Leveraging LLMs in Unstructured Claims Data: The CAS Issues a Research RFP for an Actuarial Solution

The Casualty Actuarial Society's (CAS) Artificial Intelligence Working Group is offering up to $40,000 for research that documents best practices for leveraging Large Language Models (LLMs) in processing unstructured data for claims analysis. Proposals are due on Wednesday, May 7, 2025.

Research Problem

Unstructured data is increasingly prevalent in actuarial work, with sources such as phone call transcripts, claim notes, images, web data (e.g., scraped data, social media posts) and scanned documents (e.g., medical records) playing a critical role in analysis. LLMs are powerful tools for processing and extracting insights from unstructured data, yet their effective implementation requires thoughtful strategies.

The CAS seeks a paper that examines actuarial use cases to present a specific LLM-based solution for converting unstructured claims data into categorical variables for reserving or ratemaking or both. The authors should provide best practices, key considerations and potential challenges in applying LLMs to unstructured data in actuarial applications, ensuring their use is both reliable and impactful.

Proposal and Work Product Requirements

Researchers will develop a paper with guidelines that address the challenges with using LLMs to process and analyse unstructured data with claims. The research should:

• Define clearly the problem being addressed (e.g., trend analysis, feature engineering).
• Provide an overview of the entire solution.
• Identify different types of unstructured data associated with claims.
• Examine key challenges of each data type.
• Discuss the tools that will be applied to the data.
• Demonstrate the formats of data the LLMs read best.
• Outline the format of data the solution produces (e.g., JSON) and how those will be fed into other actuarial workflows.

Final Deliverables

A research report documenting the classification of unstructured data, pitfalls, methodologies and best practices for using LLMs in actuarial work that also includes:

• Exploration of actuarial use cases demonstrating the effectiveness and limitations of LLM-based methods.
• A clear, step by step solution that practitioners (with assumed existing skills) to reproduce the solution.
• Codebase hosted on CAS GitHub.
• An executive summary (1-2 pages) suitable as a blog post or magazine article, highlighting the solution, best practices and provide additional helpful information understandable by a non-technical audience.

Submitting proposals

Interested researchers should submit a proposal including:

• A detailed outline of the proposed deliverable by Wednesday, May 7, 2025.
• Estimated out-of-pocket expenses (e.g., cloud storage, LLM API usage, fine-tuning models, software licenses, etc.) required to complete the work.
• Expected compensation for labor.
• Resumes of the researcher(s), indicating how their background, education and experience demonstrate their qualifications to undertake the research.

Interested researchers should submit their proposals and any questions to Annmarie Geddes Baribeau, CAS Research Manager and copy Elizabeth Smith, CAS Director of Publications and Research, by Wednesday, May 7, 2025. Please write “LLM AI Research Proposal” in the subject line.

Receipt of proposals will be acknowledged in a timely manner. Respondents who are not awarded the contract will be informed shortly thereafter.

A CAS contract will be awarded to the respondent who, in the judgment of the CAS Artificial Intelligence Working Group and entirely based on the written proposal, is best able to perform the work as specified. If the group determines that no proposal meets the requirements of the RFP, no contract will be awarded.

Compensation

Compensation will be commensurate with the time required to carry out the work.

Authors must report specific use of artificial intelligence, if any, while producing research. Authors will be required to upload their final paper electronically in the CAS’s Scholar One system.

Presentation, Ownership and Publication of Report

As a condition of selection, the CAS requires that all rights, title, and interest, including copyright and patent, in and to the report be owned by the CAS. The selected researcher(s) must sign a formal research agreement that assigns all such rights to the CAS.

In any publication of the report, the researcher(s) will receive appropriate authorship credit. The CAS may publish the report in its entirety, or any sections thereof, in any format and medium as it finds fit, including, but not limited to CAS publications, and electronic versions on its website or physical storage media. Publishing outside the CAS requires permission from the CAS, and the authors are to acknowledge previous publication.

To aid research adoption, the final work product’s code and data will also be placed in the CAS’s GitHub repository, under the MPL2.0 license.

The researcher(s) should make every effort to be available to present the report at a CAS meeting or seminar.

Timeline

ACTION DEADLINE

About the Casualty Actuarial Society (CAS)

The CAS was organized in 1914 as a professional society for the promotion of actuarial and statistical science as applied to insurance other than life insurance, such as automobile, liability other than automobile, workers compensation, fire, homeowners, commercial multiple peril, and others. Such promotion is accomplished by communication with those affected by insurance, presentation and discussion of papers, attendance at seminars and workshops, collection of a library, research, and other means. The membership of the CAS includes over 10,000 actuaries worldwide, employed by insurance companies, industry advisory organizations, national brokers, accounting firms, educational institutions, state insurance departments, the federal government and independent consultants.

About the Artificial Intelligence Working Group

The CAS Artificial Intelligence Working Group was established to fulfill the CAS mission to “advance the body of knowledge” on a technology that is transforming actuarial practice. Its objective is to encourage the exploration of Artificial Intelligence in actuarial practice through research to help educate members, build knowledge, provide practical insight and establish CAS as thought leaders.