What's Happening?
A recent survey by Moody's Ratings indicates that while artificial intelligence (AI) is widely adopted across core operations by most North American insurers and brokers, many are unable to demonstrate its financial return on investment. The survey, which
included 58 companies across various insurance sectors, found that approximately two-thirds of respondents have multiple AI applications in production, with 34% deploying AI broadly throughout their businesses. However, only 41% of these companies quantify and track the benefits for individual AI use cases, while another 34% quantify benefits without consistent tracking. A significant 24% reported that the benefits were primarily qualitative. Moody's suggests that productivity gains from AI may not immediately translate into financial results due to the need for workflow redesigns and the simultaneous operation of old and new processes. Larger insurers are better positioned to absorb AI investments due to their larger technology budgets and access to more data, potentially leaving smaller companies at a disadvantage.
Why It's Important?
The findings from Moody's highlight a critical challenge within the U.S. insurance industry: the gap between AI adoption and measurable financial impact. This situation has significant implications for business strategy, investment decisions, and competitive landscapes. Companies that cannot effectively track AI's payoff risk misallocating resources and failing to optimize their technological advancements. The disparity between larger and smaller insurers in their capacity to invest in and leverage AI could lead to a widening competitive gap, potentially consolidating market power among larger entities. Furthermore, the identified obstacles, such as skills shortages, data quality issues, and integration with legacy systems, point to systemic challenges that could impede broader technological progress and efficiency gains across the sector. The inability to prove ROI could also deter future AI investments, slowing innovation in an industry that could greatly benefit from enhanced automation and data analysis.
What's Next?
Insurers will likely need to prioritize developing more robust methodologies for quantifying and tracking the financial benefits of their AI implementations. This could involve investing in advanced analytics tools, refining key performance indicators (KPIs) specifically tailored to AI-driven outcomes, and establishing clearer timelines for expected returns. Addressing the identified obstacles will also be crucial. Companies may need to focus on upskilling their workforce to mitigate talent shortages, invest in data governance and quality improvement initiatives, and strategically plan for the integration of AI with existing legacy systems. Regulatory bodies may also increase their focus on AI governance and ethical considerations, potentially leading to new guidelines that insurers must adhere to. The ongoing evolution of AI technology, particularly in areas like generative AI, will continue to present both opportunities and challenges, requiring insurers to adapt their strategies to remain competitive and demonstrate tangible value.
Beyond the Headlines
Beyond the immediate financial implications, the struggle to quantify AI's payoff in the insurance sector touches upon broader themes of technological disruption and organizational change. The qualitative benefits cited by a quarter of respondents suggest that AI is delivering value in less tangible ways, such as improved customer experience or enhanced decision-making, which are harder to measure but nonetheless crucial for long-term success. This raises questions about how businesses define and measure 'value' in an era of rapid technological advancement. The reliance on third-party models and cloud providers also introduces new dimensions of operational and cyber risk, underscoring the need for comprehensive risk management strategies. The potential for AI to exacerbate cyber risks, as highlighted by the example of Anthropic’s Mythos model identifying vulnerabilities, suggests that while AI offers efficiency, it also demands heightened vigilance regarding security and vendor dependency. This evolving risk profile will necessitate a re-evaluation of traditional insurance models and potentially lead to new types of insurance products designed to cover AI-specific risks.
