The insurance industry is confronting unprecedented challenges: Property claims volume in 2024 surged by reported 36%, largely driven by a 113% increase in catastrophe-related claims. Auto claim costs have been on the rise for the past few years, too. The trend, especially for property claims, is likely to continue as the frequency and severity of natural disasters intensifies.
The rise in claim volumes and costs necessitates that insurers become more proactive in modernizing their claims management process, including adopting generative artificial intelligence (AI), which can help companies more efficiently process vast amounts of data, reduce operational costs, and enhance customer experiences. In fact, we estimate that by enhancing generative AI capabilities it could generate savings of nearly 40%.
Claims and operations are greatly benefiting from the adoption of these solutions — including the ability to rapidly analyze vast amounts of data, reduce claims leakage or over payments, and help leaders make more well-informed financial decisions. By leveraging actuarial capabilities, large language models (LLMs) and machine learning, today insurers are reducing costs, streamlining process times, improving customer experience, and enabling their company to operate in a more integrated way.
We believe that generative AI is a tool that compliments human expertise, and that AI is not a replacement for claims adjustors and customer service representatives. Rather the technology should be used to augment processes, automate routine tasks, and free up talent to focus on more value-added activities.
Generative AI also provides insight for senior leadership to understand and act on claim severity and the root cause of loss. While working closely with clients, we have found that generative AI enables a culture of continuous improvement — ultimately leading to enhanced operational efficiency and better financial performance.
How insurers can use generative AI to significantly cut costs
Insurers that are in the early stages of adopting generative AI for claims management could see savings of 5% to 25%, according to an analysis from our property and casualty actuarial team. Those adopting more aggressive strategies could achieve cost savings between 20% to 40%.
Additionally, generative AI has the power to automate routine tasks and create new efficiencies for claims adjusters, safety professionals, and risk managers. We estimate a potential for 5% to 20% time saving benefits, depending on type of claim, line of business, and the percentage of claims documentation that can be automated.
Chief financial officers and treasurers are often frustrated with the lack of insights regarding insurance and risk-associated costs across the business. With the current market volatility and increasing catastrophic floods, fires, and hurricanes, insurers need adequate claims reserves and a good understanding of potential liabilities. Additionally, predicting large claim severity and understanding the root cause of loss with very limited data has been a significant challenge. Today, claims leaders and risk managers are finding it more difficult to gain actionable insights from disjointed claims information and exposures.
Generative AI solutions enable insurers to gain earlier claims resolutions and improve the overall customer experience. The technology has allowed them to have stronger integrated data and modeling for underwriting decisions. For organizations that want to leverage generative AI in some way, it may be overwhelming where to begin. Here, we present three ways to get started — as well as a video on generative AI and claims management, with Adam Lewis, Partner and Global Strategy Lead for our P&C Actuarial team, Adam Lewis.
Three AI-driven strategies to optimize claims management
- Leverage large language models (LLMs) bespoke to your needs. Models can be trained to augment processes, read adjuster notes, medical records and other free-form text associated with claim files, and accelerate the next steps in the claim lifecycle. Additionally, LLMs may indicate the systemic issues leading to losses. These models offer an instant feedback loop of information — that can efficiently analyze feedback from direct customer communications, surveys, and social media threads to proactively address customer concerns and improve satisfaction. Machine learning algorithms can help detect fraud, mitigate risk, maintain premiums, and improve the overall bottom line for an insurance business.
- Automate your claims assessment process to enhance customer experience. Insurers need comprehensive, unified data and modeling for managing liability claims and workers’ compensation claims. Using generative AI, organizations are gaining a deeper understanding of the root cause of loss. For example, claims adjusters can review the correlations between specific variables such as claim types, customer behavior, and external conditions, and the impact on loss ratios. This can lead to safety program initiatives and prevent those claims from happening in the future. Additionally, by streamlining workflows, adjusters can improve efficiency and focus on more value-added analysis and activities instead of tedious tasks.
Generative AI and image recognition technologies can enhance claims estimation and outcomes. For example, after an automobile accident, insurers can quickly upload images, identify damaged vehicles, assess repair needs, and estimate associated costs — including labor and parts. - Develop proactive actions for early identification of large claims. Data visualization tools enable senior leadership to quickly grasp complex data and identify trends or anomalies. Using predictive models, insurers can forecast future losses based on historical trends and current data. These models help leaders gain agility, anticipate future risks and take proactive measures to mitigate them.
The future of claims management driven by advanced analytics
By integrating predictive analytics, insurance organizations can improve decision making and better determine if a claim is likely to become severe. For example, “what if” scenarios can identify the underlying causes of the claim and understand how different factors influence losses. Data and correlations can be aggregated from multiple sources, such as claims history, economic indicators, market trends, and customer interactions. It allows leadership to recognize patterns and determine the different factors that may influence losses.
Currently, with manual processes, the data may not be efficiently captured in structured data fields and does not make it into claims management systems. At Oliver Wyman, our LenAI solution leverages the latest publicly available version of the LLM models underpinning ChatGPT. This technology enables us to help insurers extract meaningful insights from unstructured data — everything from claims adjustor notes to claims records. This information can be captured within a claims management system — allowing insurers to make future predictions and classify claims risk. With Quotient – AI by Oliver Wyman we can also help our clients harness the value of AI at every stage of their transformation journey.
Integrating generative AI is table stakes for the modern claims management landscape. As technology continues to evolve, insurers that embrace these advancements will likely enhance operational efficiency, gain a competitive advantage, and can provide their customers with stronger overall experiences.
