As AI and Microsoft Copilot become embedded in daily operations, organizations should be seeking to gain greater control over the data AI can access, process, and expose. The challenge is no longer just about identifying sensitive information. It is about establishing the right foundations for data governance, classification, protection, and risk management at scale.

In this session, we will introduce the Purview maturity journey: a practical framework for how organizations can evolve from fragmented and manual data protection practices toward a more automated, scalable, and AI-ready operating model. We will discuss the key capabilities, decisions, and governance mechanisms that influence success along the journey.

Building on real-world customer experiences, we will then explore practical implementation patterns, lessons learned, and common pitfalls. Attendees can gain insight into what works, what to avoid, and how to simplify classification, reduce oversharing, automate protection, support compliance requirements, and prepare data for the secure adoption of AI.

Rather than focusing solely on technology, the session will highlight the operational, governance, and change management considerations that differentiate successful implementations from those that struggle to deliver lasting value.

AI is fundamentally changing how organizations modernize applications and create business value. What once may have required years of investment and transformation can now be accelerated through AI-powered modernization, enabling enterprises to reduce technical debt, unlock innovation, and build a foundation for an AI-first future. Leveraging KPMG Blaze and Microsoft Azure, organizations are modernizing legacy applications faster, accelerating cloud adoption, and creating intelligent, resilient, and scalable business solutions.

This session explores how leading enterprises are transforming application portfolios, accelerating Azure value realization, and preparing their organizations for the next wave of AI-driven growth. Through real-world client examples and practical technical insights, attendees can learn how AI is reshaping modernization and changing the economics of digital transformation. 

We've sat in this exact meeting dozens of times. Different organizations, different industries, different locations, but the same pattern keeps repeating. The business redesigns the operating model, but implementation stalls because the underlying data was never addressed. Technology teams deliver impressive AI tech solutions, but they automate the same broken processes faster. And even when the technology and process design are right, adoption often struggles because the change journey was treated as an afterthought.

In this session, we'll share a framework that KPMG member firms have seen determine the success or failure of AI transformation: the Data Layer, AI Orchestration Layer, Process & Business Layer, and Change Layer.

Using a real-world composite engagement, we'll explore what happens when organizations focus on only one or two of these layers, and what changes when all four are designed and delivered together as a single transformation journey.

Attendees can leave with a practical way to identify which layer(s) their organization may be overlooking today, understand the consequences of that gap, and learn how to align data, AI, business processes and people to drive meaningful and sustainable transformation.

As organizations scale Microsoft Copilot and AI agents, the question is shifting from whether AI works to whether its growing consumption can be translated into measurable business value.

This session introduces the Frontier value method: a practical approach for selecting the right workflows, modelling the cost of human-agent operating models, and underwriting investment against successful business outcomes rather than tokens, licenses or theoretical time saved.

Attendees can see how leaders can connect AI consumption, operating-model redesign and evidence of realized value to make better investment decisions and scale what works with confidence.

Agentic AI is moving out of supervised pilots and into unattended production. The moment an agent books, approves, provisions or commits on the organization's behalf, a question that used to be theoretical becomes an operating requirement: when it acts on its own and gets it wrong, who answers for it?

This session seeks to take the question past policy. We look at what changes in practice when agents inherit human permissions — identity and entitlement, the audit trail, the kill switch, the escalation path, and the boundary between what an agent may propose and what it may execute without a human in the loop.

Panelists will share where accountability actually sits in their own deployments today, what they had to put in place before they would let an agent act , and what they will not automate at any level of model performance.

Attendees should leave with a practical checklist of the controls that should exist before autonomy is granted, and a clearer view of which decisions should stay human by design rather than by default. 

Disruption frequently moves faster than periodic reporting, and fragmented data can leave leaders without the shared context to see what matters, understand what is changing, and act in time.

This session will cover how trusted, governed enterprise data and knowledge can be connected with live operational signals and AI to create reusable intelligence — turning context into faster, evidence-based decisions and coordinated action at scale.

Panelists will share examples of real applications in operational resilience and digital twins: what had to be connected, what the intelligence layer actually reuses across cases, and where a live operating picture changed a decision that periodic reporting may have missed.

Attendees can gain a view of what an enterprise intelligence foundation needs to contain before resilience, and digital-twin use cases can run on it, and where decisions may be happening too late.