Laying the Groundwork for Enterprise AI in Banking and Finance – with Leaders from EPAM and Edward Jones

This interview analysis is sponsored by EPAM and was written, edited, and published in alignment with our Emerj sponsored content guidelines. Learn more about our thought leadership and content creation services on our Emerj Media Services page.

While AI stands poised to transform even legacy financial institutions, many organizations across BFSI spaces struggle to get started in deploying intelligent systems. 

Recent research from Harvard Business Review highlights that the primary barriers to organizational AI readiness are a shortage of skilled talent and insufficient investment in workforce upskilling. For example, 52% of surveyed HR leaders identified a lack of AI expertise as the top obstacle. In comparison, only 35% felt their organizations were effective at reskilling employees to keep pace with AI demands.

Furthermore, 72% reported that AI adoption has exposed critical technical skills gaps, and just 21% said HR leadership plays an active role in shaping AI strategy, revealing a disconnect between HR’s potential and its current involvement. Addressing data literacy, adaptability, and change management is considered essential for building an AI-capable workforce.

Complementing these insights, peer-reviewed research in Nature underscores the central role of trust, transparency, and ethics in AI adoption within organizations. Lack of trust, stemming from AI’s “black box” nature, regulatory concerns, and unclear institutional responsibilities, remains a significant barrier.

Despite many financial institutions labeling themselves as “advanced” in AI maturity, a recent white paper from EPAM reveals a critical gap in workforce readiness. According to their 2024 survey of 925 financial services executives and engineers, training and enabling current employees (21.9%) and hiring new AI talent (21.7%) ranked as the top two challenges in AI adoption. 

Emerj Editorial Director, Matthew DeMello, recently hosted a special conversation with Zar Toolan, General Partner and Head of Data & AI at Edward Jones; Dana McClure, Managing Principal Consultant, Wealth Management Practice Lead at EPAM Systems; and Chris Tapley, Vice President and Head of Financial Services Consulting for North America at EPAM.

During their panel-style discussion, each executive spoke to the fundamental requirements and challenges to scale AI responsibly in the enterprise. They each cite unique scenarios around gaps in talent, AI governance, leadership understanding, and the need for long-term organizational commitment. Ultimately, all participants agree that scaling AI isn’t just a technical challenge; it requires cultural, structural, and ethical alignment.

This article examines three key insights from their conversations for leaders across financial spaces driving AI transformation at their organizations:

• Choosing partners who scale tech and empower people: Selecting partners who integrate technical expertise with effective change management ensures that AI is not only implemented but also adopted and scaled across the organization.
• Enhancing human touchpoints with AI: Prioritizing AI that improves advisor and client experiences drives stronger adoption and long-term value than focusing only on internal gains.
• Embedding AI governance to scale responsibly and compliantly: Embedding AI governance across teams and workflows to build organizational trust, ensure model transparency, and sustain long-term AI adoption and impact.

Guest: Zar Toolan, General Partner and Head of Data & AI, Edward Jones

Expertise: Finance, AI-Powered Transformation, Leadership

Brief Recognition: Zar has led large organizations through transformational change, mergers and acquisitions, and market disruptions. His experience includes work with the U.S. Navy and Wells Fargo. He holds a Bachelor’s degree in Mechanical Engineering from Columbia University and a Master’s in Finance from the Raymond A. Mason School of Business at William & Mary.

Guest: Dana McClure, Managing Principal Consultant, Wealth Management Practice Lead, EPAM Systems

Expertise: Data Analytics, Behavioral Finance Techniques, Digital Enablement

Brief Recognition: With 25 years of experience in wealth management, Dana has spent her career working at the intersection of data analytics, experience design, and behavioural finance technique. She has previously worked with LPL Financial and TIAA. She holds a Master’s degree in Business Administration from Wake Forest University. 

Guest: Chris Tapley, Vice President and Head of Financial Services Consulting for North America, EPAM

Expertise: Risk Management, Business Analytics, IT Strategy

Brief Recognition: With over two decades of experience, Chris has served as a senior banking leader for technology and operations businesses, focusing on strategy, delivery, and operations. He earned his Bachelor’s in Science from the University of Georgia. 

Choosing Partners Who Scale Tech and Empower People

Zar opens the conversation by highlighting that the foundation of any build versus buy versus lease decision lies in strong data and model governance. In regulated industries, especially, the ability to scale AI depends on overcoming data-related barriers. 

He highlights the complexity of today’s AI landscape, describing it as both multimodal and multi-model, meaning AI is delivered across various channels and involves different models tailored for specific tasks. To manage this effectively, he stresses the need for clean, governed data that can flow across systems with consistency and intent:

“When I think about the build versus buy decision, it comes down to two things: for anything that can’t drive a significant competitive advantage for your business – buy it. For anything that can add to your secret sauce? Consider building it or strategically partnering. 

And there could be bridge solutions that sit in between those two, where you might be “renting or leasing” a solution from a strategic provider to help you advance faster into specific strategic spaces that accelerate your AI transformation.”

– Zar Toolan, General Partner and Head of Data & AI at Edward Jones

He also urges companies to move away from measuring success by the number of AI use cases and instead focus on scalable patterns and platforms that deliver clear outcomes. He notes that many organizations are still counting use cases, such as 25, 100, or more, without linking them to actual results. 

Shifting toward outcome-driven thinking, he says, helps clarify whether a solution should be built or bought based on its ability to accelerate strategic goals and improve experiences for clients, users, or internal teams.

Zar says organizations are over-focused on AI tools and underinvesting in the mindsets and skill sets needed for real transformation. He argues that for AI to succeed, it must be seamless in the workflow, and users shouldn’t even realize they’re using it. The real shift begins with adopting an AI-first mindset and recognizing the massive upskilling effort required in this new era of intelligence.

He also argues that choosing the right partners means finding those who can support both the technical and the human aspects of change, helping organizations manage transformation not just technically, but also culturally and operationally.

Lastly, he emphasizes that when choosing AI and tech partners, it’s not enough to focus on those who can scale platforms and integrate data systems. The real value lies in finding partners who can also help engage the movable middle — the 70 – 80% of users who need to be brought along on the transformation journey.

He says effective partners must combine technical expertise with strong change management capabilities to ensure adoption and integration into daily workflows.

Enhancing Human Touchpoints With AI

Dana McClure then explains that many firms struggle to monetize and measure the impact of AI initiatives, especially when comparing front-office enhancements, such as meeting prep or note-taking for advisors, with back-office improvements, such as automation in software development or testing.

Dana goes on to point out that firms often prioritize internal, engineering-focused automation first, because it’s easier to quantify, while delaying AI  that directly affects advisors or clients, even though those could offer meaningful value. 

She emphasizes the importance of establishing a clear framework to assess and prioritize AI opportunities throughout the organization.

She says the fundamental AI mindset shifts occur when employees experience the benefits firsthand and feel that AI is enhancing their work. While C-suite buy-in is essential, she points out that the impact on employees is what truly drives adoption, and this aspect is often overlooked in AI transformation conversations.

Dana further argues that the focus should be on outcomes that matter to both advisors and their clients. Advisors are already using AI in their personal lives and are clear about where improvements are needed in their workflows. She emphasizes that AI initiatives should be tied to meaningful client outcomes, such as improved retirement planning or increased savings, as these directly help advisors grow their books of business.

“Clients want to work with humans. The research indicates that we want to use AI to enhance that relationship and make our advisors more relatable, contextual, and have more touchpoints.

There are still many relationships where clients only get to talk to their advisor once or twice a year. AI can vastly enhance that touch point to make personalization feel less digital and more connected, without clients even knowing that AI is being used. It can work to benefit everyone in the relationship and solidify that future advisor-client personalization.“

– Dana McClure, Managing Principal Consultant, Wealth Management Practice Lead, EPAM Systems

Embedding AI Governance to Scale Responsibly and Compliantly

Chris Tapley mentions that to unlock AI’s potential truly, companies must stop confusing usage with capability; using AI tools isn’t the same as building or managing them. He emphasizes that companies need to:

• Educate leadership on the real requirements of becoming an AI-native organization.
• Distinguish between AI users and AI builders—power users aren’t automatically AI engineers or prompt experts.
• Invest in training and tools to create AI-empowered workers, not just consumers.
• Address the hidden costs of AI, including the acquisition of new tools and upskilling of talent.
• Prioritize AI governance, which is often overlooked but critical to the responsible and secure deployment of AI.

“We have to educate – not only our people and our partners, but also our customers. If customers don’t trust AI, they won’t adopt it. 

So why are we building it? Education builds comfort, and comfort builds adoption.”

– Chris Tapley, Vice President and Head of Financial Services Consulting for North America, EPAM

Chris emphasizes that once organizations have laid the foundational work for AI, such as identifying use cases and building technical capabilities, the next critical step is governance. He highlights that governance is one of the most significant current gaps in AI adoption across organizations, comparable to the much-discussed talent gap. 

However, he clarifies that AI governance differs from traditional data governance. While data governance focuses on managing data quality, access, and compliance, AI governance must ensure that models are safe, transparent, fair, and secure.

Chris emphasizes that effective AI governance is essential to ensure data privacy, model transparency, and regulatory compliance, particularly in highly regulated industries such as finance.

But governance isn’t the final step. Chris explains that many leaders fail to recognize the commitments required after governance is in place. He suggests that leadership often overlooks the broader organizational changes necessary to sustain and scale AI successfully. 

These changes include building trust among employees, partners, and customers; aligning AI goals with business outcomes; and preparing the enterprise for cultural and process shifts. Without these long-term commitments and ongoing oversight, even well-governed AI initiatives can fail to reach their full potential.