Tuesday 9:00 AM - 12:30 PM · Room 447-448

AI That Actually Works: How to Get More from Copilot

Most organizations already have Copilot and AI tools enabled. Very few are getting consistent, repeatable value from them.

This half-day workshop focuses on how people actually use AI day-to-day—not demos, not hype, and not “prompt tricks” that fall apart outside a sandbox.

Participants will learn how to:

• Work with Copilot as a collaborator, not a search engine
• Structure questions so AI produces usable, trustworthy outputs
• Recognize when Copilot is helping and when it’s quietly misleading
• Apply AI effectively across documents, data, meetings, and decisions
• Build personal and team habits that compound AI value over time

The session is practical, non-technical, and grounded in real workplace scenarios using Microsoft Copilot and general-purpose AI tools. Attendees leave with repeatable patterns they can immediately apply, not just ideas they’ve heard before.

Who This Is For

• Business professionals already using Copilot “a bit”
• Managers and team leads responsible for output quality
• Analysts, knowledge workers, and decision-makers
• Anyone frustrated that AI should be helping more than it is

No technical background required.

What Makes This Different

• This is not a feature tour
• This is not a prompt library
• This is about changing how work gets done with AI

Agenda (Subject to change as product changes happen)

1. Why Copilot Feels Underwhelming (But Isn’t)
2. Asking Better Questions Is a Skill (Not a Trick)
3. Copilot in Real Workflows (Not Demos)
4. Trust, Accuracy, and Oversight
5. From Individual Use to Team Impact
6. The One-Week Reset

Tuesday 9:00 AM - 5:00 PM · Room 421-422

Extend Microsoft 365 Copilot with Custom Agents: Copilot Studio and the Agents Toolkit Side-by-Side

Microsoft 365 Copilot knows what's in Microsoft Graph: your emails, Teams chats, and some SharePoint files. But the data that actually runs your business, your SOPs, CRM records, inventory systems, and line-of-business applications, lives outside that boundary. Extending Copilot to reach that data is where agents come in.

The challenge? Microsoft offers two primary tools for building agents, Copilot Studio and the Microsoft 365 Agents Toolkit (ATK) for VS Code, and their documentation rarely gives you a clear side-by-side comparison. Which one fits your team? Which one fits your scenario? What are the real tradeoffs?

In this full-day workshop, Microsoft MVPs Mark Rackley and Andrew Connell answer those questions by building the same agent with both tools, live, throughout the day. You'll see every decision point, every capability difference, and every tradeoff as it happens. This isn't a pitch for one tool over the other. It's an honest comparison designed to help you make the right call for your organization.

We'll start with how Microsoft 365 Copilot actually works and what customization options exist. Then Mark and Andrew will work through the same scenarios in parallel: adding custom instructions, connecting knowledge sources, integrating external data through actions and connectors, and extending Copilot's capabilities for real business use cases. We'll also cover the practical realities that matter for planning: licensing requirements, cost implications, and the skills your team needs for each approach.

This workshop is designed for a broad audience. Whether you're a developer evaluating pro-code tooling, a power user building agents in Copilot Studio, or a project manager charting your organization's Copilot extensibility strategy, you'll walk away knowing what's possible, what each option requires, and which approach fits your situation.

Tuesday 1:30 PM - 5:00 PM · Room 447-448

Microsoft 365 Copilot for Work: How to Get Real ROI with Better Prompts and Workflows

Everyone’s talking about Microsoft Copilot, but most organizations are struggling to get real value from it. In this full-day, highly interactive workshop, you’ll learn how to move beyond “trying Copilot out” and start using it as a true productivity partner.

We’ll cover: • Where Copilot fits into M365 chat, Word, Excel, Outlook, PowerPoint, Teams, and SharePoint • The five-part framework for writing strong prompts that deliver better results • The most common mistakes users make with Copilot - and how to fix them • Five core skills every professional should master (summarizing, transforming, analyzing, creating, and planning) • How to map a real workflow in your job and build prompts around it • Simple governance rules to stay safe and compliant

By the end of the day, you’ll leave with five reusable prompts, a mapped workflow you can run immediately, and a practical governance checklist - plus the confidence to use Copilot effectively in your daily work.

Wednesday 8:30 AM - 9:40 AM · Ballroom 2

Keynote : Building What’s Next for Work

You’ve been using and building with AI. Now let’s talk about what’s actually possible at scale. In the opening keynote, Microsoft CVP Jason Moore takes a hands-on look at how business users, AI leaders and makers are rewriting the rules of work — improving personal productivity, transforming business systems, and orchestrating complex processes across people, applications, data, and agents. The Microsoft 365 community has always led in delivering value from productivity tools, and with AI now woven into every layer of the stack, what you can build and how fast you can build it has changed fundamentally. Live demos and real customer stories bring it all to life: how Microsoft 365 Copilot, Copilot Cowork, Power Platform, and Copilot Studio come together to put serious capability in the hands of makers at every level. And because none of it matters without trust, you’ll also see how governance keeps it all running at enterprise scale. If you’re building with AI and wondering how far it can go, this is where you find out.

Wednesday 10:20 AM - 11:30 AM · Room 423-424

How to Avoid AI Disasters: ChatGPT, Data Security, and Unintended Consequences

Say your company rushes in the coming months to implement a private version of ChatGPT over their Intranet. An employee asks, "Hey ChatGPT, is my company going to have a layoff?" How will it answer?

This scenario becomes plausible as ChatGPT / GPT-5.3 can be connected to data sources for Q&A activities. If the data related to layoffs is accessible by the AI, it's possible that an employee could get an answer to their question. You should take great care before connecting it to your own internal data.

Unfortunately hearing ChatGPT, AI and Machine Learning terms thrown about like doubloons at a Mardi Gras parade in the news is giving many organizations FOMO. We'll discuss what you should know before you jump in.

In this session, we'll address:

• What are these technologies and how they will impact you
• How can these technologies be used in your organization
• Data and security considerations
• Preparing for unintended consequences

Join us to learn about this brave new world and how you could apply it in your own work.

Wednesday 10:20 AM - 11:30 AM · Room 320-321

Lessons from the Field: Building AI-Ready Microsoft Data Estates

Many organizations are already using Microsoft Fabric, Power BI, SQL Server, Azure data services, and cloud data platforms. The challenge now is not whether these tools can work. It is how to turn a fragmented, fast-changing data estate into a trusted foundation for analytics, automation, Copilot, and AI agents.

Across client engagements, we see organizations moving toward the same desired end state: governed data, reliable pipelines, trusted semantic models, clearer ownership, production-ready operations, and business teams that can act on the information in front of them.

The path there is rarely clean. It often involves legacy SQL Server environments, Power BI sprawl, inconsistent definitions, disconnected source systems, migration decisions, platform overlap, and governance models that have not caught up with AI.

In this session, we will share lessons from the field on what actually breaks, what teams should fix first, and how technical teams can move from platform adoption to durable business value.

You will learn:

• How organizations are rationalizing SQL Server, Fabric, Power BI, and cloud data platforms into a more coherent data estate • What separates initial Fabric adoption from a production-ready operating model • Why trusted semantic models, data quality, and ownership matter more as Copilot and AI agents enter the picture • Where modernization efforts commonly break down across migrations, integrations, governance, and operational support • How AI changes the stakes by exposing weak definitions, poor access controls, and fragile data foundations • Practical lessons for moving from fragmented reporting and platform pilots toward governed, AI-ready analytics

This session is designed for data architects, DBAs, Fabric and Power BI practitioners, analytics leaders, and technical decision makers responsible for modernizing Microsoft data environments while preparing for AI-driven work.

Wednesday 11:40 AM - 12:50 PM · Room 335-336

AI Driven Data Engineering with Spec Driven Development

Explore how Spec-Driven Development is transforming data engineering by making specifications the foundation of the Software Development Life Cycle (SDLC). This session demonstrates how business requirements, user stories, architecture decisions, and acceptance criteria can be captured as structured specifications that guide development from concept to production.

Through a customer case study and live demo, attendees will see how specifications drive solution design, AI-assisted code generation, testing, documentation, deployment, and governance. The session will also showcase how Harness enables orchestration of AI-powered development workflows and how QE/Eval frameworks establish automated testing, validation, and quality guardrails for AI-generated outputs.

Participants will gain practical insights into improving engineering productivity, accelerating delivery, reducing risk, and ensuring enterprise-grade quality through specification-first, AI-driven development practices.

Wednesday 1:20 PM - 1:40 PM · Solution Studio

How AI-Driven Master Data Management with Microsoft Fabric Transforms Business Data & Use Cases

Organizations often struggle with duplicated, inconsistent master data spread across ERP, CRM, and legacy systems. This session explains how AI-driven Master Data Management (MDM)—built on Microsoft Fabric and Azure OpenAI—helps businesses create a trusted, governed single source of truth at scale.

Attendees will learn how modern MDM unifies and standardizes critical business entities such as customers, vendors, and products, ensuring analytics, reporting, and operations are powered by reliable data. Instead of manual data cleanup, AI continuously monitors data quality, detects anomalies, identifies duplicates, and recommends corrections using intelligent confidence scoring.

The session provides a business-level view of how the solution works: Microsoft Fabric consolidates data across systems; AI-powered match-and-merge capabilities resolve duplicate records; Azure OpenAI enriches metadata with clear, business-friendly descriptions; and configurable approval workflows ensure all changes are reviewed and governed.

Security and compliance are built in through role-based access controls, full audit trails, external address validation, and governance workflows aligned with enterprise standards.

Business leaders will see how AI-enabled MDM reduces operational risk, improves compliance, accelerates decision-making, and transforms master data into a scalable, enterprise-grade asset—without requiring users to be technical experts.

Wednesday 1:20 PM - 1:40 PM · Collab Studio

How a K-12 District Built a Trusted, AI-Ready Data Foundation: 90% Fewer Reporting Errors, Lean Team

Fragmented source systems, manual reporting bottlenecks, and data that nobody quite trusts: these aren't education problems, they're enterprise problems.

In this 20-minute session, Monterey Peninsula Unified School District shares how a lean two-developer team consolidated siloed systems, automated pipeline development on Microsoft SQL Server, and reduced reporting errors by more than 90%; all while building a governed foundation ready for AI.

The technical lessons translate directly to any organization wrestling with resource constraints or the challenge of establishing data trust at scale, and you'll leave with practical takeaways on accelerating pipeline development, improving data reliability, and architecting infrastructure that's audit-ready today and AI-ready tomorrow.

Wednesday 1:50 PM - 3:00 PM · Room 440-442

Build Your First Agent with Microsoft 365 Agents Toolkit in 70 Minutes

What A fully hands-on 70-minute lab where every participant builds and deploys a real IT Service Desk Agent using the brand-new Microsoft 365 Agents Toolkit. Your agent will read support requests from Teams/Outlook, check ticket status, reset passwords, create Planner tasks, and reply intelligently — all running natively inside Microsoft 365. How

Step-by-step guided experience (no coding required) Start from the official M365 Agents Toolkit template Connect to real M365 services via Microsoft Graph Test live in the Agents Playground Deploy with one click to your own tenant (Teams + Outlook) All assets, prompts, and cheat sheets provided

Why The M365 Agents Toolkit is the fastest, most secure, and most governed way to create custom agents that live inside Microsoft 365. Perfect for citizen developers and pros who want production-ready agents without leaving the M365 ecosystem or fighting DLP policies. Who

Citizen developers & Power Platform makers IT pros & service desk teams Anyone curious about the new wave of M365-native agents (Level 100–200 – zero AI experience required)

Key Takeaways – You will walk away with

Your own fully deployed IT Service Desk Agent running in your tenant Ready-to-use template pack + prompt library Clear decision framework: M365 Agents Toolkit vs. Copilot Studio Confidence to build and ship real business agents on Monday morning

Laptops required – bring your Microsoft 365 tenant (trial accounts work perfectly). Limited seats – come build the future! 🚀

Wednesday 1:50 PM - 3:00 PM · Room 423-424

Smart Governance for Smarter AI: Managing Data Effectively in the Copilot Era

Data governance has always mattered – but with Microsoft Copilot now woven into the fabric of Microsoft 365, managing your content has never been more critical. In this session, we’ll explore how governing data strategically enables AI to deliver more relevant, secure and meaningful results.

We’ll share practical strategies and lessons learned in managing ROT (redundant, obsolete, trivial data), mitigating risks around overly shared sensitive content, and creating lifecycle policies that balance retention needs with AI readiness. You’ll learn how tools in Microsoft Purview and SharePoint Advanced Management can be used to classify, protect, and streamline enterprise content across varied use cases.

Finally, we’ll look at how effective governance not only supports compliance but also empowers Copilot to surface the right knowledge at the right time – transforming how users find, use, and trust information.

Wednesday 1:50 PM - 3:00 PM · Room 431-432

Top 10 Features You Need to Know About SharePoint AI + Agents

SharePoint AI, also known as SharePoint Premium, is a robust framework that helps you build AI-infused solutions and automate your content-heavy processes. It’s all about helping you extract critical information from your business documents and turning it into actionable insights for your entire organization. With the latest innovations—including AI Agents, OCR, eSignature, AI Models, and Autofill—you can get more value out of your content and manage it more efficiently than ever before. But that’s not all. Join my session to see how SharePoint AI harnesses cutting-edge intelligence to transform document processing across your organization. This is a fully demonstration-driven session where you’ll quickly learn how to enhance your Enterprise Content Management (ECM) and navigate the full spectrum of SharePoint’s AI capabilities. If you're ready to embrace the future of content management, this is one session you won't want to miss!

Wednesday 1:50 PM - 3:00 PM · Room 335-336

You Can't AI Your Way Out of Bad Data: A Practitioner's Blueprint for AI-Ready Data on Azure and SQL

Every organization wants to build AI-powered analytics. Most are discovering that AI doesn't fix bad data, it scales it. The real blocker isn't the model, the GPU, or the orchestration layer. It's the warehouse underneath.

In this session, Kevin Marshbank,Principal Consultant at The Data Vault Shop and a data warehouse architect with over 15 years building enterprise data environments, will walk through what it actually takes to structure a data environment that AI can work with. Drawing on hands-on implementations across Azure Synapse, Microsoft Fabric, and SQL Server, Kevin will break down the architectural decisions that determine whether your AI initiatives succeed or stall; regardless of the modeling methodology you've chosen.

This is not a session about picking the right data model. It's about the properties that any well-built warehouse must have to serve AI reliably: full historicity, traceable lineage, rich metadata, enforced governance, and semantic clarity. Kevin will show how these properties are achieved in practice, and why they're harder to retrofit than to build in from the start.

A central theme of the session is metadata, not as a documentation afterthought, but as the architectural connective tissue that makes AI workloads possible at scale. Kevin will cover how metadata-driven development approaches, which he applies in production using WhereScape's automation tooling on Azure, allow teams to generate pipelines, lineage, governance artifacts, and documentation from a single design layer, reducing manual coding by over 90% while producing the kind of governed, traceable data structures that AI systems actually need.

Attendees will walk away able to:

1. Audit their own data architecture against concrete AI-readiness criteria
2. Understand the metadata and governance gaps most likely to derail AI initiatives in production
3. Evaluate automation approaches that reduce manual development burden while building AI-ready structure into the warehouse from the ground up
4. Apply practical patterns Kevin has used across real Azure-based implementations to accelerate AI readiness without starting over

Wednesday 1:50 PM - 3:00 PM · Room 333-334

Start your adoption journey with adoption.microsoft.com

Don’t start from scratch. Discover the resources, frameworks, and proven guidance available on adoption.microsoft.com to help you successfully drive the adoption of Microsoft 365, Copilot, and AI solutions in your org. Learn where to begin, what resources to use, and how to build a sustainable adoption strategy that drives lasting value for your organization.

Wednesday 3:10 PM - 3:30 PM · Insight Studio

Accelerating Migration to Microsoft Fabric with MigrateFAST and Microsoft Funding Programs

Modernizing analytics platforms can be complex, costly, and time-consuming. In this 20-minute session, discover how MigrateFAST, MAQ Software’s Fabric migration accelerator, helps organizations accelerate their journey to Microsoft Fabric through an automation-first approach.

Through a customer case study and live demo, attendees will see how MigrateFAST streamlines migration across three critical modernization paths:

Data Integration – Modernizing ETL/ELT pipelines and data movement workloads

Analytics – Migrating data engineering, data science, and advanced analytics workloads

Reporting – Transitioning BI and reporting assets to a unified Fabric ecosystem

The session will highlight how organizations can reduce migration effort, improve accuracy, minimize risk, and accelerate time-to-value while adopting a scalable, future-ready data platform on Microsoft Fabric and leverage Microsoft funding programs to reduce migration costs.

Wednesday 3:50 PM - 5:00 PM · Room 423-424

MCP Fundamentals: The What, Why, How, and the OMG Impact

The Model Context Protocol (MCP) is reshaping how we extend AI systems, and this session breaks down everything you need to know to get started. Join Andrew Hess, as he will explore the core fundamentals of MCP. What it is, how it works, and why this new standard creates powerful benefits for builders, developers, and organizations adopting AI.

You’ll learn why MCP came into existence, the challenges it solves, and how it provides a unified, secure, and scalable way for models to interact with tools, data, and systems. We’ll also compare MCP vs. the connectors you’re already using in Power Platform and Microsoft 365, clarifying where each shines and how MCP changes the future of extensibility across the AI ecosystem.

To make it real, we’ll walk through live examples of existing MCPs, demonstrating how they work, what they unlock, and the “OMG” moments that make this protocol such a game changer.

Whether you’re new to Model Context Protocol or looking to elevate your Copilot strategy, this session will give you the understanding and excitement to confidently put MCP to work.

Wednesday 3:50 PM - 5:00 PM · Room 431-432

Make Content Copilot‑Ready: From ROT to ART with Purview, PowerShell & SharePoint

Copilot success isn’t a license problem—it’s a content readiness problem. In this fast, real‑world session, we’ll show you the same patterns we used to script and build our own multi‑tenant content governance system for Copilot: how to discover what Copilot can “see,” reduce oversharing surprises, and apply guardrails that scale—using tools you already have: SharePoint, Microsoft Purview, and PowerShell.

We'll show you how we built our own internal app to prepare and manage your content for Copilot - and other AI systems. Why we built it, how we built it, and how to make sure your content stays high-quality AFTER rollout too.

You’ll learn how to move your content from ROT (redundant, obsolete, trivial) to ART (accurate, relevant, trusted)—so Copilot produces better answers, users trust what they get back, and the business sees measurable value.

And you'll leave with free tools to help jumpstart your own content readiness program.

We’ll cover practical readiness moves like policy‑driven controls that automatically adjust as sites change (including governance approaches that shape what Copilot can and cannot access), plus simple scripting strategies to inventory risk signals and prioritize fixes—without paralyzing everyday collaboration.

Wednesday 3:50 PM - 5:00 PM · Room 435-436

Getting Your Microsoft 365 House in Order Before Copilot Moves In

Copilot doesn't invent new access. It works with what your environment already allows, which means permissions gaps, data sprawl, and unclear ownership all shape what AI surfaces and how much you can trust the output. Most organizations jump straight to enablement and spend weeks or months cleaning up what Copilot exposes. This session flips that sequence.

We'll walk through the governance foundations Copilot depends on, show you how to assess readiness using Microsoft-native tools like Purview, Entra, and SharePoint, and give you a practical framework for building trust in Copilot outputs before you roll it out broadly.

• How to identify the governance gaps Copilot will expose
• Using Purview, Entra, and SharePoint to close them before enablement
• A phased readiness framework that scales with your rollout

Wednesday 3:50 PM - 5:00 PM · Room 335-336

Copilot Cowork and the Rise of Personal Agents

Copilot Cowork changes how we work with AI. Instead of writing one-off prompts, you build a personal library of skills that can fetch, read, summarize, translate, and act across multiple steps. It's the difference between asking for help and having help ready to go.

In this session, we'll cover what Copilot Cowork is, how it uses Markdown-based skills stored in OneDrive, and why this points to a bigger shift toward personal agents that work alongside you. You'll see live demos of real skills in action, including multi-step workflows that go beyond what any single prompt can do.

Whether you're new to agentic tools or already deep in the Copilot ecosystem, you'll leave with a clear understanding of where this is headed and ideas you can put to work for you.

Thursday 9:50 AM - 10:10 AM · Collab Studio

When AI Inherits Everything: Closing the Oversharing Gap in Microsoft 365

AI does not break your permissions. It reads them faster than any user ever could. The day you switch on AI in Microsoft 365, every legacy "Everyone except external users" link, every broken inheritance, and every orphaned site becomes a one-prompt path to data nobody meant to share. This session is for the admins who own that cleanup.

We explain how agents gain access to content through the oversharing that happens natively in SharePoint and Microsoft 365, then show how Orchestry surfaces what you cannot see from the admin center, remediates sharing links, external access, and broken inheritance across thousands of workspaces in bulk, and governs agent inventory and lifecycle so the gap stays closed.

You leave with a clear picture of how Orchestry takes a tenant from discovery to remediation to ongoing control.

Thursday 10:20 AM - 11:30 AM · Room 421-422

10 Microsoft Copilot Scenarios That Deliver Real ROI for Your Information Workers

Many of your employees already have access to the premium Microsoft Copilot across Word, Excel, PowerPoint, Teams, and Outlook and All of your users have access to the free Microsoft 365 Copilot chat.

However, most people aren’t using it in ways that really move the needle and get ROI. Leaders want measurable productivity lifts, but often feels invisible. The challenge isn’t the technology - it’s helping employees understand what’s in it for them and how Copilot can directly improve their daily flow of work.

This session shows you 10 practical, real-world scenarios that drive actual results inside the Microsoft 365 apps your teams already live in. You’ll see examples that shorten report creation in Word, automate analysis in Excel, refine presentations in PowerPoint, summarize meetings in Teams, manage communication overload in Outlook, and help your sales team close deals faster in Dynamics 365 CRM. Beyond the features, we’ll focus on adoption tactics, success metrics, and the leadership actions that turn Copilot from shelfware into impact.

Thursday 10:20 AM - 11:30 AM · Room 343-344

Making AI Agents Smarter with Fabric IQ, Foundry IQ, and Work IQ

Connecting Copilots and Microsoft Foundry AI agents to company data across SharePoint, databases, and documents is complex. Foundry IQ + Fabric IQ + Work IQ provide secure, unified access through one API. Learn to connect agents to multiple sources, configure permissions, use Agent Factory for deployment, and ensure agents understand business context, not just raw data.

Thursday 11:40 AM - 12:50 PM · Room 423-424

Beyond the Prompt: Power User Strategies for Copilot Success

Copilot isn’t just another AI. It’s a productivity partner built to amplify human potential. In this hands-on session, we’ll explore how Copilot differs from traditional AI systems and how power users can unlock its full potential. Drawing from real-world deployments, we’ll walk through advanced prompting techniques that go far beyond simple queries.

You’ll learn how to: • Use multistep reasoning to guide Copilot through complex workflows • Generate graphics, presentations, and documents with precision and flair • Apply vibe coding to shape tone, style, and emotional impact • Build reusable prompt patterns for repeatable business outcomes • Understand Copilot’s contextual intelligence and how it adapts to your work • Design and deploy your own agents to automate tasks, personalize workflows, and extend Copilot’s reach across business processes

I helped launch Copilot while I was at Microsoft, and I help organizations around the world maximize the value of their AI investment. Whether you're in marketing, operations, or IT, this session will equip you with the tools to become a Copilot power user—ready to teach, lead, and scale AI adoption across your organization.

Thursday 2:00 PM - 3:10 PM · Room 421-422

Governing Data in the Age of Copilot

As organizations rush to deploy Microsoft Copilot, the most significant barrier to success isn't the AI, it's the data. Without robust governance, over-sharing becomes an immediate enterprise risk. This session explores the critical intersection of traditional data security and the new world of Generative AI. We will discuss how to move beyond basic permissions to a "Zero Trust" data posture.

In this session, we’ll explore the critical intersection of traditional data security principles and the new world of Generative AI Generative AI. Attendees will learn why legacy approaches to permissions and access control are no longer sufficient and why adopting a "Zero Trust" data posture is essential for safeguarding sensitive information in the age of Copilot. Through real-world examples and best practices, participants will leave equipped with the knowledge to bridge the governance gap, mitigate risk, and unlock the full value of AI while keeping enterprise data secure.

Thursday 4:00 PM - 5:10 PM · Room 423-424

Creating a Copilot Training Center using Microsoft 365 Learning Pathways

Copilot has the potential to be a powerful tool to help users be more productive but not without training. In this session we will learning how to use Microsoft 365 Learning Pathways to create a Copilot training center to help your users learn how to use Copilot effectively. We will show how to install Learning Pathways and surface that content in Microsoft Teams. We will look at the content provided by Microsoft and learn how to augment that content with community-based content as well as your own custom Copilot prompt suggestions specific for your organization.

Thursday 4:00 PM - 5:10 PM · Room 335-336

Stop Feeding AI Your Corporate Junk Drawer

Your storage bill has a dirty secret, and your AI has a stale-content problem. They're connected, and that connection might just be your biggest opportunity.

Your M365 storage quota didn't die in one dramatic moment. Slowly, over time it was overrun by abandoned workspaces, sites that outlived their project by three years, content that multiplied faster than anyone could govern it. And underneath all of it, version history quietly doing its thing; every draft, every minor edit, every "just in case" save still sitting there, stacking up against your quota, unmanaged and unloved. Most organizations have no idea how bad it is until the bill arrives.

The AI problem is different. Copilot and other AI tools are grounding on whatever's still sitting in active storage. Inactive workspaces, orphaned sites, outdated content nobody's reviewed in years. If it's active, it's fair game.

M365 Archive solves both. Move stale content out of active storage and out of your AI's grounding scope in one move. And because not every problem can be solved at the site level, file-level archival gives admins the precision to go after exactly what's causing the bloat without disrupting the workspaces still doing real work.

Two problems. One strategy. Zero excuses for that storage bill.

Thursday 4:00 PM - 5:10 PM · Room 333-334

Copilot, Cowork, or Scout? Choosing the Right AI tool for Every Task

Microsoft is rapidly expanding the AI toolkit available to organizations, but knowing which tool to use and when can be just as important as knowing how to use them. In this session, we'll demystify the roles of Microsoft Copilot, Copilot Cowork, and Scout, helping IT leaders, Microsoft 365 administrators, and business decision makers understand where each solution delivers the greatest value.

Through real-world scenarios, you'll learn how to match the right AI experience to common workplace needs—from personal productivity and content creation to deep research, knowledge discovery, and organizational decision-making. We'll explore the strengths, limitations, governance considerations, licensing implications, and adoption strategies for each tool so you can confidently guide your organization’s AI journey.

Whether you're planning your AI roadmap, enabling users, or driving business outcomes, you'll leave with a practical framework for deciding when to use Copilot, when Cowork is the better choice, and when Scout can unlock insights that other tools can't.

Friday 9:00 AM - 10:10 AM · Room 421-422

Copilot in SharePoint: A Practical Guide to What’s Possible

This is a demo-heavy session for you to discover how Copilot in SharePoint can helps create, improve, and manage content. We’ll focus on practical, everyday ways to use Copilot across SharePoint: from generating pages and summaries in and outside of SharePoint, to using Copilot inside web parts like the Text web part, to tapping into new SharePoint‑specific agents such as the Knowledge Agent to help organize and hone your site and site contents. Whether you build sites, publish content, or support teams, you’ll leave with clear examples and repeatable patterns you can apply immediately—no technical background required except the willingness to try new things.

Friday 9:00 AM - 10:10 AM · Room 423-424

From Catch-Up to Takeoff: How to Take Charge of an Existing M365 Copilot Adoption Journey

Starting a Microsoft 365 Copilot adoption journey is one thing when you're there from the beginning—but what happens when you step into an initiative that’s already underway?

In this session, Lindsay Shelton shares her experience stepping into a newly created role focused on Microsoft 365 adoption and training—and how she took the lead in shaping and driving the company's M365 Copilot adoption efforts. When she stepped into the role, multiple adoption efforts were already underway—but many had stalled or lacked clear ownership. She identified these gaps, reignited momentum, and provided the structure needed to turn scattered initiatives into a cohesive strategy that empowered users at scale.

You'll gain insights into assessing an ongoing adoption effort, fostering collaboration, and implementing training that empowers employees to make the most of Copilot. Whether you're launching an adoption plan or enhancing an existing one, you'll walk away with practical strategies to guide your journey.

Key takeaways: • How to assess and document an ongoing adoption effort • Strategies for gathering data around existing usage across departments • Best practices for user engagement and training at scale • Lessons learned from a real-world Microsoft 365 Copilot adoption case study

Friday 9:00 AM - 10:10 AM · Room 431-432

Getting Started with Work IQ - Integrating Microsoft Content Into Any Agent

Using Work IQ you are able to ground agents in Microsoft 365 content through MCP-based tools. With Work IQ you can give any agent access mail, meetings, documents, Teams content, and other Microsoft 365 data. In this session attendees will learn the process for wrapping Work IQ access in a custom skill or tool and calling Work IQ from an agent - any agent, not just Copilot. You’ll leave with a clear, practical starting point for extending your own agents with Microsoft 365 data in a way that respects existing permissions and keeps the architecture simple.

Friday 9:00 AM - 10:10 AM · Room 433-434

Unlocking AI: A Practical Guide for Business and IT Pros

"Treb was very knowledgeable and provided great insight on the past and future of AI."

"Great overview of what's possible with AI."

"I thought Copilot and ChatGPT was all there was to AI. Now I know there's a lot more." --Attendee Quotes from Feedback Survey

Many are struggling with AI A recent study from Slack showed that core AI literacy is a blocker for many AI projects. Discover why AI literacy is a game-changer for businesses worldwide. Join our workshop to demystify AI concepts, overcome common challenges, and gain a competitive edge. Let me guide you through the intricacies of AI and empower you to lead with confidence.

Problem #1: Lack of Clarity on AI's Practical Impact. Many organizations still struggle to understand impactful AI use cases beyond initial experiments.

Problem #2: Integration Into Existing Processes. Difficulty in smoothly integrating AI into current workflows without major disruptions.

Problem #3: Skepticism and Resistance. Employees and leaders have mixed feelings, ranging from fear of job loss to skepticism about AI's abilities.

Yet we know we work better with AI, we still struggle.

"...those who are trained to use AI are up to 19 times as likely to report that AI is improving their productivity." Workforce Index Study 2024, Slack.

AI Literacy with Focus on the Real World Embark on a transformational journey with AI literacy tailored for business and IT professionals. Overcome the hurdles of AI technology adoption and pave your path to business excellence. Gain practical insights and strategies to thrive in the evolving digital landscape.

You'll learn: Understand AI Concepts Discover the diverse applications of AI across industries and how they drive innovation and efficiency in business operations.

Enhance Communication with AI Master the art of AI communication, transitioning from search queries to insightful dialogues with ease.

Master AI Training Techniques Elevate your skills by understanding the distinctions between training, fine-tuning, and grounding practices.

Curriculum

• Introduction
• What is AI, where we cover the different types of AI and how they are used
• What are the best AI uses today?
• Basics of how LLMs work, hallucinations, and more
• What's next in AI?
• Conclusion

Friday 9:00 AM - 10:10 AM · Room 437-439

Better Instructions = Better Agents: Prompt Engineering & Debugging for Creators of any Background

Every agent platform gives you a text box to write instructions. Most people treat it like a suggestion field. But those instructions are the primary way you shape your agent's behavior, and there's a real craft to writing them well.

In this session, I'll break down what makes agent instructions effective: how Markdown structure helps the model parse your intent, when and how to use few-shot examples, how your word choices signal the depth of reasoning you expect, and why formatting decisions that seem minor can have outsized effects on agent behavior. We'll also tackle the part nobody talks about enough: debugging. When your agent ignores instructions, calls the wrong tool, or gives a weird response, you need a systematic way to figure out why. I'll show you techniques for using the agent as its own debugger, getting it to reflect on its instructions, explain its decision-making, and help you identify where your guidance is ambiguous or conflicting.

This isn't tied to any one platform. Whether you're working in Copilot Studio, creating Declarative Agents for Copilot with the M365 Agents Toolkit (ATK), Microsoft Foundry, OpenAI’s ChatGPT, Anthropic’s Claude, or any other agent framework, you'll walk away with a repeatable approach to writing instructions that make your agents more reliable, more useful, and easier to troubleshoot.

Friday 10:20 AM - 11:30 AM · Room 421-422

Strategic AI Adoption: Lessons Learned from Scaling Copilot Across the Enterprise

AI is no longer a future initiative – it’s a present priority. As AI transforms the modern workplace, enterprises face critical decisions around adoption, governance, and long-term strategy. In this session, we’ll share our organization’s journey to scale Microsoft Copilot – from pilot to enterprise-wide roll out.

Learn how we developed a practical roadmap, established governance and data readiness, and ensured security and compliance every step of the way. We’ll explore how we’re measuring adoption and ROI, using training and change management to drive sustained engagement, and embedding AI capabilities into day-to-day workflows.

We’ll also discuss our approach to data governance and information protection, share lessons learned, and provide a look ahead at how we’re expanding Copilot use cases and building toward a broader AI-enabled future.

Friday 10:20 AM - 11:30 AM · Room 423-424

How Copilot and AI are Evolving Intranets and How Our Users Interact with Them

different that we consume it today. Users have expectations on what answers they should get when they search or how an agent should react to their prompts. Copilot and other agentic experiences are changing the way users interact with information and their expectations. How do we evolve our intranets to meet these changing expectations and do it in a way that doesn’t hurt usability and accessibility? In this session we’ll cover

• The changes AI has brought to finding and searching for information • How AI can be used for creating content and design • What stays he same even with the use of Copilot and what may change • The added value of human-centered design

Friday 12:30 PM - 1:40 PM · Room 443-444

Bring Copilot Studio Agents to Life with Power Automate & Azure Functions

What

A 70-minute, demo-driven session that shows practical integration patterns to connect Copilot Studio agents with Power Automate and Azure Functions. The session emphasizes real use cases — including IT onboarding and offboarding — and demonstrates how agent → tool-call wiring and server-side functions enable agents to take verifiable actions in enterprise systems.

Why

Agents are most valuable when they can act on business processes: create tickets, provision accounts, perform lookups, trigger approvals, or call APIs. Repeatable integration patterns reduce engineering effort, limit privileges, make actions auditable, and lower operational risk — so teams can safely automate common workflows.

How (Demo -> use-case focused)

Pattern demos (three integration patterns):

Agent → Power Automate flow — demo Agent → Azure Function (HTTP/webhook) Hybrid (Flow → Function / orchestrator)

Use cases illustrated: IT onboarding & offboarding, service desk ticket creation & escalation, password resets, HR onboarding tasks, approval workflows (expense/time-off), order-status lookups, and knowledge-base retrieval with safe redaction.

Operational patterns shown: authentication options (service principal vs. managed identity), least-privilege permissioning, telemetry and correlation (linking agent transcripts to flow runs and function executions for audit), and production checklist items (error handling, idempotency, retries, and monitoring).

Implementation is shown at a demo level; full lab materials are available after the session for teams that want hands-on follow-up.

Who should attend

Developers, integration engineers, Power Platform implementers, solution architects, and platform owners who need to enable Copilot Studio agents to perform actions across Teams, web portals, and backend systems.

Friday 12:30 PM - 1:40 PM · Room 431-432

Become a SharePoint Agents Rockstar

Have you ever wondered how to make your SharePoint sites feel more dynamic and intuitive? We’re all about to discover how SharePoint Agents are completely changing the game. Join us as we explore what these intelligent, AI-powered assistants can do—whether you're using a built-in agent or a custom one. Through practical examples and real-world scenarios, you'll see how these agents can help you transform your business by boosting efficiency and reducing manual work, making collaboration more intuitive than ever before. This session is a must-attend for Makers and information workers. Still, it's also designed for IT teams, administrators, and business leaders who want to build a more innovative and responsive SharePoint environment.

Friday 1:50 PM - 3:00 PM · Room 421-422

Stop Oversharing: Designing an AI‑Safe Information Architecture for Microsoft 365

Copilot does not create new risk - it exposes the risk that already exists. Broken inheritance, broad access groups, and unclear content boundaries become far more visible when AI starts answering questions across Microsoft 365.

This session introduces an AI‑safe information architecture playbook designed for Copilot‑enabled environments. Rather than focusing on individual tools, we look at architectural principles: how content should be structured, how access should be modeled, and where collaboration boundaries must exist to prevent accidental exposure without slowing teams down.

We will explore real‑world access patterns that cause oversharing, common design anti‑patterns, and practical ways to redesign sites, libraries, and workspaces for AI‑driven discovery. The goal is not lockdown—it is intentional openness. Attendees will leave with a set of IA principles and decision frameworks they can apply immediately to reduce risk, improve Copilot response quality, and support sustainable collaboration at scale.