AI transformation in financial services: 5 predictors for success in 2026

Financial services companies are among the most advanced adopters of AI globally, even though they operate in one of the most heavily regulated environments. The reason is simple: these firms understand the long-term transformative power of AI in a disruptive, rapidly evolving industry landscape.

Today, financial services has the highest concentration of Frontier Firms -organizations that embed AI agents across every workflow to drive speed, agility, and scalable innovation. These are companies that have demonstrated greater business impact from AI by blending human judgement with AI agents. A November 2025 IDC study commissioned by Microsoft shows that Frontier Firms report returns on their AI investments roughly three times higher than slow adopters.¹

In my previous blog, I detailed a three-phase roadmap for banks to become Frontier Firms. Now, I'd like to share essential practices in AI adoption that we believe are predictors for financial services success in 2026. These are based on hundreds of conversations I've had with customers, industry leaders, and technologists worldwide over the past three months. Organizations aiming to maximize AI's potential in 2026 and beyond should consider these practices as part of their approach to AI transformation.

The new essentials for AI success in financial services

In 2026, success won't come from experimenting with AI, it will come from re-architecting core business processes to be human-led and AI-operated. Frontier Firms are already heavily invested, with 70% of organizations across industries saying that in the next 24 months, they plan to increase their budgets for generative AI and agentic AI.¹

In financial services, there is pressure from senior executives and boards to move at greater pace and scale-to differentiate their firms by infusing AI into the fabric of their business.

In response, our customers are eager to learn how Frontier Firms do it. What are they doing to drive real impact faster than others? How do they make the right investments? What can they not afford to miss?

Regardless of when and how a firm acts, we see five essential keys to success, not all of them purely technical.

1. Value creation will drive innovation in agentic AI

As the adoption of AI evolves, Frontier Firms are re-thinking how they measure value creation. The traditional approach to crafting business cases is giving way to a new, more dynamic model as Frontier Firms are now measuring the impact of AI as use cases are deployed. A/B testing (comparing two versions of a use case to understand which delivers greater impact) has helped organizations like Investec quantify meaningful benefits, including saving bankers up to 200 hours a year with Microsoft Copilot for Sales. Focusing AI enablement on customer-facing teams is a practical way to move beyond internal productivity gains and begin influencing top-line growth.

Frontier Firms in financial services are now focusing on the measurable impact of AI-revenue growth, increased margins, and market share gain from new products, differentiated customer experiences, and empowered employee workflows. This goes well beyond use cases focused on efficiency. According to IDC, 36% of financial services firms are planning AI use cases in the next two years to boost revenue with new business models, products, or services.¹

To scale AI to more powerful, multifunction workflows, Frontier Firms are creating agentic operating models that embed AI more deeply across the business and the workforce. Working under the direction and oversight of humans, these new AI agents can reason, plan, and act across critical workflows.

Embedding agents correlates with value creation by aligning AI with core processes and key metrics. A good example is Generali France, a key player in the insurance sector in France. The organization is powering strategic use cases in customer relations and core business expertise with AI agents. In their helpdesk operations, they've developed a 24/7 voice assistant that's capable of reassuring claimants before a human steps in. Nearly 1.3 million calls-representing 30% of requests-are resolved directly by clients, with no human intervention needed.

IDC predicts such adoption of agentic AI will triple in the next two years.¹ Winning firms will anchor their innovation to business outcomes that matter in financial services (such as safer payments, faster credit decisions, and decreased fraud, to name a few), and report outcomes in quarterly scorecards so teams see purpose, not just tools.

2. Skilling and AI fluency will maximize workforce value

Even the best technology will fall short if workers aren't trained and supported to embrace it.

Successful transformation addresses the human aspect-ensuring everyone understands the benefits and feels part of the journey. Change management is both a top-down and bottom-up process. Leadership must set the vision while making sure that employees at every level are empowered to contribute.

Organizations should start with skilling, with a focus on "learning in the flow of work." Leaders can foster learning by embedding it into daily tasks so that skills stick and compound over time. They should consider building learning pathways focused on AI fluency for all employees, plus specialized tracks for specific roles, then reinforce adoption with incentives that reward employees for integrating AI into everyday workflows.

Lloyds Banking Group offers a powerful example. Departments competed for a limited number of Microsoft 365 Copilot licenses with bids based on business cases. The firm built a network of 1,000 volunteer "flight instructors" and hosted weekly "promptathons" to share best practices. The impact: over 10,000 employees trained, with 93% daily usage among 30,000 licensed users.

3. Innovation will expand across business processes

Frontier Firms are quickly moving beyond single function use cases and innovating across seven business functions on average. Focused on expanding impact across the business, AI innovation in financial service will map to key functions such as research automation in capital markets, claims in insurance, and anti-money laundering or fraud in banking. Plus, more than 70% of firms are using AI in customer service, marketing, IT, cybersecurity, and product development.¹

This broad approach is delivering better outcomes for Frontier Firms on many critical fronts: top-line growth (88%), brand differentiation (87%), cost efficiency (86%), and customer experience (85%).¹ Interestingly, it also opens innovation to drive new opportunities, such as transforming support functions into revenue generators through new customer experiences.

The impact of advanced AI spans the financial services industry. In capital markets, it improves market research and analytics, personalizes the client experience across digital channels, and helps tailor services to individual needs. BlackRock, for instance, is transforming its investment lifecycle by embedding AI into its Aladdin platform, used by tens of thousands of users. Likewise, LSEG and Microsoft have built tools that let financial professionals quickly build custom agents leveraging 33-plus petabytes of trusted market data.

In banking, AI is equipping financial institutions with new tools for personalized service and stronger client relationships. AI will help to improve the effectiveness of targeted marketing campaigns, streamline lending and mortgage processes, and safeguard assets with advanced fraud analysis. AI agents will continue to proliferate, thanks to efforts such as Argentina's Banco Ciudad, which launched a new AI Center of Excellence that delivered 10 agents in six months to improve customer service, workflow automation, and cross-team integration.

In insurance, AI will accelerate value by automating complex, high-value processes such as underwriting, claims management, and policy administration while improving risk modeling and compliance. It will become more effective in helping insurance agents better serve clients, and in helping customers make the best policy choices. Fraud detection and decision making will advance, thanks to innovations such as a new service from Shift Technology designed to help insurance companies by automating classification and extraction of unstructured data.

4. Responsible AI and regulatory readiness will be competitive advantages

The firms that lead in AI will also lead in governance. IDC predicts 1.3 billion AI agents will be in business workflows by 2028.² As they become part of the organization, business leaders need to think of them as employees in many ways. They will require identities, permissions, and oversight. They'll need to be trained, monitored, and auditable.

Proactive compliance is now an imperative, if not a competitive advantage. In 2026, regulatory complexity will only intensify, meaning that trust must be the foundation for scale and innovation. Frontier Firms embed responsible AI frameworks into every stage of the lifecycle, from design to deployment and monitoring. Leaders will integrate data privacy, encryption, and access controls across AI innovation from day one.

Bradesco's Bridge is an example of how a bank can responsibly operationalize agentic workflows. Bridge uses Microsoft Azure AI to provide a governed API layer to enforce consistent policies and secure data access. The result is 83% resolution rates for digital service and a 30% reduction in tech costs.

Complementing this, Microsoft's new Agent 365, announced at Microsoft Ignite 2025, addresses the critical need for control at scale. Agent 365 is a unified control plane that extends enterprise management and security to partner-developed agents and even those running outside the Microsoft Cloud. Integrated with Microsoft Entra, Microsoft Purview, and Microsoft Defender, it enforces identity, permissions, and data protection while surfacing telemetry for ROI and compliance. All agent activities are logged into Microsoft Sentinel and Purview audit logs, giving security operations teams a full audit trail to investigate incidents. It also minimizes the risk of "shadow AI" and helps ensure that agent deployment is compliant.

5. Data strategy will unlock AI at scale

Perhaps the single most important requirement for success with agentic AI is data readiness. Without the right strategy to ensure data interoperability and real-time intelligence, an organization's most important initiatives are destined to fall short.

To derive the most value from AI, the first step is to unify all of the organization's data. Fragmented systems-core banking, risk models, compliance archives, customer relationship management-create blind spots. The traditional approach to doing this-moving all of an organization's data to a single location-has often proven costly and resource intensive. The new approach is to use a unified data platform, such as Microsoft Fabric, which connects data wherever it sits, giving organizations a single source of truth, even when the data resides on other platforms or systems. This approach empowers organizations to deliver faster insights, lower costs, and unify governance, accelerating their AI deployment.

LSEG leveraged Microsoft Fabric to modernize their data infrastructure and accelerate time to market. LSEG uses Apache Spark on Fabric to process around 80,000 files daily, consuming approximately 280,000 capacity units per day. Usage is growing rapidly, with month-on-month consumption increasing by more than 50%, signaling the start of a transformative journey.

Finally, organizations must embed governance and security. Identity-based access, audit trails, and adaptive risk controls are non-negotiable.

AI at scale is not just about models-it's about the foundation: data, cloud, and governance. Microsoft delivers on all these imperatives: Microsoft Fabric IQ for unified semantics, Foundry IQ for contextual knowledge, Azure for scalable performance, and Microsoft Agent 365 for governance. The mandate is clear-modernize your data foundation today to avoid challenging consequences later.

Now is the time for agentic AI in financial services

Financial institutions have long pursued productivity gains to reinvest in growth, and that imperative remains. But AI is no longer just about cost savings-it's about reinventing how organizations engage customers, redefine services, bend the innovation curve, and create competitive differentiation.

To empower our customers to accelerate to scale, Microsoft has built a full stack enterprise AI platform that is fully integrated. Our approach is architected around 3 objectives:

• AI in the flow of human ambition
  Copilot can be used by end users to develop personal agents to execute tasks on their behalf. For low-code development, customers can use Microsoft Copilot Studio to build agents and assistants that can answer questions and execute workflows that integrate with their data and line of business systems. Both are powered by Work IQ-an intelligence layer that customers can use to give real-time insights into how teams are working across applications and processes and improve operational performance.
• Ubiquitous innovation
  To enable ubiquitous innovation, Fabric is an AI-powered, end-to-end data and analytics platform that breaks down the data siloes and unifies data that resides in different places. It is powered by Fabric IQ, a semantic intelligence layer that enables customers to organize data around the language and meaning of their business (such as relationships, entities, and logic) so teams can build agents that are grounded in the same semantic understanding of the business.

  Ubiquitous innovation is also enabled by Microsoft Foundry, Microsoft's unified Azure platform for building, deploying, and optimizing pro-code AI apps and agents-bringing models and tools together. Foundry IQ makes it easier for teams to build reliable agents and apps that use trusted governed knowledge. Foundry IQ is a knowledge grounding layer for agents: it connects AI apps and agents to content across many sources (like documents, policies, Microsoft SharePoint, OneLake, and external stores) using an Azure AI Search-powered knowledge base and a single grounding API, with permission-aware access.

• Observability at every layer
  As organizations navigate the agentic era in financial services, the question isn't just how to deploy AI agents-it's how to do it securely, at scale, and in a way that positions the organization as a leader. This requires a connected, trusted, and scalable platform to orchestrate AI across the business, which Microsoft offers. With Agent 365 as a unified control plane, organizations gain a single governance layer that extends enterprise-grade identity, security, and compliance to every agent-whether it's built in-house or running on external clouds. This means organizations can confidently embrace AI innovation without sacrificing oversight or regulatory alignment. Microsoft's approach means organizations aren't just adopting tools; they're anchoring their business to the platform for the open agentic web-a trusted, interoperable ecosystem where agents and humans collaborate seamlessly.

The firms that embrace this shift-modernizing data, embedding governance, and preparing their workforce-won't just adapt; they'll lead the next era of financial innovation.

Learn more

¹ IDC, What every company can learn from Frontier firms leading the AI revolution, sponsored by Microsoft, November 2025.

² IDC Info Snapshot, sponsored by Microsoft, 1.3 Billion AI Agents by 2028, May 2025 #US53361825.