The global (re)insurance industry is entering a new era of risk analysis that is changing how leaders approach underwriting, portfolio management and capital allocation decisions, says Marsh Re’s Dan Becker
Steady advances in technology have expanded access to information about the world around us while creating a rich ecosystem of specialist providers for models and data. AI is now significantly lowering the practical barriers to using vast amounts of information to surface risk insights across enterprises.
Discussion of AI’s impact often focuses on automation initiatives and improving cost efficiency. While both opportunities matter, AI’s deeper impact is its ability to create a richer situational context for decision-makers. The context available to support decisions is now far less constrained by a firm’s proprietary data, the robustness of its process or human capacity for synthesis. Leaders now have an unprecedented opportunity to fundamentally expand the depth, breadth and speed of their risk analysis frameworks.
The technical foundation for capturing this opportunity is similar to existing best practices in data strategy, but the value of solving it has increased dramatically. Companies need to move beyond fragmented silos and connect and govern the intelligence they consume, purchase, generate and distribute across the enterprise. When data on the underlying market, the competitive environment, risk models and internal business metrics are brought together into a connected intelligence framework, leaders can assess underwriting, portfolio and capital allocation alternatives with significantly broader context and speed.
The (re)insurance industry has historically filled information gaps through shared foundational views of risk, whether through industry groups, regulatory data submissions or broadly adopted third-party vendors. For most companies, building an independent view of risk was expensive and expert-intensive. Even when firms sought to deviate from those shared views, differentiation depended on dedicated experts capable of translating product strategy, portfolio experience and institutional knowledge into a distinct perspective on risk. With a stronger data foundation and connected intelligence, those experts can now interrogate institutional intelligence at scale. As a result, a more proprietary and strategy-specific view of risk is now within reach for far more firms.
It is tempting to project these trends into a future of automated decision-making and capital allocation. But the fundamental nature of risk and uncertainty limits mechanised decision-making at the portfolio level. When probabilities of “expected” outcomes can be readily produced, uncertainty itself can become underappreciated. Yet uncertainty is precisely why strategic decisions remain grounded not only in quantitative frameworks, but also in human judgment, organisational psychology and leadership conviction. The enduring value of human expertise lies in helping leaders navigate the trade-offs that uncertain outcomes create.
The value of these experts with deep practical experience is being significantly amplified by technology. By automating the extraction of insights and accelerating high-dimensional analysis, AI allows companies to spend less time producing analysis and more time applying judgment. The effect becomes even more powerful when firms make expert methods explicit and reusable: how questions are framed, evidence weighed, assumptions tested and tolerances applied. The firms that outperform over the next five years will not be those that automate analysis most aggressively, but those that best combine modern analytical platforms with the irreplaceable nuance of human judgment.