AI has moved quickly from possibility to priority, but many leaders are now confronting a more practical question: how ready is the organization to turn isolated wins into repeatable, governed progress? Early pilots and productivity gains matter, but they do not automatically create the leadership alignment, data foundations, governance practices, workforce readiness, or operating discipline required to scale AI across the business.

The AI Strategy Roadmap offers a practical guide for advancing Frontier Transformation. It helps leaders understand where they are today, identify the capabilities they may need next, and strengthen the conditions that can support AI as a durable business capability.

What the AI Strategy Roadmap is

The AI Strategy Roadmap translates insights and patterns observed among organizations advancing Frontier Transformation into practical strategies and actions leaders can consider. Its recommendations are informed by qualitative research conducted by Emerald Research Group on behalf of Microsoft, including in-depth interviews with 70 business and IT decision-makers actively leading AI transformations, as well as Microsoft’s own internal AI transformation experience and the work we’ve done supporting customer transformation efforts across industries.

The Roadmap reflects challenges described by leaders in transformation efforts and provides practical, experience-informed guidance for navigating them.

Why it matters now

This roadmap goes beyond the basic question of where AI can create value. It addresses the organizational conditions that can help teams move from isolated initiatives toward more repeatable, governed progress.

The research surfaced a consistent pattern among the organizations interviewed. Leaders described AI transformation as business transformation that occurs—and improves—over time. Organizations working through the shift often have strong ideas, but can get stuck controlling scope, aligning leaders, managing complexity, and building the workforce capability required to make change happen.

“We didn’t start with a big, fully baked AI strategy. It was much more about identifying a single area where we could realistically make a difference and testing it there first. Once we had something working and could show tangible impact, that’s what helped us expand and bring others along.”

Senior VP of Growth Marketing, Financial Technology

The five drivers of AI value

Every organization is somewhere on the AI readiness journey. Across the research and customer experience reflected in the Roadmap, five core drivers emerged as useful lenses for assessing readiness, prioritizing investment, and planning practical next steps.

1. Business strategy. Align AI investments to measurable business priorities and prioritize use cases that can earn sponsorship, funding, and momentum.
2. Technology and data strategy. Build trusted, governed, AI-ready data foundations and infrastructure designed to support growth, so AI initiatives can become more repeatable over time.
3. AI strategy and experience. Build trust through visible learning, experimentation, and repeatable AI delivery practices that can support user confidence, leadership engagement, and sustained adoption.
4. Organization and culture. Develop AI literacy, change management, leadership alignment, and operating-model practices that can help teams adopt AI more consistently.
5. AI governance and security. Embed responsible AI, governance, security, and compliance from the start, with transparency and human oversight that can help build trust as AI initiatives expand.

Developed in parallel, these drivers can help teams assess readiness, manage risk, and plan the next stage of AI transformation. Each one maps to a maturity model—Exploring, Planning, Implementing, Scaling, and Realizing—so leaders can assess where they are today and identify where to invest next.

Key research highlights

• Culture outweighs code. According to the Microsoft 2026 Work Trend Index, organizational factors drive 67% of the realized value from AI—twice the impact of individual behavior. For many organizations, culture can be a significant barrier to scaling AI initiatives.
• Leadership alignment shows up in the results. In the AI Strategy Roadmap research, more than 70% of organizations with advanced strategy and execution readiness say leadership communicates an AI vision clearly tied to business strategy.
• Direction beats enthusiasm. In the AI Strategy Roadmap research, among organizations that struggled with execution complexity, 15% named a lack of executive buy-in and strategic alignment as a barrier—a reminder that ideas without sponsorship can fragment.

Wherever you are on the journey

Frontier Transformation is a journey, not a destination—and no two organizations start in the same place. By pairing the five readiness drivers with a clear maturity model, the guide helps leaders locate their organization today, identify roadblocks, and focus on practical next steps for moving from experimentation toward more repeatable systems for delivering value at scale.

Start where you are

The next phase of AI transformation will require organizations to understand where they are ready, where they may be exposed, and what capabilities to build next. The AI Strategy Roadmap gives leaders a practical way to assess their current maturity, align teams around the next stage of progress, and strengthen the trusted foundations that can support AI initiatives as they scale.

Download the AI Strategy Roadmap

Discover the key actions leading organizations are taking to scale AI initiatives and advance their journey toward frontier leadership.

Download the roadmap

The AI Strategy Roadmap draws on qualitative research conducted by Emerald Research Group on behalf of Microsoft—70 in-depth interviews with IT leaders and business decision-makers across industries, conducted from February through March 2026—alongside Microsoft’s own Frontier Transformation experience.

The post The AI Strategy Roadmap: Five drivers of successful AI transformation appeared first on The Microsoft Cloud Blog.

This is our idea behind the AI Railroad Brain concept: a network-level intelligence layer that could help rail operators move from localized, reactive decision-making to system-wide orchestration. Instead of treating dispatching, maintenance, safety, workforce planning, and energy optimization as separate problems, it connects them into one operating picture so leaders can make faster, more consistent, and more profitable decisions.

Five freight rail challenges AI can help solve

Across freight rail, five challenges are limiting performance. Each is familiar on its own. What makes them urgent is how they now compound across the network.

1. Reduce network congestion and railcar dwell time

Rail networks lose value when yards, terminals, crews, and equipment are optimized locally instead of as one system. A decision that improves one terminal can create delays downstream. A small disruption can cascade across regions. The result is lower velocity, higher dwell time, underused assets, and constrained capacity.

The issue is not that railroads lack data. They already collect vast operational, asset, and network information. The gap is turning that data into real-time, coordinated decisions that improve the entire network, not just one node at a time.

2. Improve predictive maintenance and asset reliability

Railroads have invested heavily in sensors, wayside detectors, and condition monitoring. These tools can detect anomalies earlier, but detection alone does not improve outcomes. Maintenance teams still need to know which issue matters most, when to act, what resources to assign, and how the decision will affect service.

Without network-aware prioritization, teams can over-maintain some assets, under-respond to others, and miss the broader operational impact. The next step is not simply better prediction. It is better decision-making.

3. Strengthen rail safety and risk management

Rail safety has improved over time, but the risk profile remains asymmetric. Many incidents are low impact, while a small number of high-severity events can create significant operational, regulatory, financial, and reputational consequences.

Most safety systems are built for compliance and response. Modern rail operations require something more: the ability to correlate signals from track conditions, rolling stock, weather, human factors, and network context to anticipate compound risk earlier. The opportunity is to shift from incident response to integrated risk intelligence.

4. Address workforce knowledge loss

As experienced railroad workers retire, organizations risk losing more than labor capacity. They risk losing institutional knowledge: how to diagnose non-obvious failures, manage disruptions under pressure, and balance trade-offs in real time.

That knowledge often lives in people’s experience, not in systems. AI copilots can help capture and scale expertise by giving dispatchers, maintenance planners, safety analysts, and operators access to relevant context, recommended actions, and explanations in natural language. The goal is not to remove people from decisions, but to help every worker make better decisions faster.

5. Optimize fuel efficiency and reduce railroad emissions

Fuel is one of the most direct levers on railroad margin and emissions. Consumption varies by route conditions, train configuration, pacing, congestion, and operator behavior. Yet many fuel optimization approaches remain static, rule-based, and disconnected from real-time network conditions.

AI can help railroads optimize train handling, routing, consist planning, and dwell reduction together. That creates a dual benefit: lower operating cost and measurable progress toward sustainability goals, often without waiting for large-scale infrastructure change.

Building an AI Railroad Brain

The AI Railroad Brain is not a single application but a concept that will help address the challenges highlighted above. It is an intelligence layer that connects operational data, digital models, optimization engines, and human workflows. Its value comes from how these capabilities work together.

• A network digital twin provides a continuously updated view of tracks, yards, rolling stock, crews, terminals, and operating constraints.
• A real-time data platform unifies telemetry, maintenance history, GPS, weather, crew data, customer commitments, and enterprise systems into a shared foundation.
• Decision intelligence uses optimization, prediction, simulation, and machine learning to recommend actions across dispatching, maintenance, safety, workforce, and energy.
• Copilots and agents bring insight into daily work by helping teams ask questions, understand trade-offs, and act with confidence, drawing on the ontology and semantic layer to return answers that reflect rail-specific terms, relationships, and operating context.

Together, these capabilities help railroads move from fragmented insights to system-wide orchestration. A maintenance decision can account for dispatching impact. A safety signal can be evaluated in the context of weather, asset condition, and traffic density. A fuel recommendation can reflect congestion, service commitments, and crew availability. This is where AI becomes operationally meaningful: not by producing more dashboards, but by improving the decisions that shape network performance.

How railroads can start to transform

Railroads do not need to transform everything at once. The most practical path is to start with a high-value operational problem, prove impact, and scale from there.

1. Choose a measurable use case. Start with dwell reduction, maintenance prioritization, safety risk detection, workforce decision support, or fuel optimization.
2. Connect the right data. Focus on the operational signals needed to improve a specific decision, not on building a perfect enterprise data estate first.
3. Embed AI into workflows. Put recommendations where dispatchers, planners, and field teams already work.
4. Redesign decision rights. Clarify when AI recommends, when humans approve, and how outcomes are measured.
5. Scale across domains. Once value is proven, connect adjacent workflows so improvements compound across the network.

Adoption will matter as much as technology. Rail operations are shaped by thousands of daily decisions made by people across shifts, regions, and functions. AI creates value only when it becomes part of how those decisions are made, reviewed, and improved.

The next rail advantage is intelligence

The future of freight rail will not be defined only by more track, more yards, or more equipment. Those investments will remain important, but the next performance frontier is intelligence: the ability to coordinate decisions across the network in real time.

The concept of AI Railroad Brain offers a practical way to unlock more capacity, reliability, safety, workforce resilience, and fuel efficiency from existing infrastructure. For railroad leaders, the question is no longer whether AI will influence operations. It is how quickly they can turn AI from isolated pilots into an operating capability that improves performance every day.

The post The AI Railroad Brain: A new operating model for freight rail appeared first on The Microsoft Cloud Blog.

As we enter a new era for AI, Microsoft’s environmental sustainability work is entering a new phase—defined not only by ambition, but by how we deliver in a period of rapid technological change. In our pursuit of becoming a carbon negative, water positive, and zero waste company that protects ecosystems, the context has evolved, and so must our approach.

The global shift toward AI is reshaping economies, accelerating innovation, and becoming foundational to how technology is built and used. It is also increasing demand for the energy, water, land, and materials required to support that growth. As a company at the forefront of this transition, Microsoft has a responsibility to help ensure that technology strengthens, rather than strains, the systems and communities on which it depends. This imperative is reshaping the context for our work.

We are approaching this moment with clarity and conviction. We believe AI can deliver broad societal, economic, and environmental benefits, but innovation at this scale must be matched by responsibility at the same scale. For Microsoft, this means designing, building, and operating infrastructure that is more efficient, more resilient, and more grounded in the realities of the communities where we operate.

We do not see these dynamics as a reason to step back. We see them as a mandate to lead differently. That requires greater operational rigor, stronger integration across our sustainability priorities, and a sharper focus on durable outcomes for the local communities where we work and the global value chains that make our work possible. It also requires being transparent about where progress is advancing, where it is more difficult, and where new approaches are needed.

The path forward will not be defined by simple tradeoffs or single solutions. It will depend on how effectively we align innovation with stewardship. The systems we build to support the future must also support the long-term health of the planet and the communities we serve. Our experience makes clear that this is possible, but only with even greater discipline, partnership, and a willingness to learn and adapt as conditions evolve.

What this moment requires
Our aim is to build technology that gives more than it uses. Lasting progress depends on how we build it and whether that growth strengthens the places where it takes root.

This thinking is reflected in our Community First AI Infrastructure approach, which is helping shape a more integrated model for community partnership, responsible operations, and environmental performance as we grow. In this way, sustainability is not separate from growth; it is part of how responsible growth is defined.

While AI infrastructure is driving demand for energy, water, land, and materials, sustainability solutions are not scaling fast enough to meet demand. This tension is real, and it is also productive.

It is forcing sharper questions: Where do we need to move faster, invest differently, or rethink our approach? Which assumptions still hold, which ones need to evolve? Five years into this work, we have more operational data, more direct experience, and a clearer view of what measurable planetary progress actually requires. That perspective helps keep us focused on outcomes rather than attached to any single pathway.

We want to be clear about what this means—and what it does not. It means being more precise about what sustainability requires for Microsoft, and more willing to refine our strategies as conditions change, data improves, and tradeoffs become clearer. It does not mean we are lowering our ambition.

Progress amid growth
Our results reflect both progress and pressure. As we scale the physical infrastructure required to power the AI economy, our emissions are shaped by the impact of that growth and the actions we are taking to manage it.

The visual that follows illustrates this dynamic by comparing our reported emissions with a modeled view of where emissions may have been in the absence of four specific interventions: carbon free electricity, sustainable fuels, XBOX console efficiency, and Surface device decarbonization. While these examples represent only a portion of our emissions reduction efforts, they highlight an important lesson from our work to date: that well-designed, targeted interventions can deliver measurable progress even as demand for infrastructure continues to rise.

In FY25, we matched 100% of our annual global electricity consumption with renewable energy[2]. Microsoft will continue to push for an expansive focus on adding all forms of carbon-free electricity (CFE) [3] to the grids where we operate, complementing and building on our portfolio of renewable energy resources. We recognize that the world’s rising electricity needs require a balanced, all-of-the-above decarbonization strategy to meet global economic growth and environmental goals, and we will continue to support this approach moving forward.

Our total emissions (Scopes 1, 2, and 3) increased 25% year over year, driven primarily by the expansion of our datacenter infrastructure and pausing our use of non-additional, unbundled renewable energy certificates as we prioritize investments that bring net new power to grids. While this decision increases our reported emissions in the near term, it enables us to increase the development of new CFE rather than relying on certificates alone. We believe this change will create more long-term sustainability benefits. Growth-related emissions pressure was expected. The more important signal is where that pressure is concentrated.

Scope 3 remains the largest share of our footprint overall, but one of the clearest changes this year was the growing contribution of Scope 2, which represents 13% of our total emissions—up from nearly 2% last year. This development highlights how important the energy systems across our supply chain are in shaping environmental outcomes.

This year’s results also made clear that progress now depends on adapting how we work.

Water is one of the clearest examples. In FY25, we replenished for the first time more water globally than we withdrew—more than 14 million cubic meters—marking a major milestone on our journey to become water positive. Reaching this point reflects years of work to improve water efficiency, expand replenishment efforts, and scale partnerships around the world.

We are proud of this achievement but also know that replenishing global volumes is not enough. The next phase of our work is increasingly local. As we move forward, we are placing greater focus on helping restore more water to the watersheds where we operate than we withdraw while strengthening long-term water resilience. We prioritize projects in water-stressed regions that are locally relevant and designed in partnership with communities, delivering benefits not only for water availability, but also for ecosystems, economies, and people. Through this approach, we aim to ensure our growth supports and helps sustain the communities and environments where we operate.

Transparency remains central to how we work and how we report. Microsoft has eliminated nearly all single-use plastics in our primary product packaging, reducing the share that remained to just 0.07% at the end of calendar year 2025.[4] But we are not rounding down. We are staying accountable to the work required to eliminate them entirely.

Across our cloud operations, we achieved 92% reuse and recycling of decommissioned servers and components for the second consecutive year, diverted 90.5% of construction and demolition waste from landfills and incinerators, and expanded our Circular Centers to seven facilities globally. These results also reflect a broader shift toward solutions that have co-benefits—reducing both emissions and resource demand over time.

Throughout this journey, we have learned that progress in one area often depends on progress in another. Clean energy investments are essential to decarbonization. Water use is linked not only to our operations, but also to the energy systems that power them. And extending hardware life through circular approaches can reduce both emissions and material demand across the value chain.

That is why our priorities extend beyond tracking progress against individual commitments on water, carbon, waste, and ecosystems as though they move independently. Our experience has made clear that progress does not happen pillar by pillar. Some of the most consequential work ahead will be measured in whether we address system challenges and help build the conditions for long-term progress: more resilient grids, stronger markets for lower-carbon materials, more effective water stewardship, and infrastructure designed and operated with local realities and community priorities in mind.

For that reason, this year’s report takes a more integrated approach—placing progress against our commitments in the broader context of how those commitments are operationalized across our infrastructure and products.

What’s next
We are proud of what we have accomplished, and we remain humbled by the scale of the challenge ahead. Responsibly building the AI future requires clear accountability for what AI demands, candor about real constraints and tradeoffs, and sustained focus on outcomes that are durable and broadly shared. The chapters that follow show how we translate that intent into execution across our physical infrastructure, products, and value chain—where our sustainability commitments become operational reality.

Read the full report: https://aka.ms/SustainabilityReport2026

[1] The solid line represents Microsoft’s reported greenhouse gas emissions (Scopes 1, 2, and 3) for FY20–FY25, prepared in accordance with GHG Protocol and management’s criteria, and uses a market-based emissions approach. The dotted line represents an illustrative counterfactual scenario of estimated emissions had select, discrete carbon reduction initiatives not been undertaken. These initiatives include energy efficiency improvements for XBOX consoles, renewable energy purchases, sustainable aviation fuel (SAF) and sustainable marine fuel (SMF) certificates, and supply chain decarbonization of Surface devices. The difference
between the two lines is an estimate of emissions avoided through these specific initiatives relative to a scenario without those initiatives occurring. This estimate is directional in nature, does not represent the full scope of Microsoft’s decarbonization efforts, and is not part of our reported greenhouse gas inventory. It should not be interpreted as a comprehensive measure of total emissions reductions or as additive to other carbon reduction or removal claims.

[2] Microsoft defines renewable energy as electricity that comes from sources that are replenished at a rate greater than or equal to their rate of depletion, such as geothermal, wind, solar, hydro, and biomass. To date, Microsoft’s renewable energy target includes two primary categories: renewable energy from contracted projects and grid mix. The first is renewable energy delivered under PPAs or similar long-term contracting mechanisms, generally for new projects where our financial involvement in the project’s development is critical for its success. This category represents more than 90% of the renewable energy applied to achieve our 2025 target. The second category is “grid mix” – renewable energy supported via our standard utility relationships and rates, inclusive of policy programs such as renewable portfolio standards and state and utility decarbonization goals. Our 2025 100% renewable target does not include purchases from short-term, so-called “spot market” renewable energy credits (RECs) sourced from operational clean energy projects.

[3] Microsoft defines carbon-free electricity (CFE) technologies as technologies with zero direct emissions and biogenic technologies with lifecycle emissions equivalent to renewables. CFE technologies include wind; solar; geothermal; sustainable biomass; hydropower; nuclear; fossil fuels with complete carbon capture, utilization, and sequestration; and storage charged with CFE generation.

[4] By weight, as designed, portfolio average. More details can be found in our Environmental Data Fact Sheet.

The post Responsibly building the AI future appeared first on The Microsoft Cloud Blog.

Summary In the days following the United Nations Micro-, Small, and Medium Enterprises (UNMSME) Day, we take a closer look at what the data shows, and why it matters for the 400 million businesses that power the global economy.

This year, I want to address something we’re watching happen in real time: small and medium businesses (SMBs), also referred to globally as Micro, Small, and Medium Enterprises (MSMEs), are stepping into AI leadership, moving quickly, and applying it directly into the work that drives their business forward.

We mark UNMSME Day to raise public awareness of their contribution to sustainable development and the global economy. The numbers deserve repeating every year: SMBs represent 90% of all businesses worldwide, 50% of global gross domestic product (GDP), and 70% of the world’s workforce.¹

SMBs are not a segment of the economy; they are a foundational part of it. They operate under real pressure. The median small business carries just 27 days of cash reserves.² There is no room to bet on the wrong transformation, they must pick the right direction and get it right the first time. That pressure is exactly why what is happening right now matters so much.

AI has moved from conversation to competitive advantage

The latest Microsoft Work Trend Index 2026 makes something clear: AI is no longer a productivity add-on. It is shifting what work is possible. 58% of AI users say they are already producing work they could not have done a year ago and 66% report spending more time on higher-value work as AI takes on execution.

In a small team, this shift affects capacity and growth. For a financial planning firm streamlining client reviews, a law firm cutting document preparation time, a title company accelerating closings, or a property management team eliminating administrative overhead, that is not a marginal improvement. It represents a structural advantage.

A different pattern is emerging among SMBs. Organizations that move beyond individual use and embed AI into how work happens across teams, workflows, and decisions are seeing step-change results. Not just better work, fundamentally different work. And because SMBs have leaner structures and shorter decision cycles, they can get there faster than large enterprise organizations ever could.

What Frontier Transformation looks like in practice

Frontier Transformation becomes clear when the SMB journey moves from simply experimenting with AI to achieving transformation at growth and scale. Frontier Firms are the ones making that shift real by embedding AI into productivity tools they already are using across their teams workflows and decision processes. The result is not just better output, but a different operating model and one that unlocks more creativity, innovation, and growth.

Here are three SMB’s doing exactly that

Turning bottlenecks into real-time workflows

At Dunaway, a multi-discipline design, planning, and engineering firm in Texas, regulatory research and compliance checks were once manual, time-consuming steps that slowed project delivery. By bringing AI agents into the workflow, engineers can access regulatory insights in real time, answer questions faster, and apply knowledge consistently across teams. The result: a 90% reduction in research time and roughly 10,000 hours saved annually. What once lived with a few experts now works for the whole team. That is what a Frontier Firm looks like in practice.

When I first saw that number, 10,000 hours, I sat with it for a moment. That is not an efficiency gain. That is an entire team’s year given back.

Scaling craft and personalization with trusted intelligence

Businesses built on craftsmanship, trust, and deeply personal service, where consistency is hard to standardize, must tackle a different kind of scale challenge. Chow Tai Fook, a 97-year-old global luxury jewelry brand based in Hong Kong with thousands of stores across Asia, integrated AI across its operations. The company moved from isolated digital efforts to a connected, real-time intelligence model—giving frontline associates the insights they need in the moment to better understand each customer and deliver more personal, relevant experiences at scale. The result is more than 70% efficiency gains across millions of monthly interactions, and a 97-year-old brand that feels personal at every counter.

For me, it always comes back to the customer. When sales associates are empowered with the right insight in the moment, we’re not just making the business more efficient, we’re creating richer, more personal experiences. That’s the power of AI when it’s done right, it doesn’t replace human expertise, it amplifies it.

Making security part of how the business runs

DT Swiss AG, a Swiss manufacturer of high-performance cycling components with teams across Europe, North America, and Asia, faced complexity from fragmented systems, manual compliance processes, and administrative overhead. By moving toward a unified security model, it made identity, access, and governance part of daily operations rather than separate layers of work. The result was a 60% reduction in administrative overhead and a stronger compliance posture. Security did not slow the business down. It made scale more practical.

This is the story I find myself telling most often right now. Security is not a tax on transformation. Done right, it is what makes transformation sustainable.

Across these examples, the pattern is consistent. The SMBs pulling ahead are not simply adopting AI earlier. They are applying it with more intention, moving from isolated use cases to integrated workflows, from individual productivity to team-wide execution, and from security as a separate control to security as the foundation for growth.

Trust is the precondition, not an afterthought

One thing these businesses share: AI adoption and security are unequivocally connected. A 2024 Microsoft Security study found one in three SMBs hit by a cyberattack in the past year, at an average cost of USD254,445. 94% consider cybersecurity critical. And 81% say AI increases the need for stronger controls.³

The businesses moving fastest are solving productivity, data protection, identity, governance, and compliance together. SMBs do not have the time or resources to make five separate technology decisions for one business outcome. Security by design is not a feature, but a foundation for lasting AI adoption.

The partner ecosystem is the multiplier

No SMB transforms alone. Across these customer stories, partners play a consistent role: they help leaders decide where to start, where to incorporate technology into real workflows, and how to support adoption after deployment.

The Microsoft Partner ecosystem brings AI, productivity, and security into one practical conversation. For many SMBs, that begins in Microsoft 365 Copilot supporting how teams create, communicate, and make decisions. Next, extend those workflows using Microsoft Copilot Studio, connect data, and add security tools like Microsoft Defender for Business and Microsoft Purview all working together on a foundation with built-in, secure AI. With more than 1,400 connectors to third-party business applications, these solutions integrate into how businesses already operate across a broader secure cloud foundation.

The Microsoft Partner blog post, “Partner-led momentum, broader availability for SMB: Microsoft 365 Business with Copilot,” has more information on our Microsoft Defender for Business bundles.

• If you are an SMB owner or leader: Start where the time cost is most visible. You do not need a grand transformation plan. You need a first process, a secure foundation, and the decision to act.
• If you are a partner: Almost every SMB conversation is now an AI conversation. Customers are ready. Many still need help knowing where to begin. The partners and Managed Service Providers (MSPs) who lead with outcomes, secure adoption, and real workflow change will be the ones SMBs trust to reach the frontier.

Recognizing UNMSME

I am grateful for the resilience and ambition of small business owners everywhere. I know firsthand, the challenges of operating a business are real. In addition to my role at Microsoft, my husband and I run a small design-build construction company. This experience shapes how I see Frontier Transformation. Together, we are proving that the AI era will not be defined by company size but by leadership. SMBs are leading this moment.

What is the one workflow your team has transformed with AI this year? Connect with me and look forward to the conversation.

¹ United Nations, Micro-, Small and Medium-sized Enterprises Day, June 27, 2026.

² JPMorgan Chase Institute, Cash is King: Flows, Balances, and Buffer Days.

³ Microsoft Security, New research: Small and medium business (SMB) cyberattacks are frequent and costly, 2024.

The post Small and medium businesses aren’t waiting for an AI invitation—they’re already leading appeared first on The Microsoft Cloud Blog.

The more time I spend working with AI, the more convinced I become that what matters most for her future isn’t how quickly she can answer. It’s whether she has the judgment to know when an answer can be trusted.

I’ve spent decades at Microsoft watching this tension play out: first building tools for other developers, then working across AI as models moved from research curiosities to systems deployed at scale. Now we’re building Microsoft IQ, where we’re exploring how an organization’s collective intelligence can become its greatest advantage. Through every one of those chapters, one thing has remained true: it’s never enough for a system to be powerful; it must also be trustworthy.

Trust is what turns assistance into delegation. When we can trust an agent to do what we intend, within the limits we set, we can hand off the work we never wanted to spend our lives on: the repetitive tasks that drain attention, the mundane work that fills a day without moving anything meaningful forward, the dangerous work humans should not have to do, the work too vast for any individual or team. Agents should take on that toil, extend our reach, and give us back our time for the work that calls for something only humans bring.

My daughter doesn’t know any of this yet. But by the time she’s grown, most of the work that rewards speed and repetition will be work we delegate. What will matter then is exactly what gave her teachers pause: the patience to stay with a hard problem, reason through it, and decide when she’s reached a conclusion she can trust. The very thing they feared might hold her back could be exactly what the next era prizes most.

So no, I’m not worried about the timed test. I hope she grows up in a world where software carries the toil and people are freed for the work that is unmistakably ours—to think, to judge, to create, to care for one another. That is the future I want agents to make real. But my hope is not evidence it will happen. The future I just described depends on a single question: can we trust agents to do the work? Trust is earned one task at a time. So, I went looking for evidence of where it’s been earned, and where it hasn’t.

We partnered with MIT Technology Review Insights on new research that draws directly from the technical leaders building this frontier: not the people talking about it, but the people doing it. We surveyed 300 technical experts across AI, data, and cloud domains, spanning 12 industries and 4 regions of the world, asking them to rank their confidence across 101 of the top tasks. What we got back is the 2026 Agent Confidence Index, an honest map of where agents are delivering real value, so our community can see what’s working and move forward together with conviction.

Learn from where confidence is highest

Across the 101 tasks measured, average confidence already lands at 64 out of 100, and thirty tasks clear 70. The highest scores cluster on work that is both predictable and draining: the late nights, the interruptions, the low-value repetition. Automated report generation leads at 83.5. Boilerplate code generation for new features sits at 82.5—the hours a developer no longer spends rewriting the same patterns, freed for the work that challenges them. Certificate expiration monitoring and renewal, at 81.5, ends the scramble that pulls engineers off high-stakes problems for something entirely routine. Real-time data stream monitoring follows at 80.5, and release note generation from commit history at 79.5—the manual end-of-sprint commit review, gone. This is where frontier teams are already delegating to agents, regularly.

The pattern holds across every discipline. In developer and AI workflows it extends to API client maintenance and code identification; in cloud operations, to ticket routing and cost optimization; in data, to anomaly detection. Wherever it sits in the stack, this is work technical teams now trust agents to own.

What matters most here isn’t what the data says about the tasks; it’s what it says about the people delegating them. When technical experts believe in something deeply enough to hand it real work, that belief ripples outward. It becomes the recommendation they make to their leadership, the solution they build for their customers, and the culture they create for their teams.

Even the toughest agent tasks are gaining traction

Here’s what strikes me most: the tasks ranked lower on the index are still high in absolute terms. Service mesh configuration and troubleshooting sits at 37.5, database schema migration scripting at 46.5, memory leak detection at 48.5. These sit at the very frontier, the interconnected, high-stakes work where investment and innovation are concentrated right now.

Consider what they demand. Service mesh configuration touches many systems at once. Database migration carries real stakes, requiring precision across data, application, and infrastructure layers at the same time. Memory leak detection means diving deep into a system’s behavior under load, accounting for conditions that shift from one deployment to the next. These are the challenges that have separated great engineers from exceptional ones—and even here, experts see agents helping. Not carrying the work alone, but contributing where it used to be unthinkable. That confidence is still climbing, and that’s telling.

We’re shipping new capabilities constantly to support this momentum. Database migration tooling in GitHub Copilot now covers not just scripts but the full application and infrastructure migration story. The Azure Site Reliability Engineering (SRE) Agent brings decades of experience operating Azure at scale and deep profiling capabilities directly into memory analysis and performance diagnosis.

Why human judgment remains paramount

When we asked technical experts how they’re navigating agent adoption, 59% named “keeping humans in the loop” as their top priority—ahead of better observability, ahead of governance documentation, and ahead of everything else. That’s a mark of maturity. Teams moving forward with clarity treat agent oversight as non-negotiable, regardless of how capabilities evolve.

The boundary itself is straightforward. Agents excel at well-specified, high-volume, reversible work: they synthesize data, automate known workflows, and surface anomalies at a speed and scale no human team could match. The moment a decision becomes high-stakes, context-dependent, or hard to undo, a human signs off. That isn’t a limitation of the technology; it’s the architecture of a trustworthy system.

What’s changing, and what remains underappreciated, is the skill it takes to draw that boundary well: the discipline of full-lifecycle evaluations and guardrails. Success means measuring agent output against intent and keeping behavior inside your business strategy. It’s new territory for most engineering teams, and it’s becoming table stakes for modern software faster than most organizations realize. The good news: the same tools generating the work can help you build the harness. Ask GitHub Copilot to write the evals and it will. Frontier teams are already doing this, and it’s why they’re pulling ahead.

Agents are opening career doors for engineering

Across system reliability and site operations, evaluations and quality assurance, and data pipeline management, 80% or more of respondents see meaningful career opportunity ahead. We believe this is one of the most significant moments in the history of building software, not because agents replace what technical people do, but because what’s left when they take on the toil is the work that defines a career: the judgment calls, the architectural vision, the reasoning to navigate complexity under pressure. That fluency will define the next generation of technical leadership.

We’re living this shift at Microsoft, right alongside our customers. Junior developers are using agents to explore codebases on their own and arriving at mentoring conversations with sharper, more sophisticated questions. Senior engineers are covering more ground because the repetitive work that used to fill their days is now delegated, and the work that’s left is harder, interesting, and consequential. Both are growing into more capable versions of themselves. For me, that’s the outcome I’ve always believed technology could deliver.

An integrated approach to intelligence and trust

Designing more sophisticated agent systems has made one thing clear: agents thrive in well-integrated environments, working best when your whole stack draws on a single source of truth. The high-confidence tasks are the ones we’ve already figured out; the meaningful frontier is the harder, interconnected work, and that’s exactly where observability, governance, security, and unified intelligence have to operate as one.

Microsoft IQ brings your enterprise context into a single, continuous intelligence layer. Within it, Work IQ builds semantic understanding of how your business operates across email, calendar, meetings, chats, files, people, and collaboration patterns. Such depth of knowledge is the reason technical teams choose us, and it’s what drives my focus and passion in learning how people actually work so their agents get them. My colleague Kim Manis, CVP of Product for Microsoft Fabric, has written specifically about what this means for data professionals, and the integral role of Fabric IQ.

It’s all part of the Microsoft Agent Platform, which is becoming the operating system for enterprise AI at scale. From building in GitHub and contextualizing with Microsoft IQ, to running in Microsoft Foundry and governing in Microsoft Agent 365, Microsoft is uniquely positioned to help customers bring together data, models, agents, and human judgment into a continuously improving and secure system.

Frontier transformation is being led by builders like you.

Next steps:

• Download The 2026 Agent Confidence Index from our partners at MIT Technology Review Insights. It is a free, ungated deep dive into all 101 tasks, broken out by role and workflow, with the patterns and reasoning behind where confidence is strongest and the frontier is expanding.

• Join us at the AI Engineering World’s Fair (June 29-July 2) where our very own Pablo Castro will keynote, and our teams will offer 16 breakout sessions and 4 labs. Swing by the Microsoft booth as well to explore an interactive 3D visualization of the Index data. We want to hear what’s working for you.

• Learn more about Microsoft IQ and how it connects across Work IQ, Fabric IQ, Foundry IQ, and the newly announced Web IQ. You can catch up on all the developer innovation from Microsoft Build through Satya Nadella’s keynote, Kyle Daigle’s blog post, and the Microsoft Build CLI.

What’s Working in Agentic AI

The 2026 Agent Confidence Index report reveals where agents are trusted, the challenges they face, and what leaders should do next

Download the 2026 Agent Confidence Index

The post The 2026 Agent Confidence Index: Where 300 builders see real momentum appeared first on The Microsoft Cloud Blog.

The shift is already underway. Major financial institutions are quickly moving beyond limited use cases toward agentic AI solutions designed to help drive impact at scale. Frontier Firms—organizations reimagining business processes around human-agent collaboration—are integrating intelligence into the workflows, data environments, and governed systems where decisions are made and work gets done.

The challenge is that in the heavily regulated constraints of financial services, access to data is tightly controlled, critical data is often fragmented across legacy platforms, and workflows span multiple systems, teams, and jurisdictions. Internal data must be integrated with external data, including market data, research, and third-party insights. Firms must be certain that the right people can access the right data in the right ways, while keeping it all safe.

Our partnership with Microsoft is shaping the next era of AI ready data and intelligence‑driven workflows. We are meeting clients where they work—embedding S&P Global’s trusted, high‑quality data directly into their workflows and unlocking agentic capabilities that turn insight into action, enabling accurate, faster, more-informed decision‑making. Together, we are accelerating how clients drive growth, manage risk, and capture opportunity in an increasingly complex and fast‑moving market.

Sally Moore, Chief Client Officer, S&P Global

Trust is paramount, and the best way to scale AI is to leverage the inherent strengths and attributes of a trusted cloud environment and the applications that professionals use to do their work. This way, data used in agentic solutions can be configured to inherit existing permissions, licensing constraints, and governance controls, rather than requiring firms to rebuild trust around new layers.

This requires viewing AI as an operating capability that unifies data, governance, tools, and workflows, enabling intelligence to show up where work happens. Approaches that simplify access to critical data and embed AI into everyday decisions and actions are best positioned to succeed.

Putting AI at the point of work—where data, decisions, and action come together

Financial work rarely happens in a single system. Professionals move across data, spreadsheets, meetings, messages, and business applications, in scenarios that span all aspects of financial services. To name just a few:

• A banker preparing for a client meeting needs more than a market summary; they need relevant signals, prior context, comparable activity, potential risks, and a path to the next best action.
• A risk professional needs intelligence that can be applied within existing controls, not outside them.
• An investment analyst needs current financial data in the spreadsheet where assumptions are being tested, not a separate answer that has to be copied and reconciled later.

AI creates more value when it is embedded in workflows across the value chain, helping to reduce handoffs, preserve context, and close the gap between analysis and action. AI that sits apart from the workflow can generate useful responses, but it still requires that data be copied, reconciled, and validated across systems.

The next phase of AI in financial services is about connected intelligence that brings the right data into the right context at the right moment. Our connectors ensure Morningstar’s independent research and trusted data sit alongside firms’ own data within their workflows, enabling investment professionals to generate deeper insights, move with greater speed, and make more confident decisions with clarity and control across the entire investment process.

Adam Wheat, Chief Technology Officer for Direct Platform, Morningstar

Our focus is to narrow the distance between trusted information, professional judgment, and the action that follows—in other words, to enable governed AI execution in context, allowing people to move from insight to action within familiar tools. This is why we, and our partners, are bringing financial data and AI capabilities directly into the everyday applications and tools many professionals use to get work done, including Microsoft Excel, Teams, Outlook, and Microsoft 365 Copilot. It enables intelligence that can be accessed in the spreadsheet where analysis is being built, in the meeting where a client discussion is being prepared, or in the collaborative space where teams coordinate the next step.

The platform for advanced AI experiences

For financial institutions, the next model for AI begins with the data that anchors the business: market intelligence, risk signals, research, client information, transaction context, policies, and operational knowledge.

Microsoft’s approach brings together trusted data, enterprise context, and intelligence on one integrated platform. It spans a rich ecosystem of financial services data providers, helps to orchestrate models, and is built upon the core strengths of the Microsoft Cloud: enterprise-grade security, compliance, and governance.

In this approach, the following key features work together to connect trusted data, enterprise context, and AI-powered capabilities:

• Federated connectors provide live data by accessing information without moving or copying it.
• Synced connectors provide enterprise data by indexing content from business and partner systems.
• Microsoft IQ provides context by understanding how people work, how the business operates, what the organization knows, and how the broader business and market environment is changing.
• Skills provide expertise by defining how specific business tasks are performed.
• Plugins combine connectors and skills to package data access and task expertise into reusable capabilities.
• Copilot Cowork helps coordinate and assist with tasks across systems, people, and workflows—bringing agentic capabilities into Microsoft 365 Copilot and, by extension, into Excel, Word, PowerPoint, and Outlook.

Together, these features empower financial services professionals to ask more precise questions, receive more relevant answers, and take action directly within their workflows, all while supporting compliance efforts and helping preserve context.

Expanding the ecosystem of financial data providers

In recent months, we’ve unveiled this strategy through milestones that reveal an expanding ecosystem of financial data providers and capabilities.

Federated Copilot connectors extend Copilot’s reach by securely integrating external real-time data directly into Copilot in Excel, Copilot Chat, and Researcher agent. Recently, we announced federated Copilot connectors from LSEG and Moody’s, and the momentum continues today with new federated Copilot connectors from CB Insights, Daloopa, FactSet (in preview), Morningstar, PitchBook, and S&P Global. Together, these integrations can provide access to up-to-date market data, company intelligence, research, portfolio analytics, investment data and more—within supported workflows, subject to provider availability, licensing, and system integration.

As financial institutions scale AI, the combination of trusted data and enterprise platforms becomes critical. Our collaboration with Microsoft enables clients to bring LSEG’s data into AI-powered workflows, supporting more consistent, informed decision-making across their business.

Emily Prince, Group Head of Enterprise AI, LSEG

Copilot Cowork plugins embed specialized partner expertise directly into task-oriented workflows, enabling Copilot to support role-specific financial analysis and decision-making. Building on the recent launch of Copilot Cowork plugins from LSEG and S&P Global Energy, we’ve now expanded the list of partners with Copilot Cowork plugins from CB Insights, Moody’s, Morningstar, and PitchBook. These plugins package proprietary data, models, and domain knowledge within defined workflows, depending on partner capabilities and implementation, allowing financial services professionals to move more readily from insight to action.

The defining question for financial institutions today isn’t whether to use AI, it’s whether they can trust the intelligence powering it. From Microsoft 365 Copilot to Excel to Copilot Cowork, Moody’s is embedding its decision-grade connected intelligence directly into the Microsoft workflows where consequential decisions are made every single day, allowing our customers to move with greater confidence and speed without sacrificing the rigor, transparency and accountability that high-stakes decisions demand. 

Cristina Pieretti, Head of Digital Content and Innovation, Moody’s

We’ve also announced new skills and capabilities for Copilot in Excel, one of the most widely used tools in financial services. In addition, we’ve introduced a starter library of pre-built finance skills, alongside the ability for organizations to create and integrate their own. Combined with new customization and control features, these enhancements enable financial services professionals to customize their workflows, apply consistent logic, and scale analysis more efficiently.

This integrated platform supports a shift from information retrieval to governed execution. For firms and institutions, this can mean more efficient preparation, more consistent use of trusted information, less manual stitching across applications, and a stronger connection between insight and execution.

Trust as the foundation

For financial institutions, this shift can only succeed if trust is built into the architecture. Organizations must be able to innovate without separating intelligence from the controls that make enterprise adoption possible.

This is where Agent 365 makes the trust model tangible. It lets organizations see which agents are operating, understand what data they are using, monitor the actions they take, and apply policies that govern how they work. It also supports the traceability and citations institutions require to understand how data was used, what sources informed an output, and how agent-driven work was generated.

As agentic workflows become the new standard for how analysis gets done, the data underneath must be trustworthy and verifiable. Through Kensho’s AI innovation and S&P Global’s domain expertise, we’ve done that work, creating a trusted data retrieval foundation designed for how AI systems reason and act. Our partnership with Microsoft is about bringing that foundation into the AI-native experiences where the industry is already headed.

Bhavesh Dayalji, Chief AI Officer, S&P Global and CEO, S&P Kensho

The same environment that brings data and AI closer to work should inherently provide the capabilities required to use agents responsibly at scale. In regulated industries, this combination is a material advantage: firms should not have to choose between moving faster with AI versus maintaining control over how data and actions are managed.

For financial services leaders, the next phase of AI will not be defined by isolated experiments, disconnected tools, or AI architectures that require firms to add a separate layer of trust, permissions, and context on top of the systems they already use. As firms move from experimentation to scaled adoption, Microsoft’s advantage is the ability to combine trusted data, context, workflow integration, and governance in one coherent operating model for AI, allowing firms to leverage the platforms they already depend on.

Learn more

• Learn how agentic AI is rearchitecting asset management at scale.
• For a practical framework on using AI to drive coordinated, enterprise-scale impact across the firm, read Frontier Transformation in Capital Markets.
• Visit Microsoft for Financial Services to learn more about how Microsoft empowers financial services organizations to achieve more.

The post AI in financial services: Bringing trusted data into the flow of work appeared first on The Microsoft Cloud Blog.

With the global banking system intermediating hundreds of trillions of dollars in funds, corporate and commercial banking sits at the center of the global economy.¹ It spans geographies, regulatory regimes, client types, and balance sheet structures. And critically, it relies on decision‑making that unfolds across extended, non‑linear processes.

Where past transformation efforts often fell short in accelerating decision-making and streamlining cumbersome processes, the emergence of agentic AI is changing what is possible. Combined with enterprise‑grade platforms, new AI capabilities are being designed to help coordinate work securely, governably, and at scale.

This shift is best understood through four ways that AI is starting to reshape how corporate and commercial banking can operate.

1. Empowering relationship managers

Relationship managers have long been the lifeblood of positive client experiences, often serving as the trusted human face of the business. In this role, the ability to synthesize insights from complexity is a critical skill. Yet even the best managers often struggle to manage the growing sprawl of data and requirements.

In many cases, relationship managers function as human middleware—responsible for manually aligning information that the organization itself cannot easily connect as environments become more fragmented across treasury and credit/lending systems, client data, risk platforms, documents, and collaboration across channels.

Agentic AI addresses this by shifting from passive or responsive assistance to active work coordination. AI-powered agents can assist in monitoring where a client, deal, or request sits across multiple systems, identify what is missing, and surface the right context at the moment decisions are made.

Standard Chartered equipped more than 6,000 bankers with a unified platform spanning 53 markets, giving relationship managers real-time insights and more time to spend on client-centered engagement. Likewise, UBS deployed Microsoft Copilot across its employee base (including relationship managers), transforming legal research with an AI assistant that surfaces precise clauses across 26 million documents using natural language queries. The system eliminates manual search and accelerates information retrieval, freeing experts to focus on judgment-intensive work and giving relationship managers more time for client interactions.

The result is not “AI advice,” but improved decision readiness for teams. Relationship managers spend less time assembling inputs and more time applying expertise, consistently, audibly, and earlier in the process.

2. Improving the quality of client interactions

Clients do not experience the bank as a set of systems. They experience it as a series of moments across onboarding, credit, treasury, and servicing. Too often, these moments are disconnected, which creates friction and can erode confidence.

Agentic AI helps close these gaps not just by connecting interactions, but by making them more meaningful. AI agents enable financial professionals to understand the client in context by maintaining continuity across all channels, while also helping anticipate needs, surface relevant insights, and making sure requests move forward without repeated inputs or dropped handoffs.

Commerzbank, for example, built an AI agent that now handles more than 30,000 customer conversations per month, resolving approximately 75% of requests autonomously. By maintaining shared context across interactions and orchestrating workflows in real time, the bank reduces the need for customers to repeat information while ensuring requests move forward without disruption. The result is a more consistent, responsive experience at scale.

AI also helps bankers maintain continuity across interactions and generate consistent, empathetic responses by unifying previously siloed workflows in sales, service, and communications. First National Bank is using Copilot for Sales to reduce fragmented touchpoints and strengthen ongoing, relationship-driven engagement with commercial clients across channels and teams.

This extends across other complex and high-stakes moments in banking. In investment banking, for example, AI-assisted meeting preparation pulls together internal context and external market data, so bankers walk into conversations already aligned to the client’s situation. In corporate and commercial banking, onboarding and KYC agents can intelligently scan documents, cross-reference sanctions lists and adverse media, and surface key findings for relationship managers.

This can turn what has traditionally been a weeks-long, manually intensive process into a more efficient and consistent experience for both the bank and the client. 

The result is a shift from fragmented, episodic service to coordinated, proactive engagement, where interactions are more timely, more relevant, and more aligned to the client’s evolving objectives.

3. Modernizing risk and core systems

Banking operations have rarely struggled because of a lack of automation alone. They often struggle because complexity compounds.

Credit assessment depends on risk inputs, risk depends on documentation, documentation depends on counterparties, onboarding depends on compliance and core systems rely on all of the above. These processes span teams, systems, and timelines that were often not designed to operate in sync, and as a result work can slow down, fragment, and stall.

Agentic AI helps address these challenges by not only automating tasks but also coordinating workflows across stages. It can track progress across processes, resolve handoffs dynamically, and surface exceptions to the right experts when judgment is required, reducing friction by connecting workflows, aligning dependencies, and helping work move more smoothly across systems and teams.

Bank of Queensland, for example, used Microsoft Copilot to streamline complex, multi-step workflows, reducing risk analysis from weeks to a single day while improving quality by 22%. By automating document-intensive tasks and simplifying cross-functional coordination, teams accelerated decision-making and enabled more adaptive, efficient execution across core banking processes.

At an ecosystem level, Swift applied federated learning and confidential computing to detect financial crime across institutions without centralizing data, which enabled more coordinated risk modeling while preserving privacy and regulatory boundaries.

At Scotiabank, rather than replacing core systems, the bank embedded AI agents into its payment operations to coordinate data transformation, reconciliation, and exception handling across fragmented workflows. As a result, processes that once required weeks can now be executed in seconds, with clear auditability and control.²

When processes are streamlined and coordination costs fall, measurable outcomes can improve. Faster credit decisions, strengthened compliance, improved reconciliation and materially better banker and borrower experiences are enabled, without requiring wholesale system replacement.

4. Transforming trade finance

Innovation in corporate and commercial banking typically falls short because complexity has historically resisted scale.

In trade finance, structured lending, and cross-border treasury, work spans multiple stakeholders, documents, and exceptions across boundaries. These are not linear processes, but complex, judgment-driven environments where traditional automation can break down.

The shift enabled by agentic AI is reimagining how these processes operate. Instead of forcing complexity into fixed workflows, banks can execute work adaptively, coordinating across participants in real time and responding to exceptions as they arise.

Agentic systems open the door to different operating models. They can address deviations, help route actions to the right specialists, and capture how complex work is completed. Over time, these patterns can be encoded into reusable playbooks, so banks can scale expertise, deliver more consistent outcomes, and maintain continuity across complex, multi-party processes.

A glimpse of this future can be seen in a Microsoft-led trade finance proof of concept with ANZ, HSBC, and Lloyds. By embedding AI agents within ERP systems, the solution parses letters of credit, cross-checks them against invoice and shipping data, flags discrepancies, and securely transmits structured, standards-aligned information to bank platforms. Rather than simply accelerating existing steps, it demonstrates how trade workflows can evolve toward a more consistent, data-driven model, ultimately reducing fragmentation, minimizing manual rekeying, and improving traceability across trade finance processes.

What emerges here is not just more efficient execution, but a potentially fundamental new way of operating, one in which complex, exception-heavy processes can be coordinated, governed, and continuously improved at scale.

Realizing this shift depends on enterprise-grade AI platforms that combine advanced AI models, secure integration across systems and partners, and scalable data and analytics foundations.

Leading institutions are using GenAI to streamline trade finance and drive the shift from paper-based to digital platforms.

From experimentation to operating advantage

Experienced banking leaders are right to approach AI with caution. The industry has seen many moments where enthusiasm outpaced operational reality. What differentiates this phase is not model capability alone, but fit. Agentic AI aligns with how corporate and commercial banking functions. It streamlines key processes, respects variability, reacts to change, and reduces coordination costs rather than oversimplifying work.

Equally important, platforms matter. Without security, governance, and integration at enterprise scale, AI coordination becomes a risk rather than an advantage.

This is where disciplined adoption—not experimentation for its own sake—will separate leaders from laggards.

Learn more

• For a practical playbook to improve customer experiences, embed AI in high‑value workflows, and act on the shift to agentic AI, read The Microsoft Business Blueprint for Agentic Assistants.
• For a practical blueprint to scale AI, drive productivity, and deploy across banking, insurance, and capital markets, read Frontier Transformation: AI in Financial Services.
• To help unlock business value and deepen customer relationships in the era of AI, visit Microsoft for Financial Services.

¹ McKinsey & Company, “Global Banking Annual Review 2025,” October 23, 2025.

² Scotiabank, “How AI Agents are Transforming Scotiabank’s Payment Operations,” September 17, 2025.

The post How AI is reshaping corporate and commercial banking appeared first on The Microsoft Cloud Blog.

Yet a gap remains. While leaders have embraced AI’s potential, much of that energy is still concentrated in isolated use cases. The result is progress that feels real but rarely scales. Efficiency improves in pockets. Insights surface in moments. But enterprise impact remains limited. Frontier Transformation begins where this pattern breaks.

It signals a shift in how organizations think about AI—from something applied to specific tasks to something embedded across the business. AI moves into the flow of work, shaping decisions, powering processes, and enabling entirely new ways of operating. This shift is not about doing the same work faster. It is about redefining what work can be.

The organizations leading this transition are distinguished not by how much AI they deploy, but by how deliberately they align it to outcomes that matter. They focus on where AI can unlock new forms of value—across employees, customers, operations, and innovation. And they build the conditions for that value to scale, grounded in both intelligence and trust.

For business decision makers, this moment requires a different lens. The question is no longer where AI can save time. It is where AI can change the trajectory of the business.

The four paths to business value

While every organization’s journey looks different, leading organizations are converging on four core areas where AI drives meaningful impact.

These four paths define how AI moves from experimentation to enterprise value.

1. Enrich employee experiences

AI is transforming how work happens.

When intelligence is embedded into the tools employees already use, it reduces friction and elevates contribution. People spend less time searching for information or coordinating tasks and more time applying judgment, creativity, and expertise.

For example, organizations are using AI-powered knowledge hubs to surface institutional knowledge from documents, meetings, and media—making it easier for employees to access and apply critical information in real time.

This shift improves not only productivity but also decision quality, enabling employees to act faster and with greater confidence.

2. Reinvent customer engagement

Customer expectations continue to rise, but traditional engagement models struggle to keep pace.

AI enables organizations to deliver faster, more relevant, and more personalized interactions across channels. It can respond instantly to routine inquiries, connect customers to specialized expertise, and generate tailored recommendations in context.

In practice, this shift is already reshaping front-line experiences. AI-powered systems can eliminate wait times for common requests while routing more complex issues to the right experts with full context, improving both customer satisfaction and employee efficiency.

For example, Alaska Airlines created a natural language destination discovery experience that helps travelers find and book trips more intuitively. The result was 90% user satisfaction and 75% less planning time, showing how AI can make customer engagement both more personal and more efficient.

3. Reshape business processes

AI’s greatest potential lies in rethinking how work gets done.

Instead of optimizing individual steps, organizations can redesign entire workflows, accelerating execution and improving outcomes. Companies applying AI in this way are already seeing measurable gains in speed, efficiency, and scalability.

In some cases, organizations have reported dramatic results, such as significant reductions in time spent searching for business-critical information and the ability to scale complex analysis without increasing headcount.

These kinds of gains illustrate how AI enables new operating models rather than incremental improvements.

4. Bend the curve on innovation

AI expands what organizations can create and achieve.

Consider how organizations are using AI to integrate vast, distributed datasets or analyze unstructured content—such as interviews and videos—to unlock new insights. This capability is accelerating how quickly teams can experiment, learn, and bring new ideas to market.

For example, Space Intelligence used Microsoft AI capabilities to accelerate large-scale forest mapping—reducing the time required to map global forests by 75% while scaling coverage to billions of hectares.

When innovation becomes faster, more accessible, and more repeatable, it begins to compound across the organization.

Together, these four paths show how AI evolves from isolated initiatives into a driver of sustained business growth.

Why intelligence and trust matter

As organizations scale AI, a familiar challenge emerges: complexity increases, data becomes fragmented, and systems grow increasingly disconnected. As a result, early gains become harder to sustain.

The difference between organizations that stall and those that scale comes down to how they build their foundation.

At Microsoft, we see two elements as essential.

Intelligence ensures AI is grounded in real work—connecting data, workflows, and business context so outputs are relevant and actionable.

Trust ensures AI can scale safely—embedding security, governance, and responsible AI practices from the start so organizations can innovate with confidence.

These elements reinforce each other: intelligence drives value, and trust enables that value to scale. Together, they transform AI from a set of tools into a durable enterprise capability.

What BDMs should do next

For business decision makers, the priority is not adopting more AI. It is realizing more value from it.

Leaders seeing the greatest impact focus on a few consistent moves:

• Start with clear business outcomes where AI can deliver measurable impact.
• Demonstrate value early through focused deployments that build confidence.
• Scale through repeatable systems that extend success across the organization.

This approach helps organizations move from pilots to platforms—and from isolated results to enterprise impact.

Moving forward

Frontier Transformation is already underway. The opportunity now is to move beyond isolated gains and use AI to reshape how the business creates value.

To learn more, read the e-book Four Paths to Business Value with AI and explore how these paths can accelerate your organization’s journey.

The post 4 paths to Frontier Transformation: From AI experimentation to real business value appeared first on The Microsoft Cloud Blog.

Safety, quality, energy, maintenance, production targets, and supply volatility now collide in real time. In many plants, the response has been more dashboards, more alerts, and more analysis. But what manufacturers need next is intelligence embedded in daily work—systems that help teams understand what matters, decide faster, and act with confidence while the plant is running. 

That is the promise of agentic AI for plant operations: AI that works alongside people, grounded in industrial context and governed by operational guardrails. At its best, it does more than explain what happened—it helps teams determine the next best step and move from insight to action in the flow of work. 

What agentic AI for plant operations means in practice

In process manufacturing, agentic AI cannot mean black-box autonomy. Plants run on physics, safety standards, and regulatory requirements that do not bend. In practice, agentic AI means human-agent teams: systems that observe, reason, and recommend—and in some cases initiate workflow steps—with the right approvals and guardrails in place. 

• Bring operational and engineering context together: Connect OT data, engineering documents, maintenance history, alarms, work orders, and shift logs so teams can see what is happening in the right plant context. 
• Guide decisions within real plant constraints: Recommendations need to reflect safety, quality, process limits, and operating conditions so they can be trusted in production. 
• Help the right teams act faster: The value is not just insight—it is helping operations, maintenance, reliability, and quality teams move on the next step with less delay and less guesswork. 
• Capture what the best operators already know: Over time, agentic systems should help turn tribal knowledge into repeatable practices that scale across shifts and sites. 

This matters most in high-pressure moments—such as troubleshooting, abnormal situation management, reliability planning, outage recovery, energy optimization, and quality response—where minutes matter and context is often scattered across systems. In plant operations, AI only matters when it helps teams act faster, safer, and with more confidence. 

What manufacturing leaders are prioritizing in agentic AI

That is why the conversation with manufacturers is changing. At Hannover Messe and ManuChem, leaders were no longer asking for definitions of generative AI. They were asking a harder question: where can this create measurable value in a live operating environment? Across those discussions, six market signals came through consistently. 

1. Focus on outcomes and guardrails. The strongest conversations were not about abstract AI potential. They were about measurable operational value and a governance model where humans remain accountable for judgment, escalation, and approvals. 
2. Start with operational friction, not architecture. The best use cases begin with real pain points—lost time finding the right drawing, inconsistent troubleshooting across shifts, slow outage recovery, or low trust in legacy system data. Technology matters, but the entry point should be a business problem that teams already feel. 
3. Brownfield reality is the default. Most process manufacturers are not starting from scratch. Winning approaches respect legacy systems, edge constraints, and the need to modernize without stopping the plant—making a flexible operating model across edge, cloud, and AI critical. 
4. Trust is the adoption lever. In industrial environments, trust goes beyond compliance—teams need identity, access control, auditability, lineage, and explainability. If a recommendation cannot be traced and bounded, it will not be used in production. 
5. Design for scale across shifts and sites. The most valuable scenarios are often not the flashiest demos—they are the ones that reduce variability, capture institutional knowledge, and help teams work more consistently from one shift, line, or site to the next. 
6. Sovereignty and ontologies are rising priorities. Customers increasingly care where data resides, who can access it, and how policy applies across environments. At the same time, semantic models and ontologies are becoming critical for connecting operational, engineering, and business context—enabling AI to reason more effectively. 

One story that captures these themes well is the work Yara and Kongsberg Digital have done together. 

Real-world agentic AI: How Yara and Kongsberg Digital built an operational intelligence layer

Yara’s Porsgrunn fertilizer plant offers a good example of what this looks like in practice. The challenge was familiar: critical plant data existed, but it was spread across systems and difficult to use in real time. Engineering documents, 3D models, maintenance history, and live operational information were not coming together fast enough to support troubleshooting and day-to-day decision-making. The result was slower response times and more field work than necessary. 

Working with Kongsberg Digital, Yara built an Azure-based operational digital twin experience that brought these sources together into a usable operating environment. That matters because, in process manufacturing, better decisions rarely come from more data alone—they come from putting the right context in front of the right people at the right moment. The goal is not more intelligence in the system—it is better decisions on the plant floor. Once that foundation is in place, copilots and agents can do something practical: help teams understand situations faster, guide the next action, and reduce the time lost moving between disconnected tools. 

The business impact is what matters—and Yara’s results are a strong signal of what becomes possible when you remove operational friction: 

• Up to 70% faster shutdown recovery by consolidating operations onto a unified, actionable platform. 
• 60% fewer field trips through improved remote visibility and diagnostics. 
• 50% efficiency gains in targeted engineering tasks—helping teams resolve issues faster and operate with greater precision. 

Access to the right data at the right time is how we define success, and our digital twin gets us there. We’re building something that can scale safely and deliver value over time.

Roar Nilsen, Program Manager, Digital Engineering, Yara 

The broader lesson is important. Before AI can guide work in a plant, teams need a trusted operational foundation that connects data, documents, assets, and process knowledge in the flow of work. Yara’s results show what happens when that foundation is in place: operational knowledge becomes easier to access, decisions become faster, and expertise scales more effectively across the organization. For manufacturers asking what to do next, the implication is straightforward. 

A pragmatic playbook for agentic plant operations 

For teams asking where to begin, the next step is not to copy every emerging use case—it is to build the conditions that allow a few high-value use cases to succeed, earn trust, and scale. In other words, the opportunity is not to experiment more—it is to operationalize what works. In my view, that starts with six practical moves. 

1. Build the context layer. Unify operational, engineering, and maintenance knowledge so systems can answer basic but essential questions about asset state, process context, constraints, and history. 
2. Use ontologies where they add business meaning. Unified data alone is not enough. Manufacturers increasingly need semantic models that connect systems, assets, documents, and work patterns in ways AI can understand. 
3. Prioritize a high-friction use case. Start where value is clearly being lost—such as troubleshooting, shutdown planning, work preparation, reliability triage, or quality response—and pilot with the expectation that successful patterns will scale. 
4. Design the guardrails early. Be explicit about permissions, approvals, explainability, and which actions an agent can recommend versus execute. 
5. Embed intelligence in the flow of work. The goal is not to force people into another portal—it is to support productivity and decision-making where teams already operate. 
6. Scale through repeatable playbooks. Capture what the best teams do, standardize it across shifts and sites, and improve it through feedback over time. 

The north star is not how many models are deployed—it is fewer hours lost to data hunting, faster recovery after disruptions, shorter resolution times, fewer unnecessary site visits, lower variability across shifts, and a safer environment where teams can focus on judgment instead of searching for information. 

What comes next for manufacturers

Across customer conversations this season, one point has become clear: manufacturers will not realize the full value of AI through pilots alone. The real opportunity is to build an operating model where intelligence is trusted, governed, and embedded in the workflows that keep plants safe, efficient, and resilient. 

If you are exploring agentic AI for plant operations, start where value is being lost. Choose a workflow, unify the operational context around it, establish the right guardrails in place, and scale what works. That is how process manufacturers move from experimentation to execution—and from dashboards to decision-making. 

The post Agentic AI for plant operations: From dashboards to decisions appeared first on The Microsoft Cloud Blog.

1. Will AI amplify the intelligence of my organization and the attributes that make my company unique within its industry to grow my business; or will it use my intelligence for its own benefit, learning from my most important business flows and leveraging my intellectual property?
2. Can I trust that the outcomes are providing durable return on investment and that these solutions are running within the confines of my governance and security standards?
3. How do I get the visibility, control, flexibility and business model innovation needed to manage the costs associated with AI and maximize value?

I consistently advise customers that they need to build their own IQ on a platform of intelligence that is model-diverse, open and heterogeneous at every layer of the stack. Models are commoditizing. No company should be dependent upon any one model or any one model’s harness. Over the weekend, Satya warned of a world where every company across every sector is ceding value to a few models that eat everything they see. AI that is intended for growth should amplify the intelligence of an organization so that it compounds from within.

Companies also need an observability platform that provides governance, management, security and Financial Operations (FinOps) to ensure the ROI with AI. This enables AI to be trusted within the environment over which it reasons and puts the business in control of the outcomes.

Intelligence + Trust is embedded across Microsoft 365 Copilot, GitHub Copilot and Copilot Studio, where model diversity aligns cost and performance to each task. Microsoft IQ optimizes workflows, so context is routed efficiently and reduces unnecessary compute. Agent 365 is the control plane to observe, govern, manage and secure agents. We have built a system to manage AI spend as a core enterprise capability, not an afterthought. It is delivered across clouds and model providers without locking customers into a single approach.

Managing costs at scale

As agent usage scales, organizations need a clear set of levers to manage cost:

Model diversity. Any given inferencing model, model harness or agentic loop on its own does not help build out an organization’s IQ in ways that compound its intelligence. Both Microsoft 365 Copilot and GitHub Copilot are model-diverse by design without locking customers into a single provider. Different models — like GPT-5.5 or Claude Opus 4.8 — serve distinct roles with different economics. Matching the right intelligence to each task optimizes performance and cost.

Your IQ. Agents struggle with raw data. Significant compute is spent interpreting structure and context before useful work begins. The Microsoft IQ platform empowers your IQ by turning raw data into usable intelligence, continuously building a semantic understanding of how your organization operates across Microsoft 365 and line-of-business systems. It provides agents with the context they need upfront rather than requiring them to reconstruct it. The result is measurable: faster execution, higher accuracy and lower token usage. This is how intelligence compounds within your organization.

Financial operations. FinOps became critical when companies moved to the cloud and requires even greater attention as AI shifts from fixed pricing to usage-driven models. With Foundry and Agent 365, we are providing tools to help our customers optimize their AI costs today.

Frontier business models

Business models are evolving as we use AI to drive business outcomes. The User Subscription License (USL) has become the foundation, providing a package of capabilities for a predictable per-user-per-month fee. Usage-based licensing has emerged for long-running, multi-tasking agents, where cost aligns directly to the work performed.

Microsoft gives customers a unique combination of business model flexibility and integrated product experiences that is unmatched in the market. Microsoft 365 Copilot and GitHub Copilot use both models — a USL offering with not only value and capabilities, but flexible consumption on top. Today we’re announcing the general availability of Copilot Cowork worldwide, which requires the Microsoft 365 Copilot USL and is then usage-based.

Our model-diverse strategy allows customers to purchase capacity with the flexibility to use the right model for the job based on model strengths, economics and the latest innovations. Microsoft Agent Factory provides a single consumption model spanning Microsoft 365 Copilot (including Cowork), GitHub Copilot and agents built in Fabric, Foundry and Copilot Studio.

Our integrated product experiences put AI in the flow of work for both knowledge workers and software developers and manage capacity fluidly across the two. Historically these personas have been distinct, but increasingly the line between them is blurring. Coding is becoming a mainstream knowledge worker skill and chat and Cowork are becoming modalities important for software development. With Microsoft 365 and GitHub, we offer market-leading tools for both roles and make it easy to seamlessly manage capacity based on availability and need.

Agent 365: The control plane

As organizations adopt agents from Microsoft, another provider or build their own, a control plane is essential. Agent 365 gives IT and security leaders a single place to observe, govern, manage and secure agents across the organization. It builds on the Microsoft stack that enterprises trust: Entra for identity, Defender for threat protection, Purview for data governance and Intune for endpoint management. We are extending Agent 365 to include cost management, so organizations can monitor and manage agent spend alongside security and compliance. As the Frontier Firm operating model takes hold, leaders will manage human and agentic work as a single system, with visibility into both performance and cost.

The two most essential elements in any AI solution are Intelligence + Trust. At Microsoft, this conviction shapes how we design every layer of our AI platform. Microsoft IQ enables organizations to harness their own unique IQ, bringing context to data and embedding AI directly into the flow of work to deliver faster, more accurate and more trusted outcomes while safeguarding assets and protecting intellectual property. Agent 365 provides that trust layer, ensuring every agent and AI artifact is observed across the environment so organizations can move decisively from experimentation to enterprise impact with confidence. As Jay Parikh put it at Build, AI alone will not change your business. The system running it will.

We have built this system for our customers and partners, where intelligence compounds from within and every agent operates with control, visibility and trust. Together, we can scale human ambition and define how AI delivers measurable business impact across every role, organization and industry.

Judson Althoff is the chief executive officer of the commercial business at Microsoft. He is responsible for the product strategy, sales, services, support, marketing, operations and revenue growth of the company’s commercial business, which operates in more than 120 regional and national subsidiaries globally.

The post Achieving success with AI appeared first on The Microsoft Cloud Blog.