The recent discussions at InsurTech NY highlighted a maturing perspective on artificial intelligence within the insurance sector. The prevailing sentiment among executives is less about AI as a revolutionary replacement and more about its disciplined integration as an enabling technology. It’s a shift from speculative hype to practical application, emphasizing governance, risk reduction, and strategic personnel management.
A critical point raised was that governance, often perceived as a bureaucratic impediment, is in fact a catalyst for agility. David Zweier of Ascot Group articulated this, noting that effective governance doesn't slow down AI adoption but rather enables flexibility and speed when implemented correctly. This reframes the conversation: governance isn't a barrier to innovation; it's the framework that makes innovation sustainable and impactful.
The mere presence of AI tools, it seems, changes little.
Patrick Gallic of Tokio Marine underscored that simply deploying off-the-shelf generative AI solutions like ChatGPT Enterprise or Claude Cowork will not, by itself, transform a business. While adoption might occur, genuine organizational change requires a deeper strategic alignment, focusing on tangible benefits rather than just technological presence. This implies a need for carriers to move beyond superficial implementation and truly integrate AI into core workflows with specific objectives in mind.
The practical application of AI, as outlined by Wil Lombardi of GBU Life, involves identifying high-volume, less risky data processes—such as address or beneficiary changes—where AI can significantly reduce manual effort. This allows for a targeted approach, focusing automation where it yields the most immediate and secure returns.
Ultimately, the accountability for risk decisions rests squarely with the carriers.
Adam Landau of Berkshire Hathaway Guard clarified the crucial distinction between AI’s role in processing acceptable, pre-defined risks and the necessity of human underwriters for more complex scenarios. AI can efficiently handle applications that fit established criteria, but it cannot, and should not, replace the nuanced judgment required for intricate risk assessment. This 'human-in-the-loop' model is not a concession to technological limitations but a fundamental recognition of where true value and responsibility lie in underwriting.
This delineation of AI’s function—handling the routine to free up capacity for the complex—represents a significant evolution in how the industry views its workforce. Rather than fearing job displacement, the narrative shifts towards augmentation. AI is seen as a tool to liberate skilled workers, including underwriters and claims representatives, from lower-value, repetitive tasks. This enables them to dedicate their expertise to higher-value activities: scrutinizing complex risks, engaging in deeper analysis, and addressing situations that previously might have been overlooked due to time constraints. For an underwriter, this means more time to evaluate a challenging risk that an agent might have flagged as suboptimal, transforming a potential 'no' into a carefully considered 'yes' or 'yes, with conditions.' This re-prioritization of human capital not only enhances the quality of risk assessment but also elevates the professional satisfaction of the workforce. The New York State Insurance Fund (NYSIF), for instance, is exploring AI for predictive modeling and claims processing with the explicit goal of not replacing staff, but making their jobs more rewarding. This perspective suggests a long-term dividend, fostering a more engaged and effective workforce, rather than merely chasing short-term cost reductions through automation. The implication is clear: the most successful AI implementations will be those that strategically enhance human capabilities, rather than attempting to supplant them entirely, leading to a more resilient and adaptable insurance operation.
This approach also places a renewed emphasis on the quality of human judgment. With AI handling the straightforward, the remaining tasks for human experts become inherently more challenging and require a higher degree of critical thinking and experience. It demands that underwriters and claims managers sharpen their analytical skills, focusing on the qualitative aspects of risk that algorithms cannot yet fully grasp.
It’s not about putting in a button just because it’s there.
The message from these executives is consistent: AI is a powerful tool, but its efficacy is entirely dependent on thoughtful strategy, robust governance, and a clear understanding of its appropriate boundaries. The future of AI in insurance isn't about automation at all costs; it's about intelligent augmentation, where technology and human expertise converge to create a more efficient, accurate, and ultimately, more responsible risk management ecosystem.
