First Hawaiian Bank (NASDAQ: FHB)

Rosemary P

Trusted BI Authority

Rosemary P is a business technology executive with over 25 years of experience leading high-performance teams and enabling business growth through data and technology-driven strategies. At First Hawaiian Bank (NASDAQ: FHB), she focuses on building process-driven data foundations that support AI adoption and decision-making.

In an interview with CIOReview, she shared insights on data quality, governance frameworks and building operational and cultural foundations for responsible AI adoption.

Strengthening Data Foundations for AI Practices 

Many organizations are rushing to adopt AI without building the strong data foundation needed to generate meaningful value. AI adoption is treated as a competitive race, where organizations move faster without understanding whether their data is ready to support it.

Business intelligence and reporting environments often focus on endpoint reports rather than understanding how data connects across the enterprise. They lack the semantic context needed for AI to interpret information effectively across business functions. Without strong semantic context across enterprise data, AI can generate inaccurate insights and unreliable decisions as it processes information at a speed far beyond human capability.

Good data management starts with building organizations around end-to-end business processes rather than isolated functions. Many companies focus only on improving individual functions, but strong data management requires understanding how processes connect across the business.  

In process-based organizations, a stronger data foundation is created through connected and well-structured processes. If processes are fragmented, the data and automation built on top of them will reflect those same weaknesses. Strong process alignment also supports better technology adoption.  

Building Data Quality into Processes 

One of the most common gaps in how enterprises approach data quality is treating it as an after-the-fact exercise, instead of building it directly into business processes. Many organizations have data governance frameworks and standards in place, but data quality is often checked only through compliance or risk-driven reviews.

Maintaining strong data quality requires organizations to embed validation directly into operational processes while reinforcing the understanding that data is a critical enterprise asset. This leads to stronger analytics, reliable decision-making and better outcomes for AI and automation initiatives. Employees generating, collecting and processing data also play a critical role in maintaining data quality.

A stronger approach begins with rethinking where data teams sit within the organization and how they engage with the business. Traditionally, data teams sit within technology departments, creating a relationship where the business requests solutions and technology simply delivers them. Recognizing data experts outside of technology and embedding business intelligence within the business creates a more strategic partnership, where data teams help interpret what the data is saying and uncover opportunities that support business objectives.

From Assistive AI to Agentic Systems

AI is often viewed as a single concept, even though different forms of AI carry varying levels of risk, autonomy and business impact. Some AI tools are designed to assist humans through summarization or workflow support. These systems can be adopted quickly as access to information is controlled.  

In contrast, agentic AI systems are capable of acting more autonomously and interpreting context beyond predefined tasks. These systems require stronger governance as they can make decisions with less human intervention.  

Leaders need frameworks that tier AI capabilities based on risk, functionality and human involvement. Before scaling to advanced AI systems, organizations must understand the quality, context and controls surrounding their data. 

Laying the Foundation for AI Adoption

A future-ready data foundation starts with going slower first in order to move faster later. Moving too quickly without strong frameworks and data context often creates operational and governance challenges later.

Organizations need to take an assessment of their data and technology foundations, people, culture and overall readiness for AI adoption. Building AI readiness requires stronger data literacy, critical thinking and a culture that helps employees work responsibly with AI.