How Leaders in Regulated Industries Are Scaling Enterprise AI – with Leaders from Searce, Blue Cross Blue Shield, US Bank, and Citi

This interview analysis is sponsored by Searce 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.

Regulated industries, such as healthcare and finance, face significant barriers to AI adoption—compliance constraints and legacy systems hinder automation, drive up costs, and impede innovation. Despite rising investment, many enterprises struggle to scale AI in a way that delivers meaningful business impact.

A 2025 report from the US Government Accountability Office highlights that financial institutions adopting AI must contend with model risk management, compliance challenges, and evolving regulatory oversight. Research from the Financial Stability Board and academic studies on AI governance indicates that many banks face structural hurdles in integrating AI while maintaining robust risk controls.

These obstacles—legacy infrastructure, shifting compliance demands, and integration costs—are not unique to the finance sector. Gartner projects that 30% of generative AI initiatives will be abandoned by 2025, largely due to poor data quality, weak risk controls, and unclear business alignment. In regulated industries, these obstacles are magnified by legacy infrastructure, evolving compliance demands, and the high cost of integrating AI into rigid workflows.

Yet the companies that succeed in deploying AI don’t treat it as a standalone tool — they embed it within core operational processes. Boston Consulting Group finds that 62% of AI’s value emerges when applied to core business workflows rather than isolated pilot projects. Companies that extend AI to support functions also outperform peers by turning automation into a strategic advantage rather than an incremental efficiency gain.

To overcome these barriers, organizations must transition from a pilot-heavy approach to a process-first AI strategy, integrating AI into core workflows rather than treating it as an isolated experiment. 

As highlighted in Harvard Business Review, AI-driven process redesign enables firms to scale improved operations significantly, but success requires careful change management and alignment between people, data, analytics, and technology. By moving beyond traditional robotic process automation (RPA) and embracing AI for end-to-end workflow transformation, enterprises can:

• Optimize costs
• Streamline operations
• Enhance productivity
• Position AI as a fundamental driver of business success in regulated industries

In the following analysis of conversations on Emerj’s ‘AI in Business’ podcast, we take a closer look at how leaders from regulated industries approach the challenges of AI adoption amid compliance, risk, and operational complexity.

The sponsored podcast series features executives with deep expertise in healthcare, finance, and insurance, including Vrinda Khurjekar and Paul Pallath of Searce, Ylan Kazi from Blue Cross Blue Shield North Dakota, Shub Agarwal of US Bank, and Kandie Ibaka of Citi.

This article synthesizes these conversations into four critical insights for leaders in regulated industries:

• Aligning AI with core operational processes: Prioritizing process-first design ensures AI deployments address real business needs, making it easier to integrate systems into regulated workflows and achieve measurable outcomes.
• Adapting to change with autonomous agents: AI agents that can learn and respond to shifting data and policy rules help organizations stay compliant while reducing manual intervention and boosting efficiency.
• Embedding responsible AI into enterprise governance: Clear frameworks around fairness, explainability, and accountability are essential for building trust and ensuring AI systems meet regulatory and ethical standards.
• Turning compliance into an AI-enabled advantage: When integrated directly into workflows, advanced AI can streamline regulatory monitoring and risk management, transforming compliance from a bottleneck into a source of agility.

Aligning AI with Core Operational Processes

Episode 1:  A Blueprint for AI-Driven Process Optimization for Regulated Industries

Guest: Vrinda Khurjekar, Vice President of North Americas, Searce

Expertise: Enterprise AI Strategy, Process Automation, Cloud Transformation

Brief Recognition: Vrinda Khurjekar is the Senior Director for the Americas business at Searce Inc. She has been instrumental in scaling the company’s operations and leading digital transformation projects across industries. With over 14 years at Searce, she specializes in leveraging technology to drive strategic growth. 

Guest: Paul Pallath, Applied AI Lead, Searce

Expertise: AI Integration, Machine Learning Architecture, Workflow Optimization

Brief Recognition: Dr. Paul Pallath is the Vice President of Applied AI Practice at Searce Inc. He has led AI-driven strategies at global companies, including SAP, Intuit, and Vodafone, driving innovation in enterprise AI adoption.

For leaders in regulated industries, aligning AI with business processes is not just about risk mitigation — it’s a prerequisite for scalable deployment and long-term value realization. According to Vrinda and Paul, a process-first mindset is critical for ensuring AI integrates seamlessly into existing governance structures, secures stakeholder trust, and supports responsible innovation.

One of the primary challenges in highly regulated sectors is operationalizing governance in a way that supports AI experimentation while remaining compliant. As Paul explains, these organizations already have substantial governance systems in place, and the key is to integrate AI within those frameworks:

“All of these companies in the regulated sector do have their compliance framework that they have already put in place. There is significant amount of all structures in place, business processes that are put together from the context of operationalizing governance in order to ensure that whatever they are doing or serving the needs of the customer, are compliant and are always evolving with the regulatory needs.

So thinking process-first allows organizations to integrate AI into their existing governance process, ensuring that the transparency, traceability and auditability is baked into the way they think AI.”

— Paul Pallath, Applied AI Lead at Searce

Thinking process-first allows organizations to build AI systems that are transparent, traceable, and auditable by design. Instead of building models and retrofitting governance afterward, aligning AI with operational frameworks ensures that AI is not only effective but accountable.

Equally important is trust and adoption. Paul notes that many AI deployments fail because they don’t gain traction with the people closest to the work, “The final thing where 90% of the implementation does not see the light of the day is because of trust and the lack of adoption.”

He adds that thinking from a process-first perspective enables companies to “involve all the people who are going to touch AI or use AI or build AI through the spectrum of business,” which helps accelerate adoption and reduce resistance.

In practice, implementation needs to be cross-functional from the outset. As Vrinda puts it:

“When you’re trying to transform end to end business process. It’s not a siloed initiative that just touches one department. So if a hospital is trying to completely revamp its patient experience, it’s not going to be just the front office that needs to be involved. They have to manage how the front office then interacts with the insurance teams, how it interacts with the internal inventory teams, and whatnot and and all of those have to come together.”

— Vrinda Khurjekar, Vice President of North Americas at Searce

In highly regulated environments, embedding responsible AI principles early is also essential. Paul stresses the need for leadership to define ethical boundaries at the outset, “It’s fundamental to think as to what is it that we want to do with AI, and what is it that we want to leave AI to do autonomously.” Without complete clarity, he warns, AI systems risk perpetuating biases embedded in historical data or making decisions without human oversight. “Machines do not have empathy… so sometimes these data-driven decisions could lead to a lot of inhuman consequences.”

Finally, both speakers highlight the importance of building productive relationships with regulators. Rather than waiting for new rules to emerge, organizations should engage early and often. As Paul notes:

“This technology is so disruptive that even the regulators who are wanting to create regulation, they’re also thinking about not stifling innovation because of strict regulation. So it’s going to be a partnership that needs to be continuous. Most of these industries should proactively lead the conversations with regulators, to get them early in the cycle as they embark on AI interventions across business.”

— Paul Pallath, Applied AI Lead at Searce

By aligning AI with existing operational structures and involving both internal stakeholders and external regulators from the outset, organizations in regulated industries can lay the groundwork for scalable, compliant, and trusted AI systems.

Adapting to Change with Autonomous Agents

Episode 2:  AI and Agentic Systems in Healthcare Workflow Optimization 

Guest: Ylan Kazi, Chief Data & AI Officer, Blue Cross Blue Shield of North Dakota

Expertise: Healthcare AI, Data Governance, Responsible AI Implementation

Brief Recognition: Ylan Kazi is the Chief Data and AI Officer at Blue Cross Blue Shield of North Dakota. He has led data-driven transformations at major healthcare organizations, including Children’s Mercy Kansas City and UnitedHealth Group, where he built AI strategies to improve patient outcomes. 

Ylan Kazi, Chief Data and AI Officer at Blue Cross Blue Shield of North Dakota, frames AI adoption in healthcare as a multifaceted challenge. Beyond technology, cultural resistance and ethical considerations shape the pace of adoption. He emphasizes transparency and patient safety as foundational to building trust with stakeholders.

Agentic AI is already influencing high-volume operational processes, such as claims management and administrative workflows. Although fully autonomous AI remains a future milestone, the strategic application of agentic tools today is helping healthcare organizations realize efficiencies and improve accuracy.

Kazi points to the importance of realistic expectations when evaluating AI’s role. “Holding AI to an unrealistic standard of perfection limits its potential. Instead of comparing AI to an impossible benchmark, organizations must evaluate its accuracy and reliability against existing human processes.”

To help healthcare leaders navigate the complexity of AI integration, Kazi recommends reflecting on key questions that address data, culture, and process:

• What are the biggest cultural and ethical challenges hindering AI adoption?
• How can transparency and patient safety be prioritized alongside automation?
• Which operational workflows are most amenable to agentic AI enhancement?
• How do human strengths and AI capabilities complement each other in decision-making?
• What standards should organizations use to measure AI performance realistically?

Through a balanced approach, healthcare providers can better position themselves to leverage agentic AI while maintaining trust and effectiveness.

“Agentic AI is reshaping high-volume operational processes, from claims management to administrative workflows. While full autonomy is still on the horizon, strategic implementation today can drive meaningful efficiencies. Holding AI to an unrealistic standard of perfection limits its potential. Instead of comparing AI to an impossible benchmark, organizations must evaluate its accuracy and reliability against existing human processes.”

— Ylan Kazi, Chief Data and AI Officer at Blue Cross Blue Shield of North Dakota

Embedding Responsible AI into Enterprise Governance

Episode 3:  How Responsible AI is Shaping the Future of Banking and Finance

Guest: Shub Agarwal, Senior Vice President of Product Management, AI and Gen AI at US Bank; Adjunct Professor, University of Southern California

Expertise: Responsible AI, Product Innovation, Financial Services Strategy

Brief Recognition: Shub A. is the Senior Vice President of Product Management, AI, and Gen AI at US Bank. He has held leadership roles at Amazon and Silicon Valley startups, driving AI innovation and digital strategy. He is also an Associate Professor of Professional Practice at USC Annenberg, specializing in AI-driven product management.

Shub shares that financial services leaders are navigating a delicate balance between embracing AI innovation and maintaining their fiduciary duty to customers and regulators. While the sector is optimistic about the potential of generative and agentic AI, there is a strong emphasis on responsible adoption — prioritizing ethical frameworks, data governance, and compliance to ensure business value and improved customer experiences without unintended harm. 

He points out that responsible AI is often misunderstood as a buzzword but is fundamentally grounded in decades of data governance practices developed by data scientists, reinforcing that “responsible inputs” reduce risks on the output side.

As agentic AI — capable of autonomously navigating workflows and delivering highly personalized, human-like interactions — gains traction, Shub highlighted how it will redefine the user experience (UX) across industries. 

Unlike traditional interfaces based on buttons and screens, agentic AI will enable more natural conversations through voice and multi-modal communication, making interactions with software more authentic and human-centered. The shift will impact operational workflows behind the scenes as much as front-end customer engagement, fundamentally reimagining how financial services operate and serve clients.

However, Shub cautioned that agentic AI will not replace human judgment. He explains:

“Agentic AI will not replace human judgment. It is going to act as automation for tasks better suited for machines, so that humans can build relationships, communicate, and do what they enjoy most. We want to make sure we are building agent AI with the caution of making sure it is goal-oriented, optimized for the better good of the customer and the company at the same time.”

– Shub Agarwal, Senior Vice President of Product Management, AI and Gen AI at US Bank

Shub’s perspective underscores the crucial role of human oversight in AI deployment, particularly in high-stakes financial environments.

The adoption of these advanced AI systems also comes with significant challenges. Financial institutions must contend with regulatory scrutiny, risk management, and differing risk appetites across organizations, all of which complicate the selection of AI use cases that are both innovative and palatable within conservative environments. 

Shub advises financial leaders to adopt a “crawl, walk, run” methodology — starting with small, targeted workflows to demonstrate real business benefits before scaling broadly. An incremental approach helps manage risks while accelerating adoption and mitigating the effects of hype-driven missteps.

Drawing from his extensive industry experience, Shub introduced his nine-step framework for AI product creation, which integrates:

• Ethics
• Compliance
• Experimentation
• Rigorous monitoring throughout the AI lifecycle

This methodology aims to democratize AI product knowledge across organizations and bridge the gap between demos and production-ready applications — a critical step for embedding AI responsibly into enterprise governance. According to Shub, the future will see all software infused with AI capabilities, making it imperative for companies to master systematic AI product development to remain competitive and compliant.

Turning Compliance Into an AI-enabled Advantage

Episode 4: How Integrated Systems Are Reshaping Global Risk Oversight

Guest: Kandie Ibaka, Vice President and Sanctions Control Officer for International Investigations and Emerging Sanctions in Advisory Controls and Executions at Citi

Expertise: Applied Machine Learning, Risk Management, Banking Operations

Brief Recognition: Kandie Ibaka is the Vice President and Sanctions Control Officer at Citi. She has led global risk oversight initiatives, balancing modernization with regulatory compliance in the financial services sector. 

Kandie Ibaka offers a nuanced perspective on how legacy systems and fragmented technology stacks continue to challenge compliance teams within financial institutions — and how AI can be strategically leveraged to turn those challenges into advantages.

She emphasizes that many legacy systems were originally designed for fixed points in time and struggle to adapt as business goals evolve.

“I’m advocating for a continuous evaluation, as I’m constantly tasked with making sure the system still serves where the business is going. The connectedness matters along with making systems more modular, transparent, and aligned. I think the goal is just to have tools that work to build an ecosystem that can help us move with us. Tools that shift, grow, and respond to whatever is coming next.”

– Kandie Ibaka, Vice President and Sanctions Control Officer for International Investigations and Emerging Sanctions in Advisory Controls and Executions at Citi

Kandie stresses the importance of a holistic tech stack, where tools aren’t isolated but integrated. Connectedness enables more effective risk management, particularly in complex compliance environments where risks are unequal and constantly shifting. She explains, “When your tools speak to each other… we’re not just chasing after the risk, we’re categorizing it in real-time and responding proportionally.”

A key capability she highlights is tiered risk handling powered by early-generation deterministic AI systems. These technologies enable rapid categorization and segmentation of risks based on customer profiles and risk types, allowing institutions to prioritize high-risk activities for immediate escalation while ensuring all risks receive appropriate oversight and governance. Kandie points out that the approach “isn’t just about automation; it’s about clarity.”

Balancing speed and control remains a critical challenge. AI can accelerate issue detection, but compliance demands “precise, timely decision-making” with human judgment layered where necessary. Kandie warns, “Speed without oversight can be just as risky,” emphasizing the need to design systems for readiness — enabling fast action without losing sight of why specific processes require deeper investigation.

Her vision for the future is an orchestrated compliance environment where AI and human expertise work together seamlessly, “moving in the right place, at the right times, in the right context.” Such an approach helps financial institutions stay ahead of regulatory changes while maintaining credibility and operational agility.

Kandie’s insights provide a realistic, forward-looking framework for financial institutions seeking to modernize compliance by blending legacy systems with AI-enabled tools — ultimately transforming compliance from a cost center into a strategic advantage.