Institutions today are racing to deploy AI, automate tasks and modernize systems. There's good reason. We've only scratched the surface of what these technologies can deliver. But before AI or any other advanced tool can deliver value, we need to build the cultural and operational foundations that allow innovation to take root. It's not the most exciting work, but it's the work that matters.

The stakes are real: 42% of 4,701 CEOs recently surveyed by PwC say their companies won't be viable in 10 years if they don't change course. As AI technologies mature, the focus is shifting from hype to something more honest: How do we enable these capabilities in ways that support real business needs and client outcomes?

The firms that thrive won't be the ones chasing every new tool. They'll be the ones willing to do the hard work of aligning leadership vision, talent development and organizational design. That's not because it's easy; it's because it's the only way new capabilities are adopted thoughtfully, safely and at scale. They will also be willing to slow down irreversible decisions so that reversible experimentation can move faster.

Related:Ignore the hype: Smarter tech bets at speed of change

Modernization starts with people, not tools

Change becomes durable when leaders create pathways for employees to learn, advance and take on new responsibilities, ensuring modernization strengthens rather than displaces institutional knowledge. This should include assigning motivated people to roles where new skills are required and supported, not just encouraged. 

Across industries, many firms are moving from narrow, siloed teams toward more connected, cross-functional ways of working to improve collaboration, speed and consistency.

At R&T, we've navigated this through post-merger integration, aligning teams from separate organizations toward a unified target architecture. As a technology organization, we're constantly balancing immediate business demands with longer-term strategy. And where people are involved, the shift has to be incremental. You can't reorganize everyone overnight, no matter how clear the target state looks on paper.

What's made the difference is recognizing that people move at different speeds. Some are eager to evolve, while others play a critical role in maintaining stability as transformation happens around them. That means aligning career goals and recognition to both realities -- rewarding those who lean into change and valuing those who provide continuity. 

Over time, we're blending teams from legacy silos into functionally aligned groups, bringing together experts who once worked separately to build something unified and stronger.

Related:The leadership disconnect paralyzing enterprise modernization

Make innovation repeatable

Building capabilities that stick requires more than good intentions. Our multi-year integration began with internal design. It meant defining the roles, disciplines and operating model needed to support innovation while still meeting regulatory, compliance and resiliency requirements. With 92% of technology roles expected to transform due to AI, according to a recent survey of 50 respondents by the ICT Workforce Consortium, getting this foundation right became even more urgent.

Building toward our target state architecture meant identifying opportunities for long-tenured employees to learn new skills and step into expanded roles. We complemented this with strategic hiring and third-party partnerships to accelerate execution. 

One key lesson on upskilling came from experience: Training without application doesn't stick. After an initial, broad approach fell short, we pivoted to a more individualized model, emphasizing internal mobility and hands-on learning in roles where new skills could be applied immediately.

R&T's Architecture Review Board (ARB) has been equally crucial. It evaluates new tools to ensure innovation strengthens our architecture rather than fragmenting it. Beyond governance, the ARB has evolved into an effective cross-functional learning forum bringing together leaders from technology, security, risk and product. When standards are clear, people can move faster because they trust the guardrails. 

Related:CES highlights what's new in hardware. CIOs decide what's worth upgrading

Strategy should come first, with AI as an enabler

With strong foundations in place, firms can shift toward a more confident, strategy-led posture on AI. That is grounded in clear business objectives and defined use cases, rather than experimentation for its own sake. In our firm, our AI Innovation Group serves as a central hub, working directly with business teams to surface practical, high-impact use cases that reflect real operational needs.

We're prioritizing AI where it can measurably improve efficiency and experience, particularly in highly manual internal functions and targeted workflows. But use cases must pass pragmatic tests: determining whether the underlying data exists and meets quality standards, and whether the business team has the appetite to experiment, learn, and iterate.

One example: We're working with our internal legal department to optimize day-to-day contract and documentation review. The data is available, the team is open to experimentation, and the potential efficiency gains are significant. We're applying "human-in-the-loop" principles throughout, freeing the team from repetitive tasks so they can focus on judgment and higher-value work, while preserving human control of the outcomes. 

In parallel, the legal department is upskilling its own staff through short AI courses and tutorials -- building fluency in how these tools work, identifying meaningful use cases, and strengthening its ability to advise the firm on responsible AI practices.

AI isn't a standalone story. It's the next step in a longer journey defined by thoughtful leadership, investment in people and a culture prepared to absorb whatever comes next. Not every institution can be a first mover on emerging technologies. We certainly don't claim that position. What matters more is having the discipline to adopt new tools with purpose, grounded in clear use cases, supported by governance frameworks that allow innovation to scale safely over time.

About the Author