What came through clearly in that conversation is that most organizations aren’t struggling to adopt agents because the technology is unsafe. They’re struggling because their governance models were built for a world that no longer exists. 

While traditional security models still depend on a clear distinction between “inside” and “outside,” and that boundary absolutely still matters. What’s changed is the pace. Agents now move fluidly across apps, data sources, and workflows often spanning environments that were designed with autonomous or semi-autonomous systems in mind.

When building an agent or app can take minutes, imposing governance models built around week-long and manual review processes quickly break down. The challenge isn’t whether to govern—it’s how. Governance has to account for blurred boundaries and apply the right oversight so teams can move fast, without losing control. 

When governance strategies boil down to “lock everything down” or “we’ll figure it out later,” the outcome is predictable: either uncontrolled adoption or shadow IT with no visibility. Neither is a win. 

The governance questions every AI agent should answer

The most effective organizations aren’t trying to stop agents. They’re figuring out how to classify risk clearly and apply the right controls at the right time. 

If governance is just a list of things people can’t do, that’s not governance—it’s a backlog of workarounds waiting to happen. Disruptors and innovators always find a way – whether it’s inside the system or outside your line of sight. When there’s no supported path to do the right thing, shadow IT isn’t a failure of discipline, it’s the natural result. Constraints without alternatives don’t stop innovation, they just push it underground, encouraging shadow IT. 

Real governance sets boundaries that let teams move fast and stay safe: 

• What data sources an agent can access 

• How broadly can it be deployed or shared 

• What actions is it allowed to take 

• What identity does it run under 

• What level of oversight applies as risk increases 

A low-risk personal productivity agent is not the same as an agent connected to a core business system. Treating them as if they are leads to predictable points of failure. You either over restrict everything and stall innovation, or you under-protect what actually matters, leaving critical systems exposed. Governance only works when it reflects the real differences in risk. 

Risk isn’t binary—and AI governance needs a risk-based model

A practical way to make this operational is a simple risk-based model. Not theoretical. Operational. 

Think in terms of graduated risk zones: 

• Low risk: constrained, self-serve scenarios where people can build and use agents with tight guardrails—limited data access, limited sharing.  In this scenario, makers don’t need to open a ticket for every idea, and IT doesn’t have to micromanage. Teams can move quickly, building with confidence, without friction, and IT can stay out of the critical path,  

• Medium risk: broader sharing, more sensitive data, more meaningful actions. These scenarios trigger review and oversight—but without resetting momentum or forcing heavyweight governance on every idea. 

• High risk: business critical workflows tied to core systems. These need deliberate control from day one. Not “nobody can build,” but “the right people build inside the right boundaries, with the right oversight.” 

The point isn’t the labels. The point is clarity. Risk is contextual. Governance should be too. 

Where governance actually gets enforced: the platform

Governance only works when it’s enforced inherently by the platform, not layered on through policy decks, emails, or spreadsheets. 

That’s why a concept of managed platform matters: Make security, governance and operations part of the platform experience—inventory, usage insight, controlled sharing, connector governance, and lifecycle management—rather than an external process held together by best intentions. 

Managed environments—our practical adoption of the managed platform concept—are a Power Platform capability, not something limited to a single product or workload. Managed environments enable teams to manage apps, automations, pages, and even agents built in Microsoft Copilot Studio 

One of the cleanest controls is also one of the simplest: sharing limits paired with a clear on‑ramp. 

If someone builds something for themselves or for their immediate teammates, that’s one risk profile. If they want to share their solution more broadly, that’s a different one. The platform needs to distinguish between those cases. When it does, you can let people experiment freely—and require deliberate promotion, review, and accountability when something is ready to scale. 

Agents don’t create permission problems—they expose them 

This bears repeating, because it matters: agents generally operate as the calling user. They don’t magically gain new permissions. Which means agents don’t create access problems; they expose the ones you already have, faster. 

If users have overly broad permissions today, agents will too. That’s not an agent problem—it’s an identity and access discipline problem. Effective agent governance only works when it’s built on solid foundations. 

Trust by design, with verification built-in 

Strong proactive controls matter, but they’re not enough on their own. You still need reactive controls: monitoring, diagnostics, and audit trails, especially when agents take actions with compliance implications. 

Trust but verify still applies. Looking at the familiar expense-approval analogy: humans unintentionally approve things incorrectly all the time. We manage that risk with audits, compensating controls, and limits on blast radius. Agent risk should be treated the same way: know what happened, understand why it happened, and contain the impact when it doesn’t go as planned. 

The takeaway 

The future isn’t “agents everywhere with no control,” and it’s not “no agents because risk.” Both fail. 

The practical path is adaptive governance: classify risk clearly, enforce it through the platform, and create promotion paths so good ideas can scale without turning into tomorrow’s incident response. 

That’s how organizations stop playing defense. If you want to learn how you can start saying “yes” safely, please watch the full interview with Fernando. 

The post ​Building trustworthy AI: A practical framework for adaptive governance appeared first on Microsoft AI Blogs.

As small and midsize businesses scale, finance and operations become harder to manage. Transactions increase, reporting requirements expand, and disconnected systems start to strain visibility and margins. What once worked becomes a constraint. 

A newly published Forrester Total Economic Impact™ (TEI) study helps quantify what many organizations are already experiencing: modernizing on Microsoft Dynamics 365 Business Central delivers measurable financial impact.

Microsoft commissioned Forrester Consulting to evaluate the potential return on investment organizations may realize by deploying Business Central. Based on interviews with four business decision makers, which were aggregated to model a fictitious composite organization¹, the study projected that the composite organization could potentially realize: 

• More than 200% return on investment (ROI) over three years 

• An estimated $460K net present value (NPV) over three years 

• Potential payback in six months 

Across finance productivity, enterprise resource planning (ERP) consolidation, improved profitability, and reporting efficiency, the composite organization modeled by Forrester realized more than $680K in three‑year, risk‑adjusted present value benefits. 

These outcomes reflect the potential impact of modernizing finance and operations on a single, integrated cloud ERP platform—while also establishing the foundation for AI‑powered experiences like Microsoft 365 Copilot and intelligent agents. 

Where the value comes from 

The Forrester TEI study highlights several areas where organizations can potentially realize tangible, risk‑adjusted benefits when using Business Central. 

Support faster, more efficient finance operations 

Manual processes often slow growing organizations. Interviewed customers reported meaningful efficiency gains across accounts payable (AP), accounts receivable (AR), billing, and financial close. 

By year three, the composite organization was projected to potentially achieve: 

• Up to 30% reduction in monthly close time 

• Up to 50% time savings for AP, AR, and billing activities 

These improvements translated to more than $215K in present value over three years in finance productivity alone for the composite organization—allowing teams to shift focus from reconciliation to higher‑value analysis. 

Standardizing data and workflows in Business Central also creates the conditions necessary for AI‑enabled automation. While AI benefits were not independently quantified in this study, interviewees noted that unified processes accelerate the adoption of Copilot‑supported approvals, variance analysis, and exception handling. 

Lower total cost of ownership through ERP consolidation 

Interviewees reported operating aging on‑premises ERP systems alongside spreadsheets and disconnected point solutions. Consolidating onto Business Central reduced infrastructure complexity and IT overhead. 

The study projected the following potential benefits for the composite organization over three years: 

• More than 10% reduction in total cost of ownership (TCO) 

• More than $170K in present value savings from retired systems and reduced maintenance 

Beyond direct savings, simplification can reduce operational risk and improved scalability—allowing organizations to grow without layering on new systems to compensate for gaps. 

Enable improved profitability through better visibility 

Unified, real‑time visibility across finance and operations enables faster, more informed decisions. 

By year three, the composite organization was modeled to potentially experience: 

• Up to 3% improvement in net profit margins 

• More than $245K in present value from improved profitability 

Better insight into costs, projects, and performance enables earlier course correction. AI‑powered experiences such as Copilot can further assist by surfacing cost variances, project overruns, or unbilled work sooner. While AI alone does not drive margin improvement, modern ERP data and standardized processes strengthen an organization’s ability to act with precision. 

Fast reporting and audit readiness 

Business Central’s integrated data model and native Microsoft Power BI capabilities streamlined reporting and audit preparation. 

Based on modeling by Forrester, by year three, the composite organization was projected to potentially reduce: 

• Audit preparation time by up to 30% 

• Time spent creating internal and executive reports 

These projected improvements were valued at nearly $50K in present value, while also enabling increased confidence in data accuracy and consistency. 

Beyond the numbers: Building an AI-ready foundation

In addition to quantified financial outcomes, interviewees highlighted broader operational improvements, including enhanced customer experience, reduced days sales outstanding (DSO), better warehouse management, and a more intuitive user experience. 

Just as importantly, Business Central provides an AI‑ready ERP foundation. 

By unifying finance and operations data and aligning processes to best practices, organizations are better positioned to leverage Copilot, Power BI, and intelligent agents to: 

• Help reduce time-to-insight—not just report creation
• Surface anomalies and trends quickly 
• Enable more proactive, data‑driven decision‑making 

While AI‑driven outcomes were not directly measured in this TEI, the study reinforces a critical principle: realizing AI value at scale depends on clean data, integrated systems, and standardized processes. Business Central delivers that foundation. 

Read the full study 

For SMB‑focused organizations and partners evaluating ERP modernization, the full Forrester TEI study provides a detailed financial framework to help quantify potential projected value—grounded in customer interviews and risk‑adjusted modeling. 

Join us at Directions North America 2026 

The Forrester TEI study highlights the potential value organizations may realize with Business Central—from faster financial processes to improved profitability and lower total cost of ownership. 

We’ll continue these conversations at Directions North America 2026, where partners, Microsoft engineering, and product leaders come together to discuss what’s next for the Business Central ecosystem and the future of AI-powered ERP. 

Join me for the keynote, where I’ll explore how Business Central is evolving into an AI-powered system of action with Copilot and intelligent agents. 

Attendees will: 

• Gain insight into the Business Central roadmap 
• Prepare for upcoming AI-powered capabilities 
• Connect directly with Microsoft engineering and product experts 
• Engage with partners across the Business Central community 

1. Composite organization assumption: Results are based on a Forrester modeled composite organization derived from customer interviews, with $50M in annual revenue, 300 employees, 15 core finance and accounting users, and 100 light users, using Dynamics 365 Business Central in a cloud deployment. All quantified benefits represent the three-year, risk-adjusted present value for the composite organization.

The post Enable accelerated growth with confidence: A Forrester TEI study projects more than 200% ROI over three years and six-month payback using Dynamics 365 Business Central appeared first on Microsoft AI Blogs.

Today, I am pleased to announce a collaboration between Microsoft and Armada to deliver a practical path to sovereign AI at the edge. Together, we are bringing Microsoft Sovereign Private Cloud capabilities to Armada’s Galleon modular datacenters (MDC), enabling customers to run secure, compliant workloads designed to operate in intermittently connected, contested, and even fully disconnected environments. This customer-controlled cloud environment delivers Azure’s operating model, security, and AI-ready capabilities where traditional cloud approaches are not feasible.

Meeting sovereign requirements anywhere

Defense, public safety, energy, and critical infrastructure operators increasingly need cloud capabilities in locations where using public clouds is not feasible. They require workloads to run in environments that are disconnected, mobile, or operationally constrained. In these settings, cloud capabilities must move closer to the point of need. These scenarios often demand:

• Disconnected or limited connectivity.
• Portable or rapidly deployable infrastructure.
• Strict data residency and regulatory controls.
• Support for modern AI and analytics workloads at the edge.

Through this collaboration, Microsoft and Armada are delivering a validated sovereign reference architecture that shows how Sovereign Private Cloud operates on and interoperates with the Armada Edge Platform, enabling customers to deploy Azure services closer to where data is created, while retaining full control over their data, operations, and governance.

Azure Local in Armada’s Galleon modular datacenters

At the core of this collaboration is Azure Local, Microsoft’s on-premises cloud platform that can be used in disconnected and sovereign scenarios, combined with Armada’s Galleon MDC and Armada Edge Platform (AEP).

Together, the solution supports:

• Azure Local control plane and managed clusters, including multi-rack scalability.
• Flexible storage architectures, including hyperconverged and SAN-backed deployments.
• Resilient multi network connectivity, spanning satellite, LTE/5G, RF, and SD-WAN.
• Security, compliance, and hardening aligned to sovereign, government, and regulated workloads.

The result is an edge platform that can be deployed in remote, mobile, or constrained environments while still benefiting from Azure’s consistent cloud operating model.

Enabling sovereign AI and mission critical workloads

Beyond infrastructure, this collaboration is focused on delivering sovereign AI capabilities at the edge.

As part of Microsoft Sovereign Private Cloud, Foundry Local and Azure Local enable customers to deploy, govern, and operate AI entirely within their own trusted boundary, supporting national sovereignty, classified workloads, and highly regulated data pipelines. With Foundry Local, customers can run AI inference and analytics locally, even when disconnected from the public cloud.

This approach helps customers:

• Process sensitive data locally to meet sovereignty requirements.
• Reduce latency for real-time decision-making.
• Operate AI workloads in austere or bandwidth-constrained environments.

By combining Foundry Local and Azure Local’s cloud consistent platform with Armada’s deployable infrastructure, customers gain a practical path to operational AI, where it matters most.

A shared vision for sovereign edge infrastructure

Customers operating in the world’s most demanding environments don’t have the luxury of choosing between sovereignty, resilience, and modern cloud capabilities, they need all three. By partnering with Microsoft, we’re combining Armada’s deployable, mission ready infrastructure with Azure Local’s consistent cloud platform to help governments and regulated industries run secure, AI enabled workloads anywhere they operate, even when connectivity is limited or unavailable. Together, Microsoft and Armada are delivering a practical path to sovereign AI at the edge, one that respects local control, supports disconnected operations, and scales from today’s mission critical needs to tomorrow’s intelligent systems.

—Dan Wright, Co-Founder and CEO of Armada

Looking ahead

Achieving digital sovereignty is no longer just about where data lives, but where intelligence runs, who controls it, and how resilient it remains under real-world conditions.

With this collaboration, Microsoft and Armada are extending Azure to the edge in a way that respects sovereignty, enables AI, and meets customers where they operate, whether that’s in remote locations, mobile deployments, or highly regulated environments.

• Learn more about Azure Local and sovereign edge solutions.
• Discover Armada MDC and edge platform.

The post Building sovereign AI at the edge: Microsoft and Armada collaborate to deliver Azure Local on Galleon modular datacenters appeared first on Microsoft AI Blogs.

For the first time, SQLCon joined the event, bringing the Microsoft Fabric and SQL communities together in Atlanta for an unforgettable week of learning, growing, and connecting.

More than 8,000 attendees gathered across nearly 300 sessions to explore the latest in Fabric and SQL. Some of the most valuable moments did not happen on stage. They happened through conversations and shared experiences across the community.

That is what makes FabCon special. It is not just what you learn. It is who you learn it with.

From packed sessions to impromptu discussions in the hallways, FabCon and SQLCon 2026 reflected the growing energy and momentum behind the Fabric and SQL ecosystems.

Here are some of the highlights from this year’s event.

If you haven’t already, check out Arun Ulag’s hero blog “FabCon and SQLCon 2026: Unifying databases and Fabric on a single, complete platform” for a complete look at all of our FabCon and SQLCon announcements across both Fabric and our database offerings.

A week of learning and collaboration

The week began with pre-conference workshops that allowed attendees to dive deeper into Microsoft Fabric and SQL through hands-on learning.

These sessions covered topics ranging from data engineering and real-time analytics to governance, AI, and advanced SQL scenarios. Participants worked directly with product teams and community experts to explore best practices and build new skills that they could bring back to their organizations.

For many attendees, the workshops set the tone for the entire conference. They offered not only technical depth, but also early opportunities to meet peers, exchange ideas, and start conversations that continued throughout the week.

As the main conference kicked off, keynotes and breakout sessions explored how databases, analytics, real-time data, semantic intelligence, BI, and AI are coming together on Microsoft Fabric, and how organizations are using the platform to turn insights into action faster.

The Expo Hall experience

At the center of the Expo Hall, the Microsoft booth became a natural gathering point throughout the week.

It was a place to connect, explore, and take a break between sessions. Attendees grabbed coffee served by a robot, sampled custom Fabric and SQL-themed Coca-Cola flavors dispensed from Freestyle machines, and spent time with engineers and product teams while picking up a bit of conference swag.

The mix of hands-on experiences and casual conversations made it one of the most active and engaging spaces at the event.

Nearby, the Ask the Experts area gave attendees the chance to connect directly with Microsoft engineers, product managers, and community leaders. These one-on-one discussions helped participants get guidance on everything from Fabric workloads and SQL optimization to governance, security, and AI strategy. Not to mention the wealth of partners also represented in the Expo Hall allowing attendees to engage with the expertise gathered across the ecosystem.

These moments of direct engagement often turned into some of the most valuable conversations of the week.

Competition, creativity, and hands-on challenges

FabCon wouldn’t be complete without a little friendly competition.

Throughout the week, attendees tested their skills in Fast at Fabric, a timed analytics challenge where participants worked through real-world scenarios using Microsoft Fabric. Competitors analyzed data, validated assumptions, and generated insights while climbing a live leaderboard. The experience blended learning with competition in an engaging format.

The Trivia Challenge in the Microsoft booth also kept the energy high. Attendees tested their knowledge of Fabric and SQL in a fast-paced game that mixed learning with plenty of laughs. Interactive dashboards showcased Fabric capabilities while participants competed for prizes and bragging rights.

These experiences turned technical learning into something interactive and memorable while encouraging attendees to dive deeper into the platform.

Celebrating creativity and storytelling with data

The DataViz World Championships delivered one of the most anticipated and electric moments of the week.

As the live finale of an international Power BI competition, the event brought four finalists to the stage to build visualizations in real time using a brand-new dataset. In front of a live audience, they combined creativity, technical skill, and storytelling to turn raw data into compelling insights.

By the end of the session, one finalist earned the title of 2026 Power BI DataViz World Champion, closing out the competition with one of the most memorable moments of the conference.

Community at the center of the experience

At the heart of FabCon was the Community Lounge, a welcoming space dedicated to connection and community-led experiences.

Throughout the week, the lounge hosted meetups, community theater sessions, certification guidance, and informal gatherings where attendees could exchange ideas and build relationships. It became a natural meeting point where conversations continued long after sessions ended.

The event also featured a Women in Data luncheon and a DEI luncheon, which created space for discussions around leadership, resilience, and inclusive growth within the technology community. These sessions encouraged thoughtful dialogue and helped strengthen connections among attendees from diverse backgrounds.

Together, these experiences reinforced what makes the Fabric community unique: a culture of openness, collaboration, and shared learning.

Unforgettable moments from the week

FabCon is known for creating memorable experiences, and this year delivered plenty of them. Power Hour, a fan-favorite event, brought together Microsoft product teams and community leaders for an energetic session filled with creative demos, storytelling, and unexpected surprises.

Meanwhile, FabCon TV broadcast live episodes from the Expo Hall throughout the event. Interviews with speakers, community members, and product experts captured the excitement of the conference and will continue to reach audiences through recordings shared after the event. Be sure to catch these episodes as they’re released on the Fabric YouTube channel!

To celebrate the week, attendees gathered at the world-famous Georgia Aquarium to take in the unique surroundings and connect. Conversations flowed easily between admiring the unique marine ecosystem and the next Fabric and SQL project, offering a memorable backdrop for new ideas and relationships to take shape.

Building on the energy of FabCon and SQLCon

FabCon and SQLCon 2026 highlighted the incredible momentum behind the global data community.

Over the course of the week, thousands of attendees learned new skills, built connections, and explored how Fabric and SQL technologies are shaping the future of data and AI.

Most importantly, the event celebrated the people who make this ecosystem so vibrant. The ideas, insights, and collaborations that began in Atlanta will continue long after the conference ends.

Thank you to everyone who joined us this year. We look forward to seeing what the Fabric and SQL communities build next and we’re already excited to see everyone at the next FabCon SQLCon in Spring 2027.

Explore additional Fabric resources:

• Visit the Fabric website
• Join the Fabric community
• Stay tuned in to the official Fabric YouTube channel
• Read the latest updates on the Microsoft Fabric blog

The post Highlights from FabCon and SQLCon 2026: Where databases and Fabric come together appeared first on Microsoft AI Blogs.

Across banking, capital markets, and insurance, some of the fastest-moving institutions are not simply “adopting AI.” They are becoming Frontier Firms, AI-powered organizations built around human-agent collaboration. In a sector where the cost of error is high, the financial services sector is emerging as an early proving ground for the Frontier Firm model.

The Microsoft 2025 Work Trend Index highlights a widening AI divide. While many organizations remain stuck in pilot mode, Frontier Firms are scaling agentic AI across their operations.

Our work with financial services leaders worldwide shows a clear pattern. The winners in the next generation of innovation will be those that combine human judgment with AI and agents, without compromising security, compliance, or customer trust. Critically, these advantages are best enabled through migration to a modern cloud foundation that can scale AI responsibly and reliably.

The crossroad: Modernize or let legacy debt grow?

Legacy systems have powered financial services for decades. Yet the very qualities that once made them indispensable—custom integrations, tightly coupled architectures, and deeply embedded processes—now create friction and fragility. Increasingly, they can be expensive to maintain, slow to change, and difficult to secure end-to-end. Worse, they can inherently constrain data access across the business, which limits advanced analytics and AI from delivering full value in key areas like customer engagement, fraud prevention, credit decisions, underwriting, and financial crime.

In many institutions, this accumulated technical debt is, in effect, an understated balance-sheet liability. It can increase operational overhead, complicate resilience planning, and broaden the cyber-attack surface. At the same time, regulators are demanding that firms prove stronger controls while, competitively, digital-native challengers are showing what’s possible when technology is designed for continuous change.

Modernization can help answer many of these challenges by helping position firms to gain competitive advantages that go well beyond cost efficiency. As workloads become increasingly cloud-native (in other words, designed to be built, updated, and scaled continuously in the cloud rather than tied to legacy infrastructure), organizations can launch new services faster, respond with agility, and use AI as part of everyday operations.

Waiting to migrate can increase risk and cost

A variety of factors are converging to increase the urgency of modernizing.

• Regulatory pressure is growing. Requirements for operational resilience, third-party risk oversight, data governance, and AI accountability are becoming more explicit and more enforceable. In Europe, the Digital Operational Resilience Act (DORA) raises the bar on stress testing, incident reporting, and information and communication technology (ICT) governance. In parallel, the European Union AI Act introduces demanding expectations for high-risk AI, including transparency, explainability, and bias mitigation. Globally, frameworks shaped by Basel guidance and securities regulators continue to push for stronger risk management, auditability, and controls across financial operations.
• Customer expectations are becoming non-negotiable. “Digital-first” now means more than building a polished mobile app. It means enabling instant transactions, proactive service, and personalized guidance—delivered consistently across channels. Doing all this at scale means that data must move securely and quickly, products should evolve continuously, and controls must be embedded rather than bolted on.
• The threat landscape is getting scarier. Threat actors are using automation and AI to increase both scale and sophistication. In a legacy environment, security improvements often arrive as point solutions, unevenly applied, and hard to validate. Cloud architectures, implemented with the right governance, help enable consistent identity controls, continuous monitoring, and policy-based protection that can be audited and improved over time.

Migration as a lever for innovation

Migration is too often framed as a technology initiative. For business and risk leaders, the more useful long-term view is as to regard it as a control and value strategy, a way to embed governance into the operating fabric of the firm.

This is why many transformation leaders manage cloud adoption as a sequence rather than a singular initiative, with a pathway from rehosting (“lift-and-shift”) through optimization and ultimately to AI acceleration. In this framing, modernization is not the finish line; it is the first step of compounding advantage.

Cloud migration, when managed well, can support a compliance‑by‑design approach, by which policy, identity, and data protections are consistently enforced. It can strengthen operational resilience through architectures that are built for redundancy, automated recovery, and continuous validation. And it can create an innovation pathway by making agentic AI practical to deploy and manage.

The AI-first divide: Cloud as operating model

As we see with Frontier Firms in financial services, innovation leaders tend to treat cloud architecture as more than an infrastructure choice. They use it as an operating model to standardize controls, build reusable platforms, and design processes that are increasingly AI-operated but human-led. The payoff can show up in faster deployment cycles, a lower cost per transaction, and predictive insights that make customer experiences more personal and operations more resilient.

Reaching that maturity typically requires progress across four transformation engines:

• Infrastructure modernization
• Legacy systems migration
• Systems modernization (including new business systems)
• Data modernization with AI integration

Financial services firms face stricter scrutiny than most industries, so the differentiator is not speed alone, it’s the ability to sustain speed while continuously demonstrating security, compliance, and control effectiveness.

We see this in practice across the industry. For example, UBS, following its acquisition of Credit Suisse, migrated a mission‑critical records platform from mainframe to a cloud‑native service on Microsoft Azure, reducing total cost of ownership by nearly 60% and improving their ability to meet regulatory demands. After LSEG migrated its high-volume, mission-critical Autex Trade Route (ATR) trading network from on-premises to Azure, the gains in scalability and resilience helped them absorb a sudden 400% surge in trading volumes with zero incidents. And the National Bank of Greece modernized document processing to improve accuracy and enable faster, more digital customer journeys. The common thread is not a single tool or model, it’s a cloud foundation that supports governed data, resilient operations, and repeatable innovation.

Turning migration into long-term value

For many firms, the hardest part of migration is not the technology; it’s making the journey auditable, repeatable, and aligned to risk appetite. That’s why a structured approach matters.

The Microsoft Cloud Adoption Framework, tailored for financial services, is designed to help institutions align cloud modernization to business outcomes while addressing the governance realities of the industry: data sovereignty expectations, operational resilience, and security-by-design. Importantly, cloud migration need not undermine data sovereignty; done right, migration strengthens locality, control, and compliance through governed architectures.

In practice, migration means helping businesses to build a compliant foundation, innovate responsibly, and maintain continuous control visibility as they scale. Microsoft supports this with financial-services-ready architectures, built-in governance and security capabilities, and a broad set of certifications and controls. Just as importantly, we work closely with customers and regulators globally to help ensure that cloud adoption can be evidenced properly in terms of risk reduction, resilience, and measurable operating improvement.

Trustworthy AI starts with the cloud foundation

Boards and regulators are right to focus on AI governance. Generative AI, agentic systems, and intelligent automation can improve productivity and customer outcomes, but only when they operate on governed data, with strong identity controls, clear lineage, and auditable policies. Those prerequisites are difficult to achieve in fragmented legacy environments.

Cloud migration creates the conditions for AI to be adopted responsibly, with modern data platforms and pipelines, elastic compute for experimentation and scale, consistent policy enforcement, and continuous monitoring.

To help institutions navigate migration with confidence, Microsoft combines a financial-services-tailored methodology with practical tooling and built-in governance. The Cloud Adoption Framework for financial services provides a proven, risk-aligned approach to planning and executing secure migrations. Azure Migrate and the Azure cloud migration and modernization programs help accelerate discovery, modernization, and execution with guidance and incentives. And capabilities like Microsoft Purview and Microsoft Defender for Cloud help establish compliance guardrails and security posture management from day one.

Lead the next generation with cloud

Migration is not the end state of digital transformation. It is the foundation for Frontier transformation, one which can enable firms to innovate faster, demonstrate stronger controls, and adapt quickly to new demands and opportunities.

The financial services firms that lead in the next generation of financial services will not be those that move the fastest in a single quarter. They will be the ones who modernize with technology that is durable, designed for operational resilience and evidence-based governance, and that makes innovation repeatable. Cloud migration is the inflection point where these powerful advantages become possible.

Learn more

• To explore a five-phase, risk-aligned approach to cloud migration that helps build the foundation for AI-first financial services, read the e-book: Becoming an AI-First Firm in Financial Services.
• Register for the Migrate and Modernize Summit to learn from industry leaders.
• Explore the Microsoft Cloud Adoption Framework for Financial Services.

The post Why cloud migration is key to realizing AI value in financial services appeared first on Microsoft AI Blogs.

We recently shared that we’re bringing the technology platform that powers Claude Cowork into Microsoft 365 Copilot. Today, Copilot Cowork—designed for long-running, multi-step work in Microsoft 365—is available via the Frontier program. Join Frontier to get early access to Microsoft’s latest AI innovations and learn more about Copilot Cowork here.

Copilot Cowork makes it easy to delegate and complete work. Describe the outcome you want, and Copilot Cowork creates a plan, reasons across your tools and files, and carries work forward with visible progress and opportunities to steer. With skills from Claude and Microsoft built in, such as calendar management and daily briefing, Copilot Cowork can handle everything from one-off tasks to repeatable workflows like a monthly budget review. Organizations like Capital Group had early access to Copilot Cowork and report already seeing its value—from planning, scheduling, and creating deliverables to preparing for executive reviews.  

We have been using Copilot since its launch in 2024, and the new capabilities in Cowork will help us automate and scale the Copilot ecosystem. This isn’t about generating content or answers. It’s about taking real action—connecting steps, coordinating tasks, and following through across everyday workflows. Because Cowork operates on our enterprise data and within our security and risk boundaries, we can experiment, learn, and scale with confidence. That allows us to move faster and focus AI in places where it actually delivers value.

Barton Warner, SVP of Enterprise Technology at Capital Group.

We are also excited to announce new features in Researcher built on multi-model intelligence. Today, Researcher helps you tackle your most complex questions by synthesizing information across sources, generating comprehensive analysis, and delivering cited, well-reasoned responses you can act on with confidence. 

Researcher’s new Critique feature takes this even further, using a combination of models from Frontier labs including Anthropic and OpenAI to separate generation from evaluation. One model plans the task and creates an initial draft, while a second model focuses on refinement, acting as an expert reviewer before the final report is produced.

The results are measurable—Researcher now scores 13.8% higher on the Deep Research Accuracy, Completeness, and Objectivity, or DRACO benchmark, the industry standard for deep research quality.  

And with Researcher’s new model Council, you can compare responses from different models side by side, instantly seeing where they agree, where they diverge, and what each uniquely brings to the table. It’s like having multiple researchers at your fingertips. Learn more here. 

Get started today 

All of this innovation is part of Wave 3 of Microsoft 365 Copilot, which marks a turning point in how AI shows up at work: intelligence that understands the context of work, and trust that allows AI to scale safely across the workforce. When intelligence and trust move together, AI stops being an experiment and starts becoming how work gets done. Visit Microsoft365.com/copilot or download the Microsoft 365 app on your mobile device to get started. 

The post Copilot Cowork: Now available in Frontier appeared first on Microsoft AI Blogs.

This is where the frontier is forming. While copilots help individuals be more effective, agents act on behalf of people. They carry out tasks, orchestrate multi-step workflows, and operate across systems continuously. And they’re moving quickly from experimentation to mainstream use. An IDC InfoBrief, sponsored by Microsoft, shows that 37% of organizations surveyed use agentic AI, another 25% are experimenting with it, and 24% are planning to use it the next 24 months.¹ Organizations that have already invested in people, skills, and responsible practices may be better prepared to operationalize agents at scale—and convert AI’s promise into real business performance.

Five strategic moves for introducing agents responsibly

The new Agents in the Workforce Handbook builds on those earlier foundations. Where the first blog in this series focused on empowering your people, and the second explored how to pair human judgment with AI systems, this third chapter looks ahead: How do you introduce agents into your workforce responsibly and intentionally? Below are five strategic moves leaders should consider. These are high-level guideposts; the Handbook goes much deeper with templates, examples, and decision frameworks to support implementation.

1. Start with your most persistent pain points

When organizations begin exploring agentic AI, a common challenge is prioritization. Imagining use cases is easy. Choosing where to start is harder. Successful organizations don’t begin with futuristic ideas—they begin with the familiar, recurring friction points that quietly drain time and introduce risk.

These are often the workflows teams have learned to “live with”: manual triage, routine follow-up, coordination across systems, repeated reporting steps, or tasks with high error potential. Leaders should observe how work truly happens—shadowing teams, reviewing process maps, and asking simple but revealing questions:

• Where do we lose time?
• What gets done manually that shouldn’t be?
• What feels broken—but no one owns?

These pain points typically offer the clearest path to early value. Addressing them not only frees capacity but also demonstrates to teams how agents can meaningfully improve the day-to-day. The Agents in the Workforce Handbook includes a readiness assessment and real-world patterns to help leaders identify and sequence the right opportunities.

2. Define your AI goal—and lead the change yourself

Introducing agents isn’t only a technical shift—it’s a leadership shift. Frontier Firms choose to align their early agent initiatives around bold, measurable goals: reducing manual work, accelerating cycle times, improving customer responsiveness, or expanding sales capacity. These goals create alignment and momentum, helping teams understand why agents matter and what success looks like.

But goals alone don’t change culture—leaders do. The organizations that move fastest are those whose executives personally model new ways of working. They use agents in their own workflows, talk openly about learnings, and recognize early adopters who demonstrate impact. They also acknowledge that change requires habit‑building. Experimenting with agents for even 20 to 30 minutes a day can materially improve adoption and confidence.

Skilling plays a central role. As Jeana Jorgensen, Corporate Vice President of Global Skilling, notes:

We’re hearing from many of our customers and partners that they expect employees across different roles to spend about 15 to 20% of their week learning and integrating AI into their daily work.

The Handbook offers guidance for identifying the roles, skills, and operating rhythms needed to support agent adoption.

3. Measure what works—and double down where it does

As with any transformative technology, early wins with agents need to be measurable and repeatable. Leaders should ensure visibility into how agents behave, how frequently they’re used, and the outcomes they produce. This isn’t about policing technology—it’s about giving teams the insights needed to improve and scale what’s working.

Effective organizations treat agent adoption like an operational discipline:

• They log and monitor agent activity.
• They measure time saved and business impact generated.
• They expand agents that demonstrate clear value.
• They refine or retire agents that don’t.

These data-driven insights help organizations move from experimentation to a consistent, enterprise-wide model for agent development—one where new ideas become shared services rather than isolated automations. The Handbook goes deeper into measurement strategies, including examples of what high-performing organizations track.

4. As agents become teammates, optimize continuously

Once an organization begins deploying agents across teams, a new challenge emerges: coordination. Agents that start out as individual productivity tools often become shared digital teammates—relied upon by multiple people, processes, and business functions. With that shift comes the need for thoughtful ownership, governance, and communication.

Successful organizations establish clear roles and responsibilities:

• Who owns each agent?
• Who can modify or update it?
• How are changes communicated to the people who rely on it?
• What happens when an agent’s behavior needs tuning?

Agents also require continuous improvement. As they’re used, they encounter edge cases, nuanced team preferences, and shifting processes. Over time, agents become more capable, and employees naturally evolve into “AI managers”—guiding digital apprentices the way they onboard and develop human teammates.

The Handbook provides deeper recommendations for governance models, centers of excellence, and cross-team alignment mechanisms that help organizations scale responsibly.

5. Reinvest the time saved—and push into innovation

While early value often shows up as efficiency, the long-term impact of agentic AI is much bigger: it creates renewed capacity for innovation. Frontier Firms understand that the goal isn’t to simply do the same work faster—it’s to free teams to pursue higher-value ideas, explore new business models, and elevate customer experiences.

Across industries, leading organizations are already demonstrating what this reinvestment looks like:

• L’Oréal is using AI agents to deliver personalized, at-scale beauty guidance.
• AT&T accelerated customer care resolution by 33%, saving millions while improving experience.

These examples highlight a crucial point: agents are not just workflow optimizers. They’re catalysts for reimagining how organizations deliver value. And the companies that begin investing now are positioning themselves for meaningful advantage.

Treat agents like teammates, not tools

The organizations achieving the strongest results view agents not as automations but as digital collaborators—systems that require feedback, tuning, and iteration. They integrate agents into team rhythms, treat them like growing contributors, and help their people evolve into confident AI managers.

This marks the natural third step in the Frontier journey: after empowering employees and strengthening the partnership between human expertise and AI (as explored in the first two blogs), organizations are now ready to bring digital teammates into the workflow in a structured, scalable way.

If your organization is ready to move from experimentation to scaled impact, the Agents in the Workforce Handbook offers the detailed guidance, examples, and templates to support your next phase of Frontier Transformation.

¹ IDC InfoBrief: sponsored by Microsoft, What Every Company Can Learn From Frontier Firms Leading the AI Revolution, IDC # US53838325, November 2025.

The post How to introduce agents into your workforce: 5 actions leaders can take appeared first on Microsoft AI Blogs.

The post For KPMG Canada’s Christine Andrew, Copilot isn’t just a time saver—it unlocks high-value impact appeared first on Microsoft AI Blogs.

The post Copilot in the C-Suite: Infobip’s Veselin Vuković on using Copilot to nurture partnerships appeared first on Microsoft AI Blogs.

Over the past several years, we have seen customer concerns evolve materially. Early conversations focused primarily on privacy and lawful data handling. Today, those concerns have expanded. Leaders are now asking how they maintain operational continuity during disruption, how they adopt AI responsibly without losing control, and how they protect national, organizational, and customer interests in an increasingly volatile global environment.

These questions are not abstract. They surface in boardrooms, procurement decisions, architecture reviews, and crisis simulations. They reflect a broader shift in how trust is evaluated in digital systems. Today in Brussels we brought together attendees from around the world—policy makers, IT leaders, and enterprises—to approach these questions from the multiplicity of perspectives to move the conversation from headlines to action.

From privacy to resilience and beyond

Privacy remains foundational. But it is no longer the sole lens through which sovereignty is assessed.

Customers are increasingly concerned about business continuity in the face of cyber incidents, geopolitical tension, supply chain disruption, and network instability. They want to understand how critical workloads operate if connectivity is constrained, if dependencies fail, or if policy conditions change with little warning.

At the same time, innovation pressures have intensified. AI is becoming central to public service delivery, national competitiveness, and economic growth. Organizations cannot afford to pause progress while sovereignty questions are debated in isolation. They need approaches that allow them to move forward responsibly, balancing opportunity with control.

What we hear consistently is this: sovereignty concerns will continue to evolve. Any approach that treats them as static is already behind.

For four decades, Microsoft has operated under some of the world’s most demanding data protection, competition, and digital governance frameworks. Working closely with European institutions, regulators, and customers has shaped how we think about sovereignty—not as a regional exception, but as a discipline that must function at scale, under scrutiny, and over time. That experience matters because many of the sovereignty questions now emerging globally were first tested in Europe, long before they became mainstream elsewhere.

A consultative approach to risk management

This is why we believe digital sovereignty must be approached as consultative risk management, not a checkbox or a predefined deployment model.

Every organization faces a unique mix of regulatory obligations, cyber risk, operational exposure, and innovation goals. Even within a single institution, sovereignty requirements differ by workload. Some demand strict isolation and local control. Others require global scale, advanced security capabilities, and rapid innovation.

Our role is to help customers navigate these tradeoffs deliberately. That means working with them to assess risk, align architecture to policy realities, and design environments that reflect both today’s constraints and tomorrow’s unknowns.

This work sits at the intersection of cybersecurity, compliance, resilience, and frontier transformation. It requires ongoing engagement, transparency, and the willingness to adapt as conditions change.

Digital sovereignty posture in practice

A digital sovereignty posture that is flexible recognizes that no single approach can address every requirement. Instead, it focuses on giving organizations options, visibility, and control across a continuum of environments.

Customers operating in public cloud environments expect clear data residency options, strong encryption and access controls, and visible operational discipline. Just as important, they look for transparency into how cloud systems are governed and how exceptional situations are managed, particularly as regulatory scrutiny increases.

Those expectations do not disappear when workloads move closer to the edge. In fact, they intensify. For workloads that require greater isolation, local processing, or operation in constrained environments, hybrid and disconnected solutions become essential. In February, Microsoft announced the expansion of disconnected operations, enabling customers to run critical workloads in air-gapped environments while retaining consistent governance and operational control. This capability extends cloud-based practices into disconnected settings, supporting operational continuity without abandoning security and innovation. 

That commitment shows up in concrete safeguards that customers can independently evaluate and apply. The EU Data Boundary is one example, supporting data storage and processing within the EU and European Free Trade Association (EFTA) regions for cloud services, alongside longstanding investments in encryption, access controls, auditability, and operational transparency. These measures provide practical mechanisms for aligning cloud operations with regulatory and risk requirements, rather than relying on abstract assurances. 

At the same time, we are expanding options across hybrid and private cloud environments to support continuity, resilience, and local control where required. These investments reflect a simple reality: customer needs are not converging toward one model. They are diversifying.

Underpinning all of this are Microsoft’s digital commitments, which frame how we approach privacy, security, transparency, and responsible AI. These commitments are not marketing statements. They guide how systems are built, operated, and governed, and they provide a foundation for long-term accountability.

Practical guidance for leaders navigating sovereignty

As digital sovereignty becomes embedded in policy and procurement decisions, leaders benefit from a practical lens. Based on what we hear from customers and stakeholders, there are a few consistent themes shaping successful approaches:

• Sovereignty requirements will continue to expand beyond privacy to include continuity, resilience, and AI governance.
• Risk management is now inseparable from digital transformation strategy.
• Flexibility and optionality matter more than rigid architectures.
• Transparency and accountability are as important as technical capability.
• Sovereignty posture must consider protections against cyberthreats.

Addressing these realities requires partners who understand the full scope of the challenge and are willing to engage over the long term. It requires platforms and collaboration designed with sovereignty in mind from the start.

So what does this mean for you?

Digital sovereignty is not a destination. It is an ongoing discipline shaped by changing technology, regulation, and global conditions.

At Microsoft, we approach this work with humility and responsibility. We recognize that customer concerns will continue to evolve, and that our own platforms and practices must evolve with them. We remain committed to expanding our sovereign cloud continuum, strengthening our cloud capabilities, and delivering solutions that balance innovation with control.

Most importantly, we remain focused on delivery. Because in moments of uncertainty, what matters most is not what technology promises, but what it allows organizations to do with confidence.

Where does digital sovereignty go from here?

The future of digital sovereignty will be defined by implementation, not rhetoric. Success will depend on collaboration between governments, industry, and civil society, as well as a shared commitment to transparency and continuous improvement.

As we look ahead, our focus remains on helping organizations turn sovereignty principles into durable, scalable outcomes. That means continuing to invest in capabilities that support trust, engaging constructively with policymakers, and listening closely to the evolving needs of our customers.

Digital trust is built over time, through consistent action and openness, and that trust is one of the most important foundations we can help create.

The post Navigating digital sovereignty at the frontier of transformation appeared first on Microsoft AI Blogs.