RBC Insurance’s AI Playbook: How governance, agility and client focus are driving measurable results

Among the use cases for AI within the wealth management and wider financial services sector is its ability to detect fraud faster and more accurately than legacy systems.

When RBC Insurance set out to embed generative AI into its operations, the goal was not experimentation for its own sake, but a focus on delivering tangible business value while maintaining strict risk and privacy controls.

According to Vik Luthra, chief digital, data & strategy officer, the rapid success of its AI-powered claims assistant came from a deliberate strategy of incremental, well-governed innovation.

Speaking with InvestmentNews during our monthly spotlight on AI, Luthra said that what allowed RBC’s genAI claims assistant to move from concept to measurable fraud savings so quickly was first down to mindset.

“With an agile mindset, RBC Insurance is building safe, secure and value-add AI incrementally, ensuring that value can be delivered early in the roll-out process,” he said. He points to CLARA — the Claims Lifecycle Automated Recommendation Assistant — as a proof point. “CLARA (our genAI-enabled Claims Lifecycle Automated Recommendation Assistant) captured over $2 million in savings related to fraudulent claims in its first year, while still in pilot phase.”

Crucially, the technology is designed to augment, not replace, expert judgment. “By using AI tools alongside our exceptional claims specialists, we’re able to accelerate the processing of claims without sacrificing data privacy or risk management,” Luthra said. “Our Insurance claims team is made up of highly trained and specialized individuals, often coming from a healthcare background. We rely on their professionalism to assist our clients when they’re experiencing a challenging time in their lives, with help from indicators from AI and other tools.”

He adds that foundational investments in data and governance enabled RBC Insurance to move quickly without compromising stability. “RBC has built out a strong foundation of risk governance and centralized data that allows us to move faster and more confidently, while achieving long-term stability.”

Startup mindset

The combination of speed and control reflects Luthra’s broader philosophy of bringing a startup mindset into a large financial institution. When asked how that experience has shaped RBC Insurance’s approach to AI innovation, he emphasized disciplined experimentation.

“When exploring use cases for emerging technology, we are always thinking about guardrails to mitigate risks inherent with accelerating innovation,” he said. “I’ve found that the best entrepreneurs are focused on limiting their risk, keeping a laser focus on their clients and solving real problems, at scale, as fast as possible.”

Luthra said that it’s about working iteratively, testing extensively at each stage and building for rapid growth. But ambition must be balanced with responsibility.

“This balance of leaning into opportunities while developing a strong risk framework is critical for long-term success,” he said. “We’re not afraid to try new things and to have ambitious goals – but we’re going to take a responsible approach to keep employee and client safety and security at the forefront.”

As AI becomes more embedded across financial services, Luthra sees it reshaping how institutions detect fraud and protect customer assets. At RBC, he describes AI as transformative at the enterprise level.

“At RBC, we see AI as a generational technology that empowers us to reimagine what a bank can do, and we’re increasingly focused on leveraging AI across our businesses to enhance how our teams support our clients,” he said. “Since launching RBC Borealis in 2016, we have built an AI foundation underpinned by data scale, exceptional talent, a culture of innovation and world-class security.”

Fraud trends

Fraud trends make this capability increasingly important. “Instances of fraud are increasing globally, including in the insurance industry, which requires higher capital reserves to protect our clients’ insurance policies,” Luthra said. Removing fraudulent claims has downstream benefits. “Identifying and removing fraudulent claims will positively impact pricing models and our ability to provide clients with better policies and better protection.”

Governance and privacy remain non-negotiable foundations for RBC’s AI strategy. Safeguards to ensure customer data privacy and responsible AI use, require formalized frameworks.

“RBC’s robust risk governance framework helps ensure that we consider, explore and build genAI tools safely,” he said. “We have also formalized a set of Responsible AI Principles that help to guide our approach to developing and deploying AI-enabled solutions. Privacy and security, accountability, fairness and transparency, and responsible disclosure guide our approach to developing and deploying AI-powered solutions.”

Luthra said that compliance is built into the lifecycle and follows protocols to ensure that AI systems are compliant with industry standards and regulatory guidelines. All AI systems must meet requirements throughout the development lifecycle, including in testing, validation and monitoring in order to continuously improve models.

“Within these responsible AI pillars, we are testing early and often, and building for scale,” he said. “This means starting with a solid foundation of centralized data and developing AI models and tools that can be reused for multiple products and initiatives.”

Looking ahead, Luthra is clear on what will separate leaders from laggards in financial services’ AI transformation: “Strong governance and risk management. An AI-ready workforce. And clear priorities focused on meaningful value for our clients, the community, our employees and our business.”

This article is part of our Monthly Spotlight series, which in January focuses on AI in Wealth. Full coverage can be found here.