AI In Microsoft 365 Episodes

AI Agents are one of the biggest shifts in artificial intelligence, moving beyond simple chatbots to software that can reason, plan, make decisions, and perform real work across multiple systems. In this episode of Microsoft Knowledge Nuggets, Mirko Peters explains AI Agents in simple terms, showing how they combine large language models, memory, tools, and automation to complete complex business tasks with minimal human intervention. The episode explores how AI agents differ from traditional AI assistants. Instead of only answering questions, agents can analyze goals, break them into smaller tasks, access enterprise data, call APIs, interact with Microsoft 365 applications, and adapt their actions based on new information. You'll learn the core building blocks of modern agentic AI, including reasoning, planning, tool usage, memory, orchestration, and autonomous execution, as well as where technologies like Microsoft Copilot, Copilot Studio, Azure AI Foundry, Microsoft Fabric, and …

This episode features Alan Buscaglia [MVP], creator of the Gentleman Programming YouTube channel, discussing why enterprise AI success depends on far more than choosing the latest language model. Together, we explore the shift from AI hype to AI Harness Engineering—a practical approach to building AI systems that are secure, governed, observable, and trustworthy. Alan explains why prompt engineering alone is no longer enough. Organizations need an AI harness that manages context, validates outputs, enforces security, integrates with enterprise data, and provides continuous monitoring. These guardrails transform AI from an experimental chatbot into a dependable business capability that teams can confidently use in production. The conversation also covers the importance of trust in AI adoption. Users must know that AI produces reliable answers, protects sensitive information, follows organizational policies, and remains transparent in its decision-making. Alan shares practical gui…

Agent-to-Agent (A2A) Communication is an open protocol that enables AI agents to discover, communicate, and collaborate with one another—regardless of the framework, programming language, or platform they were built on. Instead of creating custom integrations between every agent, A2A defines a common standard for exchanging messages, delegating tasks, sharing capabilities, and coordinating complex workflows across distributed AI systems. In this episode of Microsoft Knowledge Nuggets, Mirko Peters explains Agent-to-Agent (A2A) Communication in simple terms and shows why it's becoming a key building block for enterprise AI. You'll learn how agents discover each other through Agent Cards, exchange structured requests and responses, manage long-running tasks, and collaborate securely across organizational and technology boundaries. The episode also explains how A2A differs from Model Context Protocol (MCP): MCP connects AI models to tools and data, while A2A enables independent AI age…

Microsoft Graph Connectors extend the Microsoft Graph by bringing data from external systems into Microsoft 365, making it searchable and accessible through Microsoft Search, Microsoft Copilot, and other Microsoft 365 experiences. Instead of limiting users to data stored in Microsoft 365, Graph Connectors allow organizations to securely index information from third-party platforms such as Salesforce, ServiceNow, Confluence, Jira, file shares, websites, databases, and custom business applications. In this episode of Microsoft Knowledge Nuggets, Mirko Peters explains Microsoft Graph Connectors in simple terms and shows how they help unify knowledge across your organization. You'll learn how Graph Connectors crawl external data sources, create searchable indexes, apply security trimming, and surface relevant content directly in Microsoft Search and Microsoft Copilot. The episode also explains the architecture behind connectors, indexing pipelines, permissions, and how external content…

The Model Context Protocol (MCP) is an open standard that allows AI applications to securely connect with external tools, APIs, databases, files, and business systems through a common interface. Instead of building custom integrations for every AI model and every application, MCP provides a universal way for AI assistants to discover tools, access data, and perform real-world actions, making it one of the key technologies behind modern AI agents. In this episode of Microsoft Knowledge Nuggets, Mirko Peters explains the Model Context Protocol (MCP) in simple terms and explores why it's becoming the "USB-C for AI." You'll learn how MCP enables AI assistants to move beyond answering questions by interacting with enterprise systems, development tools, cloud platforms, and business applications in a secure and standardized way. The episode covers key concepts including MCP clients, MCP servers, tools, resources, prompts, and the client-server architecture that powers the protocol. It…

Azure AI Search is Microsoft's intelligent search and information retrieval service for building modern search experiences and AI-powered applications. It combines traditional keyword search with vector search, semantic ranking, and AI enrichment, making it a core component for Retrieval-Augmented Generation (RAG), AI agents, and enterprise search solutions. In this episode of Microsoft Knowledge Nuggets, Mirko Peters explains Azure AI Search in simple terms and shows why it has become one of the most important services for generative AI applications. You'll learn how Azure AI Search helps AI models retrieve relevant information from your own documents, databases, and knowledge bases instead of relying only on their pre-trained knowledge. The episode covers key concepts including search indexes, indexers, full-text search, vector search, hybrid search, semantic search, AI enrichment, embeddings, and data connectors. It also explains how Azure AI Search integrates with Azure Open…

Microsoft AI Agent Harness is the runtime layer that transforms a large language model into a capable AI agent. While an LLM can generate text, an agent harness provides everything needed to complete real work—planning tasks, calling tools, managing memory, handling approvals, and maintaining context across long-running workflows. In this episode of Microsoft Knowledge Nuggets, we explain Microsoft AI Agent Harness in plain English and show why it's becoming a core building block for enterprise AI applications. You'll learn what an agent harness actually is, how it differs from an AI model, and why modern AI agents need much more than prompts to operate reliably. We cover key concepts including planning, tool calling, context management, conversation history, memory, safety policies, approval workflows, and autonomous execution. You'll also discover how the Microsoft Agent Framework provides these capabilities out of the box, allowing developers to build robust AI agents without cr…

Microsoft 365 Copilot Agents have the potential to transform the way organizations work—but only when they move beyond simple chat experiences and become deeply integrated into real business processes. In this episode of the M365 FM Podcast, Mirko Peters is joined by Microsoft MVP Steve Corey to explore how organizations can build AI agents that deliver measurable business value rather than simply generating impressive demonstrations. Together, they discuss why successful AI adoption depends on solving business problems, not just deploying new technology. You'll learn how Microsoft 365 Copilot Agents extend the capabilities of Microsoft Copilot by automating multi-step workflows, accessing enterprise knowledge, interacting with Microsoft 365 services, and supporting employees across their daily work. The conversation explains how organizations can combine Microsoft 365, Microsoft Graph, SharePoint, Teams, Power Platform, and Copilot Studio to build intelligent agents that understan…

Microsoft Agent Builder makes it possible to create custom AI agents without writing code, allowing business users to extend Microsoft 365 Copilot with organization-specific knowledge and instructions. In this episode of the M365 FM Podcast, Mirko Peters explains what Microsoft Agent Builder is, how it fits into the Microsoft AI ecosystem, and when it's the right choice compared to more advanced tools like Copilot Studio. The episode demonstrates that creating useful AI agents can be as simple as defining a purpose, connecting trusted knowledge sources, and describing how the agent should behave. You'll learn how Agent Builder enables employees to build assistants for tasks such as answering internal questions, summarizing documents, guiding business processes, and providing department-specific expertise. The episode explains how these agents use company knowledge from Microsoft 365 while respecting existing permissions, ensuring users only receive information they are already auth…

Microsoft Copilot is bringing AI directly into enterprise resource planning, making complex financial and operational processes easier to manage. In this episode of the M365 FM Podcast, Mirko Peters explains how Copilot enhances Microsoft Dynamics 365 Finance & Operations by helping finance, procurement, and operations teams work faster through natural language interactions, intelligent automation, and AI-powered insights. Instead of navigating complex ERP menus, users can simply ask questions, generate reports, and automate repetitive tasks using everyday language. The episode explores practical use cases including financial reporting, invoice processing, forecasting, procurement assistance, workflow guidance, and data analysis. You'll learn how Copilot reduces manual effort, surfaces relevant business information instantly, and helps employees make faster, better-informed decisions. The discussion also highlights how AI integrates with existing business processes rather than repl…

Microsoft Fabric is Microsoft’s unified analytics platform, bringing together data engineering, data integration, data science, data warehousing, and business intelligence into a single cloud service. But where does Copilot fit into the picture? In this episode of Microsoft Knowledge Nuggets, we break down Copilot for Microsoft Fabric in simple, practical terms. You’ll learn how AI helps data professionals and business users generate reports, build data pipelines, write SQL and DAX, create Power BI visuals, and uncover insights using natural language instead of complex code. We also explain how Copilot works across different Fabric workloads, making advanced analytics faster and more accessible for everyone. Using real-world examples, we show how organizations can reduce manual work, accelerate reporting, and empower teams to ask questions about their data without needing deep technical expertise. We also discuss the importance of clean data models, governance, and human validat…

Power BI Copilot brings generative AI directly into Microsoft's business intelligence platform, making it easier for users to analyze data, build reports, and uncover insights using natural language. Instead of manually creating complex visualizations or writing DAX queries from scratch, users can ask questions in plain English and let Copilot generate charts, summaries, measures, and report suggestions. This episode explains how Power BI Copilot works, where it delivers the most value, and why successful AI-powered analytics depend on far more than simply enabling a feature. You'll learn how clean data, well-designed semantic models, governance, and Microsoft Fabric create the foundation that allows Copilot to deliver accurate and trustworthy business insights. The discussion also explores the current limitations of Power BI Copilot, including its reliance on high-quality data, the need for human validation, and scenarios where traditional BI expertise remains essential. Whethe…

Copilot in Dynamics 365 Business Central brings generative AI directly into the ERP system that many organizations use to manage finance, sales, purchasing, inventory, and operations. Instead of navigating complex menus or manually entering large amounts of information, users can ask Copilot for help using natural language and complete everyday business tasks more efficiently. In this episode, you'll learn what Copilot in Business Central is, who it's designed for, and how it improves daily work for finance teams, operations managers, sales staff, and business users. Rather than replacing the ERP system, Copilot acts as an intelligent assistant that helps users complete tasks faster while staying inside familiar Business Central workflows. Using practical, real-world examples, the episode explores how Copilot can generate product descriptions, summarize records, assist with data analysis, suggest information during data entry, answer questions about business data, and help users…

Copilot Cowork represents the next evolution of Microsoft 365 Copilot. Instead of simply answering questions or generating content, it can take ownership of multi-step tasks and carry them out on your behalf across your Microsoft 365 environment. You describe the outcome you want, and Cowork creates a plan, performs the work, and keeps you informed along the way. In this episode, you'll learn what Copilot Cowork is, how it differs from traditional Copilot chat, and why Microsoft describes it as an AI coworker rather than just another chatbot. While standard Copilot responds to prompts, Cowork is designed to execute longer-running tasks that may involve multiple applications, files, emails, meetings, and business processes. Using simple, real-world examples, the episode demonstrates how Cowork can prepare meeting briefings, research customers, draft presentations, organize project documentation, schedule meetings, summarize conversations, coordinate follow-up actions, and create …

Copilot Skills are one of the newest ways to make Microsoft Copilot more capable by giving it reusable expertise instead of relying on long, repetitive prompts. Rather than explaining the same instructions every time, a skill acts like a predefined capability that teaches Copilot how to perform a specific business task consistently. This episode explains Copilot Skills in simple terms, showing how they differ from prompts, plugins, and AI agents. While prompts solve one-off requests, skills package knowledge, instructions, and business logic into reusable building blocks that anyone in an organization can use. They help standardize tasks, reduce prompt engineering, and ensure more predictable results. Through practical examples, you'll learn how organizations can create skills for common scenarios such as writing project updates, generating executive summaries, preparing customer proposals, documenting meetings, answering HR questions, or following company-specific processes. In…

Microsoft Scout is Microsoft's vision for the next generation of AI assistants—moving beyond chatbots to autonomous agents that can actively complete work on your behalf. In this episode, we explain what Microsoft Scout is, why it represents a major shift in Microsoft's AI strategy, and how it differs from Microsoft 365 Copilot. Rather than simply answering prompts, Microsoft Scout can work across your files, Microsoft 365 data, browser, and even your local computer to execute multi-step tasks. From organizing documents and managing emails to running workflows and interacting with business applications, Scout is designed to become an intelligent digital teammate that understands context, remembers preferences, and safely performs actions with your approval. We explore the concept of Microsoft's new "Autopilot" agents, how Scout combines cloud intelligence with desktop capabilities, and why security, permissions, and governance are central to its design. The discussion also cover…

AI agents are rapidly becoming autonomous digital workers, but most organizations still manage them like traditional applications or shared service accounts. In this episode, we explore why that approach no longer works and why the future of enterprise AI depends on giving every agent its own governed identity through Microsoft Entra Agent ID. You'll learn why AI agents require far more than authentication. They need ownership, lifecycle management, least-privilege access, auditability, and clear governance boundaries. The discussion explains how Agent IDs fundamentally change enterprise architecture by allowing organizations to assign permissions, monitor activity, enforce security policies, and retire agents just like human employees—without relying on insecure shared credentials. The episode also examines the structural shift from chat-based AI to autonomous execution systems. As agents begin interacting with Microsoft Graph, SharePoint, Power Platform, and business applicati…

Artificial intelligence is fundamentally changing how organizations design applications with Microsoft Power Platform. In this episode, Tchesco Ayih explores why traditional user interfaces built around forms, menus, and manual workflows are giving way to AI-first experiences where users simply describe what they want to accomplish. Instead of navigating software, people increasingly interact with intelligent agents that understand context, intent, and business processes. The discussion explains how Power Apps, Power Automate, Copilot Studio, AI Builder, and Microsoft Dataverse are evolving into a unified platform for creating intelligent business applications. AI is no longer just an assistant layered on top of existing software—it becomes an active participant capable of generating interfaces, orchestrating workflows, reasoning over enterprise data, and supporting better decision-making. A major focus is the changing role of UX design. Designers must move beyond creating stati…

In this episode of M365.FM, Mirko Peters talks with Microsoft MVP Sara Lagerquist about how Power Apps is evolving beyond traditional forms and screens into AI-powered, enterprise-ready business experiences. Instead of treating AI as a feature that is simply added to an app, Sara explains how organizations should design applications where AI becomes a natural part of the user experience, helping employees make better decisions, automate repetitive work, and interact with business data more intelligently. The conversation explores Microsoft's latest innovations, including Generative Pages, Copilot capabilities, Dataverse integration, and modern Power Apps architecture. Sara shares practical guidance on balancing low-code productivity with enterprise requirements such as governance, security, scalability, and maintainability. Rather than focusing on building visually impressive demos, she emphasizes creating business applications that deliver measurable value and remain manageable as…

Artificial Intelligence is evolving far beyond single prompts and chatbots. In this episode of the M365 FM Podcast, Mirko Peters speaks with Microsoft MVP Reshmee Auckloo about how organizations can design enterprise-grade multi-agent AI solutions using Microsoft Copilot, Copilot Studio, SharePoint, Microsoft Graph, and Azure AI Foundry. Rather than focusing on prompt engineering, the conversation explores the architecture required to build AI systems that are scalable, secure, and trustworthy. Reshmee explains why successful AI initiatives begin with well-governed data, structured knowledge, and clear information architecture. SharePoint plays a central role as the enterprise knowledge foundation, while Microsoft Graph provides the contextual intelligence that enables AI agents to understand relationships across people, documents, and business processes. The discussion also covers how multiple AI agents can collaborate to solve complex tasks, when orchestration is preferable to a …

Most organizations think they're building AI agents, but in reality they're still creating advanced chatbots. This episode explains why the future of Microsoft AI lies beyond conversational interfaces and focuses on Graph-powered, enterprise-grade agents that can understand business context, execute actions, and collaborate across multiple systems. The discussion explores how Microsoft Graph serves as the operational backbone for intelligent agents by providing secure access to Microsoft 365 data, identities, files, Teams, Outlook, SharePoint, and business applications. Rather than relying on prompts alone, successful agents combine Graph, semantic context, orchestration, memory, and governance to deliver reliable business outcomes. Listeners learn why enterprise AI requires architectural thinking instead of chatbot design. Topics include agent orchestration, delegated versus application permissions, event-driven workflows, Microsoft Graph APIs, knowledge retrieval, security bou…

Traditional cybersecurity often focuses on vulnerability counts, alerts, and compliance dashboards. Attackers don’t. They look for weak identities, excessive permissions, exposed cloud resources, forgotten devices, and misconfigurations they can combine into a successful attack. In this episode, Mirko Peters is joined by Microsoft Security MVP and MCT Uros Babic to explore cybersecurity from an attacker’s perspective and explain how Microsoft Security Exposure Management helps organizations stay ahead of modern threats. Rather than discussing security in theory, the conversation walks through a realistic attack chain—from reconnaissance and phishing to credential theft, privilege escalation, lateral movement, ransomware, and data exfiltration. Uros demonstrates how Microsoft Defender XDR, Microsoft Sentinel, Security Copilot, Microsoft Entra ID, and Zero Trust principles work together to detect, prevent, and disrupt attacks before they reach critical assets. A major focus is Mic…

Most organizations believe they are building AI agents, but in reality they are still creating advanced chatbots. In this episode, we explore why the next generation of Microsoft AI is no longer about individual bots, but about intelligent runtimes that coordinate specialized agents across business processes. You'll learn why a single chatbot cannot effectively manage complex enterprise workflows and how agent runtimes provide the orchestration, memory, governance, and lifecycle management needed for production-ready AI systems. Instead of simply responding to prompts, agents can collaborate, maintain context, trigger actions, and work together to complete end-to-end business tasks. The episode also explains the architectural shift from isolated conversational interfaces to connected agent ecosystems powered by Microsoft technologies such as Copilot Studio, Azure AI Foundry, Microsoft Graph, and Dataverse. Topics include agent communication, orchestration, long-term memory, secu…

This episode explores why Computer-Using Agents (CUA) could fundamentally change the future of enterprise software. Instead of humans navigating dashboards, menus, and forms, AI agents will increasingly interact with applications on our behalf. The user interface becomes optional, while AI focuses on achieving business outcomes rather than simply responding to clicks. The discussion explains how traditional SaaS applications were designed around human interaction, but agentic AI shifts the focus toward APIs, automation, and orchestration. Rather than opening multiple business applications, users will simply describe what they want, and AI agents will coordinate the required systems, execute workflows, and return the final result. The episode also examines why this transformation challenges existing SaaS vendors. Products will compete less on beautiful interfaces and more on the quality of their data, business logic, integrations, governance, and security. Companies that expose t…