Insurance Claims Lose the Paper Chase as AI Gets to Work

Insurance claims processing has long been a labor-intensive and error-prone process, fraught with thousands of pages of documentation to review and potential fraudulent activities. The sheer volume of documents continues to surge. To address this challenge, artificial intelligence is being employed to augment and streamline the workflow.

This transition is underway within the industry's leading reinsurers, state insurance regulators, and banking and insurance carriers utilizing agentic AI to automate repetitive and high-risk tasks.

A Swiss Re report authored by Florian Maurer, chief Digital & Transactional officer, and Vincent Plantard, GL head of Performance, Analytics, and Delivery for Claims Corporate Solutions, outlines the development and deployment of a tool called ClaimsGenAI to automate corporate insurance claims handling. With more than 40,000 corporate insurance claims annually, Swiss Re recognized the need for a significant improvement in their claims management strategy.

The system immediately triages new documents, extracts crucial data, and organizes information to enhance the efficiency of claims handlers. Built upon two decades of unstructured claims data, the tool identifies keywords typical of corporate insurance loss scenarios that usually lead to successful recoveries. Within its first year of operation, ClaimsGenAI generated over 1,000 alerts for potential irregularities, which could result in fraud savings estimated in millions of dollars, and identified hundreds of third-party recovery opportunities that were not detected by human claims handlers.

Despite these advancements, human decision-making retains its crucial role. Swiss Re's Responsible AI strategy ensures that ultimately, the final decisions lie with the individuals supervising the AI system. Allianz and Lloyds are also adopting agentic AI for claims processing. Allianz Partners reported that an AI claims tool reduced processing time from days to mere minutes while still maintaining human oversight.

Lloyds Banking Group plans to fully implement agentic AI across its enterprise by 2026, anticipating a value of £100 million by automating fraud investigations, reserving complex human intervention for the most nuanced escalations. The shift towards agentic AI is no longer a debate about whether to implement but rather how quickly to proceed and the necessary governance framework to establish.

U.S. state insurance regulators are also adapting to the rapid adoption of AI across the industry. The NAIC's Big Data and Artificial Intelligence Working Group is developing an AI Systems Evaluation Tool to assist regulators in evaluating insurers' AI usage and governance practices. As of March, 12 states are piloting the tool, with full adoption expected at the 2026 Fall National Meeting.

A survey across various lines of business, including auto, homeowners, life, and health insurance, revealed that 88% of auto insurers, 70% of home insurers, and 92% of health insurers utilize or plan to utilize AI in their operations. In claims processing, property and casualty insurers are leveraging AI for accident image analysis, estimating claim settlement values, and detecting fraudulent activities.

Health insurers are utilizing AI for claims adjudication, prior authorizations, and risk adjustment modeling. The article reiterates that while insurers may embrace AI, they remain accountable for adhering to all relevant insurance laws, regulations, and consumer protection rules, underscoring the enduring importance of human oversight in insurance decision-making.