Pith. sign in

REVIEW 2 major objections 4 minor 2 cited by

Advanced Applications of Generative AI in Actuarial Science: Case Studies Beyond ChatGPT

T0 review · 2 major / 4 minor · reviewed 2026-08-06 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read Four implemented case studies show generative AI measurably supports actuarial work across text, reports, images, and code.

desk verdict Four worked actuarial GenAI case studies with honest limitations; the weakest link is the stochastic-output validation in the code-migration study, but the paper is worth refereeing. read the letter →

arxiv 2506.18942 v3 pith:MELTFVWT submitted 2025-06-22 cs.CY q-fin.RM

Simon Hatzesberger , Iris Nonneman This is my paper
classification cs.CYq-fin.RM
keywords generativeAIactuarialsciencelargelanguagemodelsretrieval-augmentedgenerationfine-tuningmulti-agentsystemsclaimcostpredictionlegacycodemigration
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

Generative AI can be put to work in actuarial practice today, not just speculated about: this paper reports four fully implemented case studies in which LLM-based systems improve real tasks. The paper claims that extracting structured injury features from free-text claim descriptions improves a gradient-boosting claim-cost model (18.2% lower RMSE on a held-out test set), that a retrieval-augmented pipeline reliably pulls solvency ratios, discount rates, and financial-strength ratings out of insurers' annual reports, that fine-tuning a vision-enabled LLM makes it outperform a convolutional baseline on car-damage classification, and that a five-agent system migrates R reserving code to Python with 100% pass rates against pre-written tests. What matters if true is that actuaries have a concrete, reproducible template for each of these four capabilities, plus a candid map of the governance and validation risks that come with them.

What carries the argument

Four named mechanisms carry the argument. First, LLM feature extraction: a commercial LLM is prompted with a strict schema to return injured body part, cause of injury, and injury count from claim descriptions, and a deterministic regular-expression codebook maps the free-form strings to 8 body regions and 13 cause classes so the features are stable and reusable. Second, Retrieval-Augmented Generation with Structured Outputs: annual-report text is chunked, embedded, and retrieved by cosine similarity, and the LLM is constrained by JSON/Pydantic schemas, with manually curated ground truth for exact-match evaluation. Third, fine-tuned vision-enabled LLM: a pre-trained multimodal model is fine-tuned on labelled car-damage images and then asked to output both class and optional location under a structured schema. Fourth, multi-agent migration system: five specialised agents (R analysis, translation, compilation, test runner, report) are wired in a hardcoded sequential graph with conditional retry loops, and validation rests on pre-written, R-verified test suites whose ground-truth values are hidden from the translation agent.

What would settle it

Design a deliberate sabotage test: take the two translated Python programs, introduce a subtle but material change (for example, an off-by-one origin-year shift, a wrong volume-weight denominator, or a different dispersion link in the GLM) that leaves all declared test assertions satisfied, and check whether the pipeline's tests still pass. If they do, the claim that the system validates the translation against the original code's outputs is falsified for that example.

Watch

Extended reading notes

Core claim

The paper's central claim is that each of the four GenAI application patterns works on the reported datasets: LLM-derived features from claim text carry real predictive signal beyond tabular data (RMSE reduction from 1.345 to 1.101 on the held-out test set, with the gain statistically significant under a corrected paired t-test); a shared RAG pipeline with Structured Outputs achieves near-perfect or perfect extraction of the studied financial fields across five commercial LLMs; fine-tuning raises a vision-enabled LLM's accuracy from 0.823 to 0.880 on six-way car-damage classification, slightly above a CNN baseline; and a multi-agent R-to-Python migration pipeline passes all 14/15 tests in every one of ten runs per example, with deterministic outputs matching R to floating-point precision. The authors present these as evidence that generative AI can improve predictive modelling, automate document-driven comparison tasks, perform image-based damage assessment, and automate code migration with validation, while emphasizing that production deployment still requires human oversight and governance.

Load-bearing premise

The code-migration claim rests on the assumption that the pre-written test suites (14 tests for chain-ladder, 15 for GLM) are comprehensive enough to catch any meaningful translation error; if untested code paths hide errors, a 100% pass rate would not prove the translation is correct.

Editorial extensions

If this is right

  • Actuaries can treat unstructured claim text as a feature source: LLM-derived body-part and cause-of-injury categories improve prediction beyond tabular data, with body-part features contributing the largest single gain.
  • Annual-report-style extraction can be automated with high reliability for well-defined fields, and retrieval quality and prompt specificity matter as much as the choice of LLM.
  • Fine-tuning a vision-enabled LLM is a viable alternative to training a CNN for damage classification, and it adds location and context extraction the CNN cannot provide.
  • Legacy actuarial code migration can be automated with a test-gated multi-agent workflow, achieving 100% pass rates on the two reserving examples while retries absorb occasional translation errors.
  • All four implementations are released as runnable Jupyter notebooks, so the claimed workflows are reproducible and adaptable by other practitioners.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The feature-extraction pattern in Case Study 1 should transfer to underwriting and fraud-detection pipelines wherever free-text notes sit unused beside tabular data, but the paper demonstrates it on only one synthetic workers' compensation dataset.
  • The 100% pass rates in Case Study 4 are bounded by test coverage; a natural stress test would be to inject subtle, test-invisible numerical errors and check whether the pipeline's validation catches them, since the paper itself concedes LLMs can pass tests while hiding errors in untested paths.
  • The RAG benchmark's near-perfect scores on three narrow financial fields may not generalize to open-ended or ambiguous document queries, where the failure taxonomy the paper cites suggests harder cases will still need human review.
  • Model and API churn is the quiet vulnerability: all results depend on pinned commercial model versions, so the reported accuracies should be re-measured whenever those versions are retired.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

2 major / 4 minor

Summary. The paper presents four implemented case studies demonstrating generative AI applications in actuarial science: (1) LLM-extracted features from unstructured claim descriptions improve claim cost prediction, evaluated with a held-out test, stratified 4-fold CV, and a corrected paired t-test; (2) a Retrieval-Augmented Generation pipeline extracts structured financial data from three insurers' annual reports, benchmarked across five LLMs with exact-match ground truth and repeated runs; (3) a fine-tuned vision-enabled GPT-4o classifies car damage types and provides damage localization, compared against a CNN baseline; (4) a five-agent system migrates actuarial R code to Python with pre-written test suites combining structural and numerical checks. The paper also surveys additional GenAI applications and discusses regulatory, security, dual-use, reproducibility, privacy, and governance risks. All case studies are implemented in Jupyter notebooks available on GitHub.

Significance. If the results hold, the paper provides a valuable, reproducible resource for the actuarial community. Its strengths include open-source Jupyter notebooks with pinned dependencies, pinned LLM version identifiers, cached LLM outputs in Case Study 1, a proper holdout with stratified CV and a corrected significance test in Case Study 1, and repeated-run stability analysis in Case Study 2. The paper is notably transparent about its limitations: Section 5.4 explicitly acknowledges the single-CNN-baseline and single-split evaluation in Case Study 3, and Section 6.4 explicitly acknowledges that LLM-generated code may pass tests while containing subtle errors in untested code paths. These self-acknowledged limitations are weighed in the assessment below.

major comments (2)
  1. [§6.3, Table 12; §6.4; §9] The 'automated validation' claim in the conclusion (Section 9) overstates what the test suites verify for the stochastic bootstrap outputs of the GLM-based reserving example. The paper states in Section 6.3 that only 'shape, type, and sanity tests' cover the bootstrap distributions, with numerical values 'varying naturally across runs'; no test compares the Python-generated bootstrap reserve distribution to the R-generated distribution (e.g., percentiles, CDF, or process variance). Since Section 6.4 itself acknowledges that LLMs 'may produce silently incorrect outputs that pass tests but introduce subtle errors in untested code paths' [28], the fourth component of the central claim is not fully supported. The authors should either add a distributional comparison test (e.g., comparing quantiles or a two-sample test between the Python and R bootstrap outputs) or narrow the claim to 'supported code migration with automated validation of deterministic outputs and structural checks for stochastic outputs.'
  2. [§5.3, Table 9] The comparison among the CNN, non-fine-tuned GPT-4o, and fine-tuned GPT-4o is based on a single train/validation/test split, with no confidence intervals, repeated runs, or error bars. The paper acknowledges this in Section 5.4, but the headline statement that the fine-tuned model achieves performance 'comparable to the convolutional neural network' should be softened or accompanied by uncertainty estimates, since the observed differences (0.837 vs. 0.880 accuracy) may not be stable across different random splits.
minor comments (4)
  1. [§3.3] The corrected paired t-test is computed on the 4-fold CV scores, but hyperparameters were selected using a separate 4-fold CV on the same training set; this selection can introduce optimistic bias in the CV performance estimates. The held-out test result is the more trustworthy evidence, and the paper should clarify that the reported p-values apply to the CV evaluation rather than to the held-out test.
  2. [§4.3] The ground-truth reference values in Case Study 2 were manually extracted by the authors. Although the annual reports are public and the values are listed in Tables 5-7, a second annotator or an explicit annotation protocol with inter-annotator agreement would strengthen the reliability of the reported pass rates.
  3. [§4.3] The exact-match pass criterion for discount rates requires all durations to be present and to match the ground truth exactly; this is strict and may penalize correct extractions with minor rounding differences. Consider reporting a partial-credit or tolerance-based analysis to complement the strict pass rates.
  4. [§10, References] The reference list is extensive and generally appropriate, but the paper would benefit from citing the original Kaggle competition and dataset pages in the reference list rather than only in footnotes, and from providing DOIs where available for the datasets used in Case Studies 1 and 3.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the four case studies use held-out test splits, manually extracted ground truth, and pre-written test suites with reference values isolated from model context.

full rationale

The paper's claims are empirical demonstrations rather than derivations, and each evaluation target is independently defined. In Case Study 1, the LLM-derived features are evaluated on a held-out test set with a deterministic codebook fixed in advance, and extraction quality is scored against a manually annotated 100-claim gold standard; no fitted parameter is relabeled as a prediction. Case Study 2 benchmarks five LLMs against ground-truth values manually extracted by the authors from the annual reports and uses exact-match scoring; the retrieval pipeline is shared and held constant, and there is no evidence that the stated threshold was tuned on the evaluation targets, so no circularity can be claimed. Case Study 3 fine-tunes GPT-4o on a training/validation split and evaluates on an unseen test set against CNN and off-the-shelf baselines, with manual verification only for the illustrative localization outputs. Case Study 4 is the strongest potential candidate, but the paper explicitly states that the test suites are 'pre-written, manually audited artefacts prepared in advance of each pipeline run' and that 'ground-truth values are held separately and deliberately kept outside the translation agent's context,' so passing the tests is not equivalent to fitting the validation data. The acknowledged limitation that LLMs 'may produce silently incorrect outputs that pass tests but introduce subtle errors in untested code paths' is a correctness and coverage caveat, not a circular step, because it does not make the validation target a function of the system being validated. No load-bearing self-citations or imported uniqueness theorems appear; the authors do not cite their own prior work as the basis for any case-study claim. Overall the paper is self-contained against external benchmarks and warrants a score of 0.

Assumptions & free parameters 4 free parameters · 3 assumptions · 0 invented entities

No new theoretical entities or fitted scientific constants are introduced. The free parameters are the paper's own hand-set pipeline constants, and the axioms are domain assumptions about the representativeness of datasets and the completeness of validation tests.

free parameters (4)
  • Cosine similarity threshold = 0.30
    Hand-selected in Case Study 2 Stage 2; retrieval includes chunks above this threshold; not fitted but affects which context reaches the LLM.
  • Top-k retrieved chunks = 10
    Hand-selected number of context chunks fed to the LLM in Case Study 2; changes prompt context and could affect extraction quality.
  • Chunk size and overlap = 2000 chars, 300 overlap
    Hand-selected document segmentation parameters in Case Study 2; influence retrieval precision.
  • Max retries for migration agents = 5
    Retry limit in Case Study 4; higher limits make success easier, so the 100% pass rate is partly conditioned on this choice.
assumptions (3)
  • domain assumption The 3,000 Kaggle workers' compensation claims are a realistic proxy for real claims text (Case Study 1).
    Performance gains on synthetic data may not transfer to real-world claim descriptions; the paper acknowledges the data is fully synthetic.
  • domain assumption The manually extracted ground-truth values from the 2025 annual reports are accurate (Case Study 2).
    Pass rates are scored against these author-generated labels; errors would change results. The reports are public, so the labels are verifiable.
  • domain assumption The pre-written test suites in Case Study 4 adequately capture the correctness of the R-to-Python translation.
    Tests cover specific structural and numerical properties; untested code paths could contain errors, as the paper itself notes in Section 6.4.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Advanced Applications of Generative AI in Actuarial Science: Case Studies Beyond ChatGPT." pith.science (2026). https://pith.science/paper/MELTFVWT

@misc{pith2026250618942,
  author       = {Pith},
  title        = {Pith review of: Advanced Applications of Generative AI in Actuarial Science: Case Studies Beyond ChatGPT},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/MELTFVWT}},
  note         = {Machine review of arXiv:2506.18942}
}
read the original abstract

This article explores the potential of generative AI (GenAI) to support actuarial practice through four implemented case studies. It situates these case studies within the broader evolution of artificial intelligence in actuarial science, from early neural networks and machine learning to modern transformer-based GenAI systems. The first case study illustrates how large language models (LLMs) can improve claim cost prediction by extracting informative features from unstructured text for use in the underlying supervised learning task. The second case study demonstrates the automation of market comparisons using Retrieval-Augmented Generation to identify, extract, and structure relevant information from insurers' annual reports. The third case study highlights the capabilities of fine-tuned vision-enabled LLMs in classifying car damage types and extracting contextual information from images. The fourth case study presents a multi-agent system that autonomously migrates actuarial legacy code from R to Python and validates the translation against the original code's outputs. In addition to these case studies, we outline further GenAI applications in the insurance industry. Finally, we discuss the regulatory, security, dual-use and fraud, reproducibility, privacy, governance, and organisational challenges associated with deploying GenAI in regulated insurance environments.

Review history (2 revisions) →

Discussion (0). Sign in to comment.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Runtime Uncertainty Monitoring for LLM-Based Multi-Agent Systems Using Bayesian Networks

    cs.AI 2026-07 conditional novelty 5.0 of 10

    A Bayesian-network monitor built on calibrated LLM log-probabilities gives workflow-level uncertainty scores for an actuarial multi-agent system, reproducing baseline RMSE but not clearly separating normal from pertur...

  2. Agentic AI and Retrieval-Augmented Models in Straight-Through Underwriting

    cs.AI 2026-07 conditional novelty 5.0 of 10

    On 635 synthetic BOP applications, multi-agent Agentic RAG reaches 86.5% decision accuracy versus 77.6% single-LLM and 76.9% naive RAG, with largest gains on multi-step and missing-information cases.

Reference graph

Works this paper leans on

66 extracted references · 41 canonical work pages · cited by 2 Pith papers

  1. [28]

    https://arxiv.org/abs/2304

    Yeti¸ stiren, B.,¨Ozsoy, I., Ayerdem, M., T¨ uz¨ un, E.: Evaluating the Code Quality of AI-Assisted Code Generation Tools: An Empirical Study on GitHub Copi- lot, Amazon CodeWhisperer, and ChatGPT (2023). https://arxiv.org/abs/2304. 10778

    2023

  2. [1]

    Annals of Actuarial Science15(2), 207–229 (2021)

    Richman, R.: AI in actuarial science – a review of recent advances – part 1. Annals of Actuarial Science15(2), 207–229 (2021)

    work page 2021

  3. [2]

    Annals of Actuarial Science15(2), 230–258 (2021)

    Richman, R.: AI in actuarial science – a review of recent advances – part 2. Annals of Actuarial Science15(2), 230–258 (2021)

    work page 2021

  4. [3]

    Working paper, SSRN (May 2025)

    W¨ uthrich, M.V., Richman, R., Avanzi, B., Lindholm, M., Maggi, M., Mayer, M., Schelldorfer, J., Scognamiglio, S.: AI Tools for Actuaries. Working paper, SSRN (May 2025). https://ssrn.com/abstract=5162304

    2025

  5. [4]

    Research report, Society of Actuaries Research Institute (February 2024)

    Carlin, S., Mathys, S.: A Primer on Generative AI for Actuaries. Research report, Society of Actuaries Research Institute (February 2024). https://www.soa.org/ resources/research-reports/2024/generative-ai-for-actuaries/

    2024

  6. [5]

    British Actuarial Journal29, 15 (2024) https://doi.org/10

    Balona, C.: ActuaryGPT: Applications of Large Language Models to Insurance and Actuarial Work. British Actuarial Journal29, 15 (2024) https://doi.org/10. 1017/S1357321724000102

    work page 2024

  7. [6]

    The Geneva Papers on Risk and Insurance – Issues and Practice47(2), 205–241 (2022) https://doi.org/ 10.1057/s41288-020-00201-7

    Eling, M., Nuessle, D., Staubli, J.: The Impact of Artificial Intelligence along the Insurance Value Chain and on the Insurability of Risks. The Geneva Papers on Risk and Insurance – Issues and Practice47(2), 205–241 (2022) https://doi.org/ 10.1057/s41288-020-00201-7

    work page doi:10.1057/s41288-020-00201-7 2022

  8. [7]

    Ethics and Information Technology25, 32 (2023) https://doi.org/10.1007/s10676-023-09703-z

    Koplin, J.J.: Dual-use implications of AI text generation. Ethics and Information Technology25, 32 (2023) https://doi.org/10.1007/s10676-023-09703-z

    doi:10.1007/s10676-023-09703-z 2023

Show all 66 references
  1. [8]

    In: Proceedings of the 31st International Conference on Neural Information Processing Systems

    Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A.N., Kaiser, L., Polosukhin, I.: Attention is All you Need. In: Proceedings of the 31st International Conference on Neural Information Processing Systems. NIPS’17, pp. 6000–6010. Curran Associates Inc., Re...

    2017

  2. [9]

    Nature323(6088), 533–536 (1986)

    Rumelhart, D.E., Hinton, G.E., Williams, R.J.: Learning representations by back- propagating errors. Nature323(6088), 533–536 (1986)

    1986

  3. [10]

    In: Advances in Neural Information Processing 37 Systems (NeurIPS), vol

    Krizhevsky, A., Sutskever, I., Hinton, G.E.: ImageNet classification with deep convolutional neural networks. In: Advances in Neural Information Processing 37 Systems (NeurIPS), vol. 25, pp. 1097–1105 (2012)

    2012

  4. [11]

    In: Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP), pp

    Pennington, J., Socher, R., Manning, C.D.: GloVe: Global vectors for word rep- resentation. In: Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP), pp. 1532–1543 (2014)

    2014

  5. [12]

    Journal of the Royal Statistical Society: Series A (General)135(3), 370–384 (1972)

    Nelder, J.A., Wedderburn, R.W.M.: Generalized Linear Models. Journal of the Royal Statistical Society: Series A (General)135(3), 370–384 (1972)

    1972

  6. [13]

    In: Proceedings of the Casualty Actuarial Society, vol

    Bornhuetter, R.L., Ferguson, R.E.: The Actuary and IBNR. In: Proceedings of the Casualty Actuarial Society, vol. 59, pp. 181–195 (1972). Casualty Actuarial Society

    1972

  7. [14]

    CAS Forum1, 179–213 (2003)

    Dugas, C., Bengio, Y., Chapados, N., Vincent, P., Denoncourt, G., Fournier, C.: Statistical Learning Algorithms Applied to Automobile Insurance Ratemaking. CAS Forum1, 179–213 (2003)

    2003

  8. [15]

    This paper has been integrated into SSRN Manuscript 3822407 (2019)

    W¨ uthrich, M.V.: From Generalized Linear Models to Neural Networks, and Back. This paper has been integrated into SSRN Manuscript 3822407 (2019). https: //ssrn.com/abstract=3491790

    2019

  9. [16]

    LIDAM Reprints ISBA 2020035, Universit´ e catholique de Louvain, Institute of Statistics, Biostatistics and Actuarial Sciences (ISBA) (Sep 2020)

    Denuit, M., Hainaut, D., Trufin, J.: Effective Statistical Learning Methods for Actuaries II: Tree-Based Methods and Extensions. LIDAM Reprints ISBA 2020035, Universit´ e catholique de Louvain, Institute of Statistics, Biostatistics and Actuarial Sciences (ISBA) (Sep 2020)

    2020

  10. [17]

    North American Actuarial Journal25(2), 255–285 (2021)

    Henckaerts, R., Cˆ ot´ e, M.-P., Antonio, K., Verbelen, R.: Boosting Insights in Insur- ance Tariff Plans with Tree-Based Machine Learning Methods. North American Actuarial Journal25(2), 255–285 (2021)

    2021

  11. [18]

    Research Paper Series 16–68, Swiss Finance Institute, Geneva, Switzerland (June 2023)

    W¨ uthrich, M.V., Buser, C.: Data Analytics for Non-Life Insurance Pricing. Research Paper Series 16–68, Swiss Finance Institute, Geneva, Switzerland (June 2023). https://ssrn.com/abstract=2870308

    2023

  12. [19]

    European Actuarial Journal15(2), 345–379 (2025) https://doi.org/10.1007/ s13385-025-00413-y

    Richman, R., Scognamiglio, S., W¨ uthrich, M.V.: The credibility transformer. European Actuarial Journal15(2), 345–379 (2025) https://doi.org/10.1007/ s13385-025-00413-y

    2025

  13. [20]

    Mitteilungen der Deutschen Aktuarvereinigung e.V.·Sonderausgabe zur DA V/DGVFM Jahrestagung 2024

    Cowling, C.: AI – Is there a future for actuaries? Aktuar AktuellSonderausgabe 1, 15 (2024). Mitteilungen der Deutschen Aktuarvereinigung e.V.·Sonderausgabe zur DA V/DGVFM Jahrestagung 2024

    2024

  14. [21]

    OpenAI Technical Report (2024)

    Brooks, T., Peebles, B., Holmes, C., DePue, W., Guo, Y., Jing, L., Schnurr, D., Taylor, J., Luhman, T., Luhman, E., Ng, C., Wang, R., Ramesh, A.: Video Generation Models as World Simulators. OpenAI Technical Report (2024). https: //openai.com/index/video-generation-models-as-w...

    2024

  15. [22]

    IEEE/ACM Trans

    Borsos, Z., Marinier, R., Vincent, D., Kharitonov, E., Pietquin, O., Sharifi, M., Roblek, D., Teboul, O., Grangier, D., Tagliasacchi, M., Zeghidour, N.: AudioLM: A Language Modeling Approach to Audio Generation. IEEE/ACM Trans. Audio, Speech and Lang. Proc.31, 2523–2533 (2023)

    2023

  16. [23]

    Nature Biotechnology43(2), 166–169 (2025)

    Lobentanzer, S., Feng, S., Bruderer, N., Maier, A., Consortium, T.B., Wang, C., Baumbach, J., Abreu-Vicente, J., Krehl, N., Ma, Q., Lemberger, T., Saez- Rodriguez, J.: A Platform for the Biomedical Application of Large Language Models. Nature Biotechnology43(2), 166–169 (2025)

    2025

  17. [24]

    https://arxiv.org/abs/2503.01245

    Huynh, N., Lin, B.: Large Language Models for Code Generation: A Compre- hensive Survey of Challenges, Techniques, Evaluation, and Applications (2025). https://arxiv.org/abs/2503.01245

    2025 arXiv

  18. [25]

    Jimenez, C.E., Yang, J., Wettig, A., Yao, S., Pei, K., Press, O., Narasimhan, K.: SWE-bench: Can Language Models Resolve Real-World GitHub Issues? In: Proceedings of the Twelfth International Conference on Learning Representations (ICLR 2024) (2024)

    2024

  19. [26]

    Variance16(2) (2023)

    Xu, S., Manathunga, V., Hong, D.: Framework of BERT-Based NLP Models for Frequency and Severity in Insurance Claims. Variance16(2) (2023)

    2023

  20. [27]

    https://ssrn.com/ abstract=4758296

    Richman, R.: An AI Vision for the Actuarial Profession (2024). https://ssrn.com/ abstract=4758296

    2024

  21. [29]

    https://arxiv.org/abs/2402.04373

    Mitra, A., Mohanty, S.P., Kougianos, E.: The World of Generative AI: Deepfakes and Large Language Models (2024). https://arxiv.org/abs/2402.04373

    2024 arXiv

  22. [30]

    Data & Policy4, 17 (2022) https://doi.org/10.1017/dap.2022.10

    Franceschelli, G., Musolesi, M.: Copyright in generative deep learning. Data & Policy4, 17 (2022) https://doi.org/10.1017/dap.2022.10

    2022 doi

  23. [31]

    Technical report, The Geneva Association (October 2025)

    Jia, R., Eling, M., Wang, T.: Gen AI Risks for Businesses: Exploring the Role for Insurance. Technical report, The Geneva Association (October 2025). https:// www.genevaassociation.org/sites/default/files/2025-10/gen ai report 0110.pdf

    2025

  24. [32]

    Technical report, Swiss Re, Zurich (2025)

    Swiss Re Institute: SONAR 2025: New Emerging Risk Insights – How Deep- fakes, Disinformation and AI Amplify Insurance Fraud. Technical report, Swiss Re, Zurich (2025). https://www.swissre.com/institute/research/sonar/ sonar2025/how-deepfakes-disinformation-ai-amplify-insurance...

    2025

  25. [33]

    Safety Science167, 106244 (2023) https://doi.org/10.1016/ j.ssci.2023.106244

    Oviedo-Trespalacios, O., Peden, A.E., Cole-Hunter, T., Costantini, A., Haghani, M., Rod, J.E., Kelly, S., Torkamaan, H., Tariq, A., Newton, J.D.A., Gallagher, 39 T.,et al.: The Risks of Using ChatGPT to Obtain Common Safety-Related Infor- mation and Advice. Safety Science167, ...

    2023

  26. [34]

    ASTIN Bulletin50(1), 1–24 (2020) https://doi.org/10.1017/asb.2019.28

    Lee, G.Y., Manski, S., Maiti, T.: Actuarial Applications of Word Embedding Models. ASTIN Bulletin50(1), 1–24 (2020) https://doi.org/10.1017/asb.2019.28

    2020 doi

  27. [35]

    Insurance: Mathematics and Economics107, 57–67 (2022) https://doi.org/10.1016/j.insmatheco.2022.07.013

    Xu, S., Zhang, C., Hong, D.: BERT-based NLP techniques for classification and severity modeling in basic warranty data study. Insurance: Mathematics and Economics107, 57–67 (2022) https://doi.org/10.1016/j.insmatheco.2022.07.013

    2022 doi

  28. [36]

    Machine Learning 52(3), 239–281 (2003)

    Nadeau, C., Bengio, Y.: Inference for the Generalization Error. Machine Learning 52(3), 239–281 (2003)

    2003

  29. [37]

    In: Advances in Neural Information Processing Systems 33 (NeurIPS 2020), pp

    Lewis, P., Perez, E., Piktus, A., Petroni, F., Karpukhin, V., Goyal, N., K¨ uttler, H., Lewis, M., Yih, W., Rockt¨ aschel, T., Riedel, S., Kiela, D.: Retrieval- Augmented Generation for Knowledge-Intensive NLP Tasks. In: Advances in Neural Information Processing Systems 33 (Ne...

    2020

  30. [38]

    https://arxiv.org/abs/2311.11944

    Islam, P., Kannappan, A., Kiela, D., Qian, R., Scherrer, N., Vidgen, B.: FinanceBench: A New Benchmark for Financial Question Answering (2023). https://arxiv.org/abs/2311.11944

    2023 arXiv

  31. [39]

    Packt Publishing, Birmingham, UK (2025)

    Raieli, G., Iuculano, S.: Building AI Agents with LLMs, RAG, and Knowl- edge Graphs: A Practical Guide to Autonomous and Modern AI Agents. Packt Publishing, Birmingham, UK (2025)

    2025

  32. [40]

    https://arxiv.org/abs/2404

    Edge, D., Trinh, H., Cheng, N., Bradley, J., Chao, A., Mody, A., Truitt, S., Metropolitansky, D., Ness, R.O., Larson, J.: From Local to Global: A Graph RAG Approach to Query-Focused Summarization (2025). https://arxiv.org/abs/2404. 16130

    2025

  33. [41]

    https://arxiv.org/abs/2501.09136

    Singh, A., Ehtesham, A., Kumar, S., Khoei, T.T.: Agentic Retrieval-Augmented Generation: A Survey on Agentic RAG (2025). https://arxiv.org/abs/2501.09136

    2025 arXiv

  34. [42]

    In: Proceedings of the 19th Conference of the Euro- pean Chapter of the Association for Computational Linguistics (Volume 1: Long Papers), pp

    Leung, K.K., Belbahri, M., Sui, Y., Labach, A., Zhang, X., Rose, S.A., Cresswell, J.C.: Classifying and Addressing the Diversity of Errors in Retrieval-Augmented Generation Systems. In: Proceedings of the 19th Conference of the Euro- pean Chapter of the Association for Computa...

    2026 doi

  35. [43]

    ACM Transactions on Information Systems43(2), 42 (2025) https://doi.org/10.1145/3703155 40

    Huang, L., Yu, W., Ma, W., Zhong, W., Feng, Z., Wang, H., Chen, Q., Peng, W., Feng, X., Qin, B., Liu, T.: A Survey on Hallucination in Large Language Models: Principles, Taxonomy, Challenges, and Open Questions. ACM Transactions on Information Systems43(2), 42 (2025) https://d...

    2025 doi

  36. [44]

    Machine Learning with Applications9, 100332 (2022)

    van Ruitenbeek, R.E., Bhulai, S.: Convolutional Neural Networks for vehicle damage detection. Machine Learning with Applications9, 100332 (2022)

    2022

  37. [45]

    Applied Sciences14(20), 9560 (2024)

    P´ erez-Zarate, S.A., Corzo-Garc ´ ıa, D., Pro-Mart ´ ın, J.L.,´Alvarez-Garc ´ ıa, J.A., Mart ´ ınez-del-Amor, M.A., Fern´ andez-Cabrera, D.: Automated Car Damage Assessment Using Computer Vision: Insurance Company Use Case. Applied Sciences14(20), 9560 (2024)

    2024

  38. [46]

    WIREs Data Mining and Knowledge Discovery15(2), 70027 (2025) https://doi.org/10

    Hasan, M.J., Nguyen, C.K., Boo, Y.L., Jahani, H., Ong, K.-L.: Vehicle Damage Detection Using Artificial Intelligence: A Systematic Literature Review. WIREs Data Mining and Knowledge Discovery15(2), 70027 (2025) https://doi.org/10. 1002/widm.70027

    2025

  39. [47]

    https://arxiv.org/abs/ 2406.09105

    Lin, C., Lyu, H., Xu, X., Luo, J.: INS-MMBench: A Comprehensive Benchmark for Evaluating L VLMs’ Performance in Insurance (2024). https://arxiv.org/abs/ 2406.09105

    2024 arXiv

  40. [48]

    Vicinagearth1(1), 9 (2024) https://doi.org/10.1007/s44336-024-00009-2

    Li, X., Wang, S., Zeng, S., Wu, Y., Yang, Y.: A Survey on LLM-based Multi- Agent Systems: Workflow, Infrastructure, and Challenges. Vicinagearth1(1), 9 (2024) https://doi.org/10.1007/s44336-024-00009-2

    2024 doi

  41. [49]

    ACM Trans- actions on Software Engineering and Methodology (2026) https://doi.org/10

    Liu, J., Wang, K., Chen, Y., Peng, X., Chen, Z., Zhang, L., Lou, Y.: Large Language Model-Based Agents for Software Engineering: A Survey. ACM Trans- actions on Software Engineering and Methodology (2026) https://doi.org/10. 1145/3796507

    2026

  42. [50]

    In: Proceedings of the Conference on Language Modeling (COLM 2024) (2024)

    Wu, Q., Bansal, G., Zhang, J., Wu, Y., Li, B., Zhu, E., Jiang, L., Zhang, X., Zhang, S., Liu, J., Awadallah, A.H., White, R.W., Burger, D., Wang, C.: Auto- Gen: Enabling Next-Gen LLM Applications via Multi-Agent Conversation. In: Proceedings of the Conference on Language Model...

    2024

  43. [51]

    In: Proceedings of the Twelfth International Conference on Learning Representations (ICLR 2024) (2024)

    Hong, S., Zhuge, M., Chen, J., Zheng, X., Cheng, Y., Zhang, C., Wang, J., Wang, Z., Yau, S.K.S., Lin, Z., Zhou, L., Ran, C., Xiao, L., Wu, C., Schmidhuber, J.: MetaGPT: Meta Programming for a Multi-Agent Collaborative Framework. In: Proceedings of the Twelfth International Con...

    2024

  44. [52]

    In: Proceedings of the Twelfth International Conference on Learning Representations (ICLR 2024) (2024)

    Chen, X., Lin, M., Sch¨ arli, N., Zhou, D.: Teaching Large Language Models to Self-Debug. In: Proceedings of the Twelfth International Conference on Learning Representations (ICLR 2024) (2024)

    2024

  45. [53]

    ASTIN Bulletin23(2), 213–225 (1993)

    Mack, T.: Distribution-free Calculation of the Standard Error of Chain Ladder Reserve Estimates. ASTIN Bulletin23(2), 213–225 (1993)

    1993

  46. [54]

    British Actuarial Journal8(3), 443–518 (2002) 41

    England, P.D., Verrall, R.J.: Stochastic Claims Reserving in General Insurance. British Actuarial Journal8(3), 443–518 (2002) 41

    2002

  47. [55]

    Technical report, OW ASP (2025)

    OW ASP Foundation: OW ASP Top 10 for Large Language Model Appli- cations. Technical report, OW ASP (2025). https://genai.owasp.org/resource/ owasp-top-10-for-llm-applications-2025/

    2025

  48. [56]

    Frontiers in Artificial Intelligence8, 1568266 (2025) https://doi.org/10.3389/frai.2025.1568266

    Bhattacharya, S., Castignani, G., Masello, L., Sheehan, B.: AI revolution in insur- ance: Bridging research and reality. Frontiers in Artificial Intelligence8, 1568266 (2025) https://doi.org/10.3389/frai.2025.1568266

    2025

  49. [57]

    Science381(6654), 187–192 (2023) https://doi.org/ 10.1126/science.adh2586

    Noy, S., Zhang, W.: Experimental Evidence on the Productivity Effects of Gener- ative Artificial Intelligence. Science381(6654), 187–192 (2023) https://doi.org/ 10.1126/science.adh2586

    2023 doi

  50. [58]

    Official Journal of the European Union, L series (2024)

    European Parliament and Council of the European Union: Regulation (EU) 2024/1689 of the European Parliament and of the Council Laying Down Har- monised Rules on Artificial Intelligence (Artificial Intelligence Act). Official Journal of the European Union, L series (2024). http...

    2024

  51. [59]

    Technical report, EIOPA (2021)

    European Insurance and Occupational Pensions Authority (EIOPA): Artificial Intelligence Governance Principles: Towards Ethical and Trustworthy AI in the European Insurance Sector. Technical report, EIOPA (2021). https://www.eiopa. europa.eu/system/files/2021-06/eiopa-ai-govern...

    2021

  52. [60]

    Technical Report NIST AI 100-1, U.S

    National Institute of Standards and Technology: Artificial Intelligence Risk Man- agement Framework (AI RMF 1.0). Technical Report NIST AI 100-1, U.S. Department of Commerce, Gaithersburg, MD (2023). https://doi.org/10.6028/ NIST.AI.100-1

    2023

  53. [61]

    Technical Report NIST AI 600-1, U.S

    National Institute of Standards and Technology: Artificial Intelligence Risk Man- agement Framework: Generative Artificial Intelligence Profile. Technical Report NIST AI 600-1, U.S. Department of Commerce, Gaithersburg, MD (2024). https: //doi.org/10.6028/NIST.AI.600-1

    2024 doi

  54. [62]

    Harvard Data Science Review 7(3) (2025)

    Hacker, P., Eber, M.: The Future of Credit Underwriting and Insurance Under the EU AI Act: Implications for Europe and Beyond. Harvard Data Science Review 7(3) (2025)

    2025

  55. [63]

    In: Proceedings of the 33rd USENIX Security Symposium (USENIX Security ’24), pp

    Liu, Y., Jia, Y., Geng, R., Jia, J., Gong, N.Z.: Formalizing and Benchmarking Prompt Injection Attacks and Defenses. In: Proceedings of the 33rd USENIX Security Symposium (USENIX Security ’24), pp. 1831–1847 (2024). https://www. usenix.org/conference/usenixsecurity24/presentat...

    2024

  56. [64]

    https://arxiv.org/abs/2108.07258

    Bommasani, R., Hudson, D.A., Adeli, E., Altman, R., Arora, S., Arx, S., Bern- stein, M.S., Bohg, J., Bosselut, A., Brunskill, E., et al.: On the Opportunities and Risks of Foundation Models (2022). https://arxiv.org/abs/2108.07258

    2022 arXiv

  57. [65]

    Technical report, International Actuarial Association (2025)

    International Actuarial Association: Artificial Intelligence Governance 42 Framework. Technical report, International Actuarial Association (2025). https://actuaries.org/app/uploads/2025/12/AITF Governance Framework Paper Final Approved.pdf

    2025

  58. [66]

    Technical report, American Academy of Actuaries, Washington, DC (2024)

    American Academy of Actuaries, Committee on Professional Responsibility: Actuarial Professionalism Considerations for Generative AI: A Professional- ism Discussion Paper. Technical report, American Academy of Actuaries, Washington, DC (2024). https://www.actuary.org/sites/defa...

    2024

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.