to tune your field descriptions before turning on case enrichment.
For step‑by‑step guidance, read the documentation:
This article is contributed. See the original author and article here.
Supply chain disruption is no longer the exception. It is the daily reality every business operates, driven by geopolitical instability, supplier failures, regulatory changes, and freight volatility. The challenge is not identifying disruption; it is synthesizing it fast enough to act. Businesses are flooded with signals across suppliers, logistics, operations, and compliance, but those signals remain fragmented across systems and teams, forcing reactive trade-offs instead of strategic action.
The leaders in this next era will be defined not by visibility alone, but by their ability to turn disruption into coordinated, cross-functional action before the cost of delay becomes too high. That’s where agentic AI comes in.
As supply chains become more dynamic, the way work gets done and operations are run is changing, and agentic AI is at the forefront. Agentic AI enables people and systems to work together more effectively, and Gartner predicts 60% of supply chain disruptions will be resolved without human intervention by 2031 (Gartner, March 18, 2026). Agents have the ability to reason over data, take action across workflows, reduce manual effort, and support faster, more consistent execution—while keeping humans in control of decisions and outcomes.
Frontier firms are moving beyond isolated AI use cases and focusing on how decisions and actions connect and orchestrate across end-to-end processes. This shift it happening in three practical ways:
More proactive risk management: With access to real-time signals, teams can identify potential disruptions earlier and take action before they impact orders, production, or customer commitments.
Faster, more coordinated execution and orchestration: Insights are surfaced directly within the flow of work, so planners and operators can act immediately.
Coordinated human and agent workflows: Agents can take on high-volume tasks across the supply chain—from monitoring conditions to initiating actions—while people stay focused on oversight, exception handling, and strategic decisions.
The result is a more responsive operating model. Let’s see how this approach comes to life in Dynamics 365 Supply Chain Management, and how organizations are using it to improve resilience, responsiveness, and end-to-end performance. Watch this video for a preview on how agentic AI with Dynamics 365 can transform your supply chain operations to a frontier firm.
// Modify player theme based on localStorage value.
let options = {“autoplay”:false,”hideControls”:null,”language”:”en-us”,”loop”:false,”partnerName”:”cloud-blogs”,”poster”:”https://cdn-dynmedia-1.microsoft.com/is/image/microsoftcorp/3797681-Agentic-AI-SupplyChain_tbmnl_en-us?wid=1280″,”title”:””,”sources”:[{“src”:”https://cdn-dynmedia-1.microsoft.com/is/content/microsoftcorp/3797681-Agentic-AI-SupplyChain-0x1080-6439k”,”type”:”video/mp4″,”quality”:”HQ”},{“src”:”https://cdn-dynmedia-1.microsoft.com/is/content/microsoftcorp/3797681-Agentic-AI-SupplyChain-0x720-3266k”,”type”:”video/mp4″,”quality”:”HD”},{“src”:”https://cdn-dynmedia-1.microsoft.com/is/content/microsoftcorp/3797681-Agentic-AI-SupplyChain-0x540-2160k”,”type”:”video/mp4″,”quality”:”SD”},{“src”:”https://cdn-dynmedia-1.microsoft.com/is/content/microsoftcorp/3797681-Agentic-AI-SupplyChain-0x360-958k”,”type”:”video/mp4″,”quality”:”LO”}],”ccFiles”:[{“url”:”https://www.microsoft.com/en-us/dynamics-365/blog/wp-json/bloginabox/v1/get-captions?url=https%3A%2F%2Fwww.microsoft.com%2Fcontent%2Fdam%2Fmicrosoft%2Fbade%2Fvideos%2Fproducts-and-services%2Fen-us%2Fmicrosoft-365%2F3797681-agentic-ai-supplychain%2F3797681-Agentic-AI-SupplyChain_cc_en-us.ttml”,”locale”:”en-us”,”ccType”:”TTML”}],”downloadableFiles”:[{“url”:”https://cdn-dynmedia-1.microsoft.com/is/content/microsoftcorp/3797681-Agentic-AI-SupplyChain_transcript_en-us”,”locale”:”en-us”,”mediaType”:”transcript”},{“url”:”https://cdn-dynmedia-1.microsoft.com/is/content/microsoftcorp/3797681-Agentic-AI-SupplyChain_audio_en-us”,”locale”:”en-us”,”mediaType”:”audio”}]};
if (currentTheme) {
options.playButtonTheme = currentTheme;
}
Enabling this model requires connecting data, decisions, and actions across the supply chain. Dynamics 365 and Microsoft 365 Copilot support this by enabling agents to work across data and business processes in a governed way—linking signals, decisions, and execution within the flow of work.
Several capabilities make this possible:
Connecting agents to data and execution systems: The Dynamics 365 enterprise resource planning (ERP) model context protocol (MCP) Server provides agents with secure, governed access to business processes and data. Using Dynamics 365 ERP MCP Server, agents can take action across the system—such as updating orders, adjusting supply plans, or initiating transfers—based on current conditions. See how Coca-Cola Beverages Africa uses Dynamics 365 and the Dynamics 365 ERP MCP server to autonomously run planning cycles, making the lives and the work easier for their end users.
Grounding agent actions in business context:Work IQ gives agents real-time understanding of how work happens across the organization by connecting the productivity tools where decisions are made, including Outlook, Teams, and Word, to live operational data in Dynamics 365. Agents see the full context of a disruption, including its impact on orders, inventory, and customer commitments, and act on it directly in the system of record.
Agents operating within business policies and constraints: Agents act within defined rules—such as service levels, customer prioritization, and cost thresholds—so decisions remain aligned to business objectives and governance requirements. Additionally, Microsoft Agent 365 gives IT teams one place to observe, govern, and secure agents in your organization.
“Our partnership with Microsoft has been instrumental in enabling Farmlands’ digital transformation. By standardizing on Microsoft Dynamics 365 and thoughtfully applying agentic AI with humans firmly in the loop, we are creating a more efficient, resilient, and scalable operating model.”
Driving impact with agents from planning to delivery
The impact of an agentic approach is best understood through how work changes across core supply chain processes.
Forecast to plan: Continuous alignment of demand and supply
Demand planning is no longer limited to periodic forecast updates. With Demand planning in Dynamics 365 Supply Chain Management, planners can incorporate external signals—such as changes in demand patterns, promotions, or market conditions—alongside operational data to continuously adjust forecasts and supply plans. Planners can focus on evaluating scenarios and making decisions instead of spending time consolidating data.
Source to pay: From supplier signal to business action
Supplier updates often require immediate analysis—what orders are affected, what production is at risk, and what actions need to be taken. When a supplier flags a component delay, the Procurement Agent in Dynamics 365 Supply Chain Management triages the supplier communication, matches it to the affected purchase order, and summarizes the downstream impact across inventory, sales orders, and production schedules. If inventory is available elsewhere in the network, the agent highlights the option. The procurement lead reviews the recommended response and moves forward spending time on decisions and mitigation, not tracing impacts across systems.
Plan to produce and Inventory to deliver: Adjusting production execution, order fulfillment, and warehouse operations in real time
Changes in material availability, labor capacity, or priorities require constant updates to production schedules and warehouse activities. Agents can help apply these changes directly within operational workflows—supporting updates to schedules, work orders, and task prioritization—so teams can keep operations running without delays often caused by manual coordination. Customer commitments depend on accurate, up-to-date information across inventory, production, and logistics. Dynamics 365 Supply Chain Management connects these data points so teams can ensure that order promising reflects current conditions—reducing the risk of overpromising and enabling earlier communication when changes occur.
“With Dynamics 365, we gain real-time visibility into our inventory across all locations, empowering faster, data-driven decision making and enhancing our overall operational agility.”
Erdal Arslan, Financial Affairs Group Manager, LC Waikiki
Order to cash: Enabling a smooth end-to-end customer order experience
Whether it is B2B or B2B2B, agentic commerce can connect customers to the right distribution channel, automate and accelerate the purchase experience from quote-to-order and the availability of accurate data increases speed and agility across the supply chain processes.
Service to deliver:Optimizing asset performance and seamless field operations
AI and agents in field service can provide timely updates on asset and work order information and coordinate field technician schedules that in turn improves overall production uptime, efficiency of delivery and customer satisfaction.
Connect with us at the Gartner Supply Chain Symposium
We invite you to engage with our leaders and explore these innovations firsthand during the symposium. Whether you are looking to learn more about the technology or for a strategic conversation on business outcomes, there are several ways to connect:
Executive Boardroom: Join Sameer Verma, Raghav Jandhyala, and your fellow supply chain leaders for a strategic conversation and best practice sharing session where we’ll discuss how agentic-based approaches can improve supply chain speed, resilience, and business impact. CSCO Boardroom: Win the Volatility Era: Build a Faster, Agentic Supply Chain.
This article is contributed. See the original author and article here.
In contact centers, operational excellence doesn’t start with the first customer interaction—it starts much earlier, during setup, configuration, and ongoing system health management. Yet for many organizations, these steps remain complex, time-consuming, and error-prone.
To solve for these, we are excited to preview Service Operations Agent, a new AI-powered capability in Dynamics 365 Contact Center transforming how administrators configure, validate, and orchestrate conversations at runtime.
Why this matters
Setting up and maintaining a contact center environment often requires navigating multiple tools, understanding complex dependencies, and manually troubleshooting issues. These challenges can result in:
Slow onboarding and delayed time to value
Configuration errors that impact routing, channels setup
Conversations managed by static rules that cannot adapt — ignoring wait time, customer tier, agent availability, or transfer history — leading to poor outcomes that no one is actively watching
Limited visibility into system health and operational readiness
Reliance on specialized expertise for basic administrative tasks
These traditional approaches fragment setup, validation, and diagnostics across multiple surfaces—making it difficult to ensure consistency and operational reliability.
Service Operations Agent addresses these by bringing these capabilities together into a unified, conversational experience.
What’s new
Service Operations Agent introduces an AI-powered administration layer built directly into Dynamics 365 Contact Center, enabling administrators to:
Set up contact center channels and manage users conversationally
Receive recommendations on next steps from agent
Dynamic orchestration of conversations to handle needs of a working contact center during run time through playbooks
Validate configurations and detect issues early when connected to Application Insights
Monitor operational health and troubleshooting guidance
Together, these capabilities give administrators a single place to set up the contact center, keep it healthy, and define how every conversation is handled at runtime — from first contact through resolution.
How it works
Service Operations Agent is available directly within the Dynamics 365 Contact Center Admin Center, where administrators can interact with it using natural language.
The agent guides users through tasks such as:
Setting up channels, queues, workstreams
Managing users and skills
Applying playbooks using provided templates for dynamic orchestration of conversations
Configuration validation and troubleshooting through Application Insights
This creates a more intuitive and streamlined administrative workflow—reducing complexity while maintaining control and governance.
Key capabilities in this preview release
In this preview release, Service Operations Agent focuses on three core areas: setup, diagnostics and conversation orchestration.
Conversational setup and configuration
Administrators can use natural language to provision and configure key components such as queues and channels—reducing reliance on manual configuration flows. This approach streamlines onboarding and helps organizations get up and running faster, even without deep technical expertise.
User prompting agent to create a queue and add users to itAgent response with details about successful queue creation and addition of users, as well as recommended next steps.
Intelligent validation and diagnostics
Service Operations Agent continuously evaluates configurations and surfaces potential issues, helping administrators:
Identify misconfigurations early
Understand root causes of setup issues
Receive actionable recommendations for resolution
Embedding diagnostics into the admin workflow reduces operational risk and improves reliability over time.
Conversation orchestration
Define how conversations are managed throughout their lifecycle, not just at entry. Playbooks let admins set conditions based on customer context, wait time, and queue state, and trigger automatic actions without manual supervisor intervention. They are authored using guided templates in the admin center — no rules to write, no code to maintain. Built-in validation catches conflicts before anything goes live. These capabilities are available now in public preview:
Dynamic prioritization: A playbook continuously monitors wait time and customer context. When conditions are met — for example, a premium customer waiting beyond a threshold with no senior agents available — their priority is automatically elevated. High-value and long-waiting customers rise to the top without anyone watching the queue.
Overflow based on CSR availability: A playbook fires the moment a queue has no eligible agents. Based on customer segment, it transfers to a backup queue, offers a callback, or sends a graceful closing message. Every customer gets a response. None of them wait in silence.
Options for setting up conversation orchestration.
Key benefits
With Service Operations Agent, organizations can:
Accelerate time to value by simplifying setup and onboarding
Reduce configuration errors through built-in validation and guidance
Improve operational reliability with proactive diagnostics
Lower dependency on specialized expertise with conversational experiences and templates for playbooks
Improve customer experience by automatically managing in-queue conversations — reducing wait times, preventing silent overflow, and ensuring high-value customers are prioritized without supervisor intervention
Continuously optimize contact center performance with ongoing insights
Together, these capabilities establish a strong foundation for building, scaling, and managing modern contact centers.
Looking ahead
Service Operations Agent is part of a broader vision to bring agentic experiences for administrators in Dynamics 365 so we can assist them in their jobs to be done. Some of these are enhanced automation, better troubleshooting, simulated conversations for validating behavior and improved insights. For orchestrating conversations, we will provide capabilities to improve the in-queue lifecycle experience, such as re-connecting customers to their last customer service representative, handling overflows for direct-dial lines, out-of-hours queues, etc. Each new capability builds on the same playbook foundation — everything configured today carries forward as the platform grows.
All these additional exciting capabilities are soon to follow!
This article is contributed. See the original author and article here.
In an increasingly connected world, organizations engage customers across diverse cultures, languages, and legal frameworks. A product or service feels most trustworthy when it shows respect for local customs, acknowledges regulatory differences and avoids culturally biased or region-specific references. That’s why we’re excited to introduce Governance for Dynamics 365 Customer Service, now available in Public Preview.
At Public Preview, Governance applies to the email channel — the channel where compliance risk and brand impact tends to be highest. We designed this new, administrator-configured policy enforcement layer to help support compliance with legal and brand standards by validating each outgoing customer email — whether composed by a service representative or generated by AI — against the specific policies your organization defines.
Why Governance? Why Now?
Customer service teams are rapidly adopting AI to streamline operations and enhance customer experiences globally. But with this transformation comes a new set of challenges: How do you ensure that AI-generated or human-authored communications stay aligned with your internal policies and external regulations? Cultural norms vary widely—from naming conventions and date formats to strict data-privacy regulations and content-control requirements. Miscommunications or unintended disclosure of sensitive information can erode trust, while non-compliance with regional regulations can lead to fines and brand damage. Governance is designed specifically for these challenges. It provides a proactive, runtime policy enforcement layer that checks outbound messages —whether written by a customer service representative or generated by AI—against administrator-defined policies. This ensures delivery of only compliant, brand-safe, and policy-aligned content to customers.
Key Capabilities
Governance introduces a suite of powerful features designed to help organizations enforce communication standards at scale:
📝 No-Code Policy Authoring: Administrators can define custom policies in plain language, like “Block any mention of competitor names” or “Require a legal disclaimer in every customer email”. The system automatically translates these into enforceable checks, eliminating the need for any coding or engineering support.
⚡ Real-Time Enforcement: Every outbound email is evaluated before it’s sent. The system flags violations or blocks the message, prompting the user to revise the content — ensuring compliance without disrupting workflows.
🔍 Auditability and Visibility: Governance logs every policy evaluation, providing full transparency to supervisors and compliance teams. It reviews what was checked, what was flagged, why it was flagged, and how it was resolved.
🧪 Simulation Mode: Test policies in a non-blocking “simulation” mode to observe how they perform in real-world scenarios. This helps teams fine-tune policies and build trust before full enforcement.
📦 Out-of-the-Box Policies: Get started quickly with pre-built policies for common compliance needs, including profanity filtering, groundedness checks, and email template adherence.
Built for the Dynamics 365 Ecosystem
Governance and Quality Evaluation are both delivered as skills within the Quality Management Agent in Dynamics 365 Customer Service. As part of the broader Quality Management suite, Governance works alongside the Quality Evaluation skill, which assesses interactions to uphold quality standards. Together, these skills create a unified framework for quality, compliance, and policy adherence.
Learn More
To explore Governance in action and start defining your own policies, learn more here. We look forward to your feedback as we continue to evolve this feature toward General Availability.
This article is contributed. See the original author and article here.
Modern customer experiences don’t live in a single system or interaction type. Whether agents resolve complex cases in Dynamics Customer Service or manage live conversations in Dynamics Contact Center, the quality of an interaction often depends on how the work is performed and not just what was said.
Across both workloads, customers have consistently asked for the same missing capability: secure, structured visibility into the agent’s on‑screen workflow during customer interactions. Voice, chat, and transcripts capture part of the story, but they don’t always reveal how agents navigate systems, apply processes, or respond in real time.
Today, Screen Recording in Dynamics 365 Customer Service and Dynamics 365 Contact Center addresses that gap by providing organizations with trusted, role‑based access to screen context that supports coaching, quality evaluation, and compliance.
Why Screen Context Matters
Customer interactions are rarely linear. Agents often work across multiple systems, reference internal knowledge, update records, and follow prescribed processes, all while engaging with the customer.
Although Customer Service and Contact Center support different interaction models, customers using both workloads face remarkably similar challenges:
Supervisors need better context to evaluate agent’s activities when helping customers.
Compliance teams require auditable visibility in regulated environments.
Organizations want to coach agents more effectively without guesswork.
Without screen context, organizations must infer intent and execution. With it, teams can:
Perform more accurate and objective quality evaluations
Identify training gaps and process inconsistencies
Reduce time to resolution by understanding workflow breakdowns
Validate compliance and required actions with confidence
Screen recording complements existing call and chat records, enabling a more complete understanding of each interaction, regardless of the channel.
One Capability, Designed for Two Workloads
We’ve built Screen Recording in Dynamics 365 as a unified capability that serves both Customer Service and Contact Center, while adapting naturally to different ways agents work.
Key Capabilities:
Automatic Screen Recording for Call Conversations
Manual Screen Recording for service reps working on multiple cases, chats, etc.
All screen recordings are securely uploaded to Dataverse
Role based access to both perform and access screen recordings
Manual screen recording when service reps are working on cases, for example:
To balance visibility with usability, recording is scoped intentionally, and configuration and operational details are managed by administrators.
For step‑by‑step setup and configuration guidance, detailed documentation is available on Microsoft Learn.
Secure by Design: Trust, Access, and Governance
Because screen recordings may contain sensitive information, security and governance were foundational to the design.
Access is controlled through role‑based permissions, ensuring only authorized users can record, review, or download recordings. Recordings are stored securely, with controls that align to enterprise and regulatory expectations.
This approach allows organizations to unlock operational insight without compromising trust between agents, supervisors, and compliance teams.
We’re excited for you to try this capability and to hear how screen recording is helping you improve quality, coaching, and operational visibility across your service experiences.
This article is contributed. See the original author and article here.
Every organization—regardless of size—incurs employee expenses for travel, meals, office supplies, mileage, per diem, and more. When these expenses are managed without a structured, automated process, companies may face higher administrative overhead and compliance challenges, compliance risks, while employees can experience delayed reimbursements, limited transparency, and additional friction.
To address this challenge, we’re excited to introduce Expense Agent, a brand‑new AI‑powered experience coming to Dynamics 365 Business Central.
What is Expense Agent?
Expense Agent in Dynamics 365 Business Central is an AI-powered agent designed to simplify how employees capture, submit, and manage expenses, while ensuring organizations maintain control, compliance, and auditability.
Our north‑star vision is simple: make expense management feel effortless for employees.
Expense Agent helps by automatically collecting, categorizing, and preparing expenses and expense reports, so employees typically only need to review and confirm the results. In many scenarios, the agent significantly reduces manual effort by preparing expenses in advance for user review. Paper receipts may still need to be scanned, but digital receipts, mileage, and allowances can be captured automatically.
Importantly, autonomy is configurable. Organizations decide how much they trust the agent and can tune its level of independence to match internal policies and comfort levels.
A new agent model for a new audience
Expense Agent follows a different design philosophy than existing Business Central agents.
Existing Business Central agents are optimized for users working inside ERP. Expense Agent, however, is built for people who are not ERP users at all—field workers, drivers, technicians, travelers, or occasional office staff. These users shouldn’t need to understand financial systems to submit expenses correctly.
As a result, Expense Agent meets users where they already work:
A dedicated web app for non‑Business Central users
Outlook
Microsoft 365 Copilot Chat (coming next)
But also, in Business Central
Business Central licenses are not required for expense submission if working outside of Business Central. Instead, usage is metered through Copilot Credits, providing flexibility and cost alignment based on actual usage.
Built AI‑first—from day one
Expense Agent is not just an agent layered on top of existing functionality. In fact, we didn’t previously have native expense management in Business Central.
This solution was designed with AI as a core component ground up, using:
Model‑driven and eval‑driven development
Continuous learning and quality validation
Deep integration between the agent and Business Central finance capabilities
The result is an experience where AI is not an add‑on—but the foundation.
How it works
When a user uploads or shares a receipt, Expense Agent automatically:
Extracts required data from the receipt across supported languages
Categorizes expenses based on configured expense categories
Add other relevant information
Optionally itemizes receipts, or add other information, if configured
Applies relevant company policies and rules in the background
Once reviewed, expenses are automatically added to an existing expense report—or a new one is automatically created when appropriate. Users simply submit the report for approval.
Expense Agent also supports:
Mileage, per diem, and other types of allowances
Cash, corporate card, and bank‑paid expenses
Project‑related expenses posted directly to project ledgers
Billable expenses that are automatically reinvoiced
Behind the scenes, the agent continuously validates expenses against regulatory and internal policy requirements—helping organizations support compliance requirements while minimizing additional complexity for users.
Availability and what’s next
Expense Agent is currently available in private preview, where we’re collecting early feedback.
Public Preview planned to start in May 2026, as part of 2026 Release Wave 1
The public preview will initially be available in English only and limited to the United States.
Additional languages and country availability are planned to begin rolling out from July 2026
We’ll showcase Expense Agent publicly for the first time at Directions NA, followed by Directions Asia and DynamicsMinds.
Public preview is not a single milestone—it’s an ongoing journey. We have a strong backlog of enhancements already in progress, with new capabilities planned to roll out regularly throughout the year as we expand language coverage, regional availability, and end‑to‑end scenarios.
Our goal is to be a full end-to‑end expense agent supporting employees, approvers, and finance teams across the entire lifecycle, from capture to posting, with minimal manual effort.
This article is contributed. See the original author and article here.
For years, contact centers have shouldered some of the most persistent challenges in customer experience. They’re expected to help differentiate brands, improve customer retention, surface new revenue opportunities, and reduce frontline burnout — all while managing rising interaction volumes and escalating costs. Artificial intelligence has been positioned as the answer, and investment has accelerated accordingly. Yet despite rapid progress in AI capabilities, many contact centers still struggle to translate that innovation into durable, enterprise-scale impact.
The problem isn’t that AI doesn’t work. It’s that it has been applied in pieces. One solution for self-service. Another for agent assist. Separate tools for quality management, analytics, and administration. Each may deliver value in isolation, but together they create fragmentation. This creates disconnected experiences for customers, service representatives, supervisors, and administrators, and AI programs that stall at pilots rather than scaling across the business.
Addressing the challenge
Microsoft is addressing this challenge by rethinking how AI shows up across the entire contact center lifecycle. Instead of delivering disconnected features, Microsoft Dynamics 365 Contact Center introduces a coordinated agent model. Purpose-built AI agents are designed to work together from day one, each aligned to outcomes across engagement, quality, and operations. This approach directly contrasts with traditional CCaaS platforms. Typically, AI is embedded as isolated capabilities operating only within the boundaries of the contact center application, limiting reuse, shared context, and enterprise scale.
Built on a shared intelligence, data, analytics, and orchestration layer, these three contact center agents are designed to continuously learn from every interaction, whether resolved autonomously or with human support. Together, these agents help form the foundation of a truly modern contact center. Because they are built on Copilot Studio rather than a CCaaS-specific AI stack, Dynamics 365 Contact Center agents are not confined to a single vendor ecosystem. They can participate in broader business workflows with consistent governance, security, and learning across the enterprise.
Customer Assist Agent: Frontline for Every Customer Interaction
The Customer Assist Agent is the primary AI agent responsible for customer self-service across voice and digital channels in Dynamics 365 Contact Center.
The newest capability in Customer Assist Agent is support for real-time voice AI. Real-time AI can listen, reason, and respond naturally with ultra-low latency. It handles interruptions, maintains context across turns, and adapts based on customer intent.
Traditional IVR can become brittle as interactions grow complex, while standalone AI voice solutions often lack enterprise control. Customer Assist Agent in Dynamics 365 Contact Center combines both. It uses deterministic logic for precise, auditable moments like payments and compliance, and generative, real-time reasoning for dynamic, multi-intent conversations. Built in Copilot Studio, real-time AI supports natural speech, interruptions, multi-language switching, and DTMF fallback. Connected to enterprise systems, voice becomes a resolution engine. Accuracy can improve over time as business users tune behavior directly in Copilot Studio using real interaction data.
When inquiries fall outside the bounds of automation or involve high-value or sensitive scenarios, the Customer Assist Agent escalates to a human Customer Service Representative (CSR). Importantly, this is not a handoff requiring customers to start over. The full conversation context, including intent, history, and progress, transfers intact so humans can pick up where it left off.
The Customer Assist Agent goes beyond inbound support by enabling proactive engagement—reaching out to customers on their preferred channels. AI agents use context, intent signals, and orchestration to initiate conversations for reminders, delivery updates, payment notifications, or issue resolution. Engagements range from simple notifications to dynamic, multi-step conversations that adapt based on customer responses across voice and digital channels.
Customer Assist Agent is Generally Available.
Quality Assurance Agent: Insight and Continuous Optimization
As AI becomes deeply embedded in customer interactions, quality management can no longer rely on sampling a small percentage of conversations after the fact. The Quality Assurance Agent supports supervisors in an AI and human collaborative model at scale, evaluating interactions in real time and post-conversation.
The Quality Assurance Agent is purpose-built to evaluate customer interactions with the Customer Assist Agent or customer service representatives at scale, in real time and post conversation. It measures conversation quality indicators such as empathy, tone, and business-defined custom quality criteria. Rather than producing static reports, it monitors for anomalies, emerging issues, and quality drops, surfacing alerts or recommended mitigation steps as needed. This is designed to help supervisors intervene earlier, address problems before they escalate, and maintain consistent service quality across both AI-led and human-led interactions.
Working in an autonomous loop with the Customer Assist Agent, the Quality Assurance Agent helps organizations identify opportunities to improve their self-service workflows and the overall quality of conversations.
Quality Assurance Agent is Generally Available.
Service Operations Agent: Accelerating Setup, Governance, and Scale
The Service Operations Agent is an AI assistant designed for contact center administrators and IT teams, focused on improving setup time, configuration, and governance. With more automation, operational complexity often increases. This means more environments to manage, more rules to govern, and more integrations to maintain. The Service Operations Agent can help administrators address this challenge.
Service Operations Agent helps provision and configure contact center environments rapidly, including trials and new deployments. It can potentially automate everything from initial environment setup to workflows and channels. This reduces manual effort and configuration errors, with the potential to significantly shorten time-to-value for new implementations.
The Service Operations Agent also introduces conversation orchestration. This continuously monitors and adapts customer conversations in real time using natural-language playbooks. It also helps improve queue management with capabilities such as dynamic prioritization of waiting conversations and intelligent overflow based on customer service representative availability.
Service Operations Agent is in Public Preview, available in US only.
“As one of the leading electronics retailers in Greece and Cyprus, we operate at scale across retail and after-sales service. Dynamics 365 helps us deliver a more proactive, conversational, and context-aware experience. With Customer Journeys and proactive engagement in D365 Customer Insights and autonomous conversation orchestration in Service Operations Agent for D365 Contact Center, we can anticipate customer needs, route customers intelligently, and carry interactions across SMS and voice, reducing friction and operational costs. At Kotsovolos, we continuously optimize our customer support to enable better living for everyone, everywhere.”
– Ioannis Papidis, CTO, Kotsovolos
Three Agents, One Copilot
Together, the Customer Assist Agent, Quality Assurance Agent, and Service Operations Agent can form the foundation of an autonomous, AI-powered contact center. Copilot ties them together, providing a single control plane where human leaders can oversee AI actions, guide learning, and maintain accountability. Each agent is priced using Copilot credits, providing flexible, consumption-based economics.
Each agent is priced using Copilot credits, providing flexible, consumption-based economics. Credits are consumed based on AI activity such as conversations handled, real-time assistance provided, summaries generated, and quality evaluations performed, rather than per-seat licensing. This allows organizations to start small, scale usage as adoption grows, and align spend directly to value delivered, while continuing to use the same agents across voice and digital channels. This outcome-based agent architecture is intended to replace disjointed tools and manual processes. It creates a coordinated system spanning engagement, quality, and operations.
The result is a blended workforce where humans and AI operate in concert, each focused on what they do best. AI handles scale, speed, and continuous optimization, while people deliver empathy, judgment, and trust. With Dynamics 365 Contact Center, Microsoft is defining a new model for customer engagement, one where AI is not an add-on but the connective tissue designed to enable faster resolution, smarter operations, and more meaningful customer relationships.
This article is contributed. See the original author and article here.
Manufacturing leaders today are under constant pressure: volatile demand, fragile supply networks, inventory risk, and zero tolerance for missed customer commitments. Plants must protect margins and maintain uptime while responding to disruption with incomplete or delayed information.
To lead in this environment, manufacturers need decision-making that surfaces risk earlier, absorb supply shocks faster, and replans quickly, so production stays aligned to demand and customer commitments.
Microsoft Dynamics 365 is your agentic enterprise resource planning (ERP) foundation
Dynamics 365 gives manufacturers a connected foundation where signals from demand, supply, capacity, and cost can drive action. As part of the Microsoft platform, it brings together data, analytics, automation, and collaboration across the business.
Imagine a manufacturing stack where every part of the business is connected, where data isn’t fragmented across systems but unified into a shared understanding of operations, and where insights don’t sit in dashboards but show up in context, in the flow of work. Where your teams can build AI agents and adapt processes in real time rather than reengineering them months later. Where every person operates with the full context of the business, demand, supply, capacity, and commitments, every decision drives aligned action across the company. This is what it means to close the gap between knowing and doing.
At Hannover Messe 2026, Microsoft is showcasing how agentic ERP with Copilot, agents and Dynamics 365 supports these operational decisions amid constant change, helping manufacturers replan faster, make better production tradeoffs, and meet or exceed customer service SLAs.
From sensing demand shifts to mitigating supplier delays, production constraints, and equipment downtime, these signals must translate into real-time fulfillment and order promise decisions. Agentic ERP can connect planning, sourcing, production, fulfillment and service commerce to execute these decisions across the manufacturing value chain.
Agentic demand planning can sense shifts earlier
Demand planning can help you forecast demand, applying a structured phase in/phase out with external signals to account for market dynamics. Copilot accelerates this data analysis, scenario modeling, and incorporates market intelligence. For structured planning, Demand Planning in Dynamics 365 Supply Chain Management provides an agentic collaborative demand planning solution, powered by best-in-class forecasting algorithms and models, an immersive user experience with AI, intelligent segmentation and signal analysis, and analytics.
“Now, we can see the item-level details and the solution populates Dynamics 365 with that information for master planning. That’s a big step forward because now we essentially have a planned budget and we can easily produce the next production cycle’s budget without manually exporting the data.”
A delayed component or supplier exception is not a paperwork problem. It is a production and operational risk. Today, Microsoft announced the Procurement Agent in Dynamics 365 Supply Chain Management is available in public preview. The Procurement Agent helps teams respond faster to supply disruption by decoding supplier communications and identifying which production orders, inventory positions, and customer commitments are at risk, while keeping humans in control. Beyond triaging supplier communications, the Procurement Agent performs impact analysis so teams can assess downstream effects and act before a supplier issue turns into an on time, in full (OTIF) failure.
“Across the team, Dynamics 365 will save us a total of 20 hours weekly. The big benefit in time savings is that the agent has read the email, found the order, found the lines affected, decoded what the vendor is trying to say, and recommended actions.”
// Modify player theme based on localStorage value.
let options = {“autoplay”:true,”hideControls”:null,”language”:”en-us”,”loop”:true,”partnerName”:”cloud-blogs”,”poster”:”https://cdn-dynmedia-1.microsoft.com/is/image/microsoftcorp/3669965-Procurement-Agent_tbmnl_en-us?wid=1280″,”title”:””,”sources”:[{“src”:”https://cdn-dynmedia-1.microsoft.com/is/content/microsoftcorp/3669965-Procurement-Agent-0x1080-6439k”,”type”:”video/mp4″,”quality”:”HQ”},{“src”:”https://cdn-dynmedia-1.microsoft.com/is/content/microsoftcorp/3669965-Procurement-Agent-0x720-3266k”,”type”:”video/mp4″,”quality”:”HD”},{“src”:”https://cdn-dynmedia-1.microsoft.com/is/content/microsoftcorp/3669965-Procurement-Agent-0x540-2160k”,”type”:”video/mp4″,”quality”:”SD”},{“src”:”https://cdn-dynmedia-1.microsoft.com/is/content/microsoftcorp/3669965-Procurement-Agent-0x360-958k”,”type”:”video/mp4″,”quality”:”LO”}]};
if (currentTheme) {
options.playButtonTheme = currentTheme;
}
For the shop floor, Copilot, connected to Dynamics 365 Supply Chain Management via Dynamics 365 MCP Server (Model Context Protocol), can provide AI-generated insights to schedules, inventory, and orders. Copilot can also surface quality-related alerts and can suggest targeted inspections to prevent recurring recalibration issues enabling supervisors to proactively mitigate resource constraints and shortages before they impact production schedules, triggering downstream inventory repositioning and order promise adjustments across fulfillment and customer commitments.
Dynamics 365 Supply Chain Management provides near real-time visibility to the production floor; with agentic experiences, supervisors can spend less time gathering data and more time on proactive exception management. This translates to higher Overall Equipment Effectiveness (OEE), fewer unplanned stops, and better on-time delivery to customers.
Equipment uptime and asset health are critical inputs to manufacturing performance. Unplanned downtime, delayed service, or poorly coordinated technician schedules can quickly cascade into constrained capacity and missed production targets. The Scheduling Operations Agent in Dynamics 365 Field Service can help maintain equipment availability for constrained production schedules by dynamically coordinating technician assignments based on geography, skills, certification requirements, capacity, travel time, and SLA commitments.
Today, Microsoft announced Researcher in Microsoft 365 Copilot is available in public preview within Dynamics 365 Field Service. With Microsoft 365 Copilot in Dynamics 365 Field Service, manufacturers can use Researcher to quickly view operational information across work orders, asset and service history, and parts availability. Researcher surfaces relevant context from Field Service data and Microsoft 365 signals to help teams understand situations as they arise.
// Modify player theme based on localStorage value.
let options = {“autoplay”:true,”hideControls”:null,”language”:”en-us”,”loop”:true,”partnerName”:”cloud-blogs”,”poster”:”https://cdn-dynmedia-1.microsoft.com/is/image/microsoftcorp/3669964-Field-Service_tbmnl_en-us?wid=1280″,”title”:””,”sources”:[{“src”:”https://cdn-dynmedia-1.microsoft.com/is/content/microsoftcorp/3669964-Field-Service-0x1080-6439k”,”type”:”video/mp4″,”quality”:”HQ”},{“src”:”https://cdn-dynmedia-1.microsoft.com/is/content/microsoftcorp/3669964-Field-Service-0x720-3266k”,”type”:”video/mp4″,”quality”:”HD”},{“src”:”https://cdn-dynmedia-1.microsoft.com/is/content/microsoftcorp/3669964-Field-Service-0x540-2160k”,”type”:”video/mp4″,”quality”:”SD”},{“src”:”https://cdn-dynmedia-1.microsoft.com/is/content/microsoftcorp/3669964-Field-Service-0x360-958k”,”type”:”video/mp4″,”quality”:”LO”}]};
if (currentTheme) {
options.playButtonTheme = currentTheme;
}
“The field technicians understand that AI can be a benefit for them. It’s really about trying to improve their overall experience as well as the customer’s”
Agentic warehouse management can fulfill demand faster
As finished goods hit the warehouse, intelligent inventory on-hand balancing in Dynamics 365 Supply Chain Management warehouse management updates slotting to align demand with pick zones. When inbound put-away happens, workers are directed to the best locations based on the intelligent inventory balancing, keeping the warehouse organized dynamically. On the outbound side, optimized pick routes shorten travel distance. The combined outcome is higher throughput with the same labor—faster fulfillment, lower picking cost.
With Dynamics 365, we trust the system. It’s easy to retrieve information and know it’s correct; people can just drive the forklift over. All the considerations we have for our picking, packing, and customer details, the system is doing it for us.
Agentic B2B Commerce aligns customer orders to production reality
As manufacturers and industrial distributors modernize how they sell, serve, and scale, digital ordering is becoming a core capability of the ERP rather than a standalone channel. Dynamics 365 Commerce enables manufacturers to support complex buying and selling models including business-to-business (B2B), business-to-business-to-business (B2B2B), and multioutlet ordering across direct customers, distributors, and dealer networks.
Agentic B2B commerce can help automate routine ordering, replenishment, and quote-to-order processes while remaining tightly connected to pricing, inventory, warehousing, and production planning. This can allow manufacturers and distributors to accelerate order cycles, strengthen partner collaboration, and scale indirect growth with accuracy and control.
Partner agents to empower manufacturers and solve unmet needs
Frontier Manufacturing Firms can extend ERP embedded‑operational decisioning using Copilot and agents built with the Dynamics 365 ERP MCP Server. Microsoft’s partners are building agents to solve high friction manufacturing workflows, including engineering change impact analysis, product recalls, and configure-price‑quote‑ processes, enabling manufacturers to execute operational decisions across sourcing, production, and fulfillment within the systems they use to run their business.
“With the Experlogix CPQ MCP Server and Dynamics 365 ERP manufacturers can turn Copilot into a revenue agent that understands complex configurations, validates engineering rules and constraints, and generates accurate quotes instantly. This accelerates sales cycles, with many customers reducing time to quote by well over 90%.”
Louis-Philippe Perras, CTO, Experlogix
“Cegeka’s Quality Impact Recall Agent turns disruptive recall events into a controlled, orchestrated process for manufacturers reliant on end-to-end traceability. Running inside Dynamics 365 Supply Chain Management, companies handling batch- or serial-controlled products can identify affected materials across plants and distribution channels in minutes, trigger automated communication flows, and accelerate quality-investigation cycles. This approach supports manufacturers in maintaining continuity, reducing risk, and protecting both customers and brand reputation.”
Stijn Geeroms, VP Business Solutions, Cegeka
“The most expensive moment in manufacturing isn’t when something breaks. It’s when an engineering change gets approved without anyone understanding the impact. Manufacturers don’t need another disconnected AI tool to fix that. They need intelligence embedded where they already work. That’s why we built the Staedean ECM+ Impact Agent natively on Copilot inside Dynamics 365 – your rules, your data, your answer – without asking teams to leave the system they run their business in.”
Michiel Toppers, Chief Product Officer, Staedean
These agents extend ERP-embedded decision execution into engineering, quality, and configure-to-order workflows that directly impact production scheduling and customer commitments.
Become a Frontier Manufacturing Firm
The manufacturers that lead this next era will be the ones who can respond to demand shifts, supplier delays, and equipment constraints by coordinating planning, production, fulfillment, and customer commitments across the value chain in real time.
Microsoft is committed to helping manufacturers make that transition, combining a connected Dynamics 365 foundation with Copilot and agents embedded in everyday workflows plus a partner ecosystem that brings manufacturing expertise to the front line. Join us and become Frontier.
Recent Comments