Is TabPFN the Silver Bullet for Insurance Pricing?

2026-05-21

AI Summary

This study presents the first application of the tabular foundation model TabPFN to motor third-party liability (MTPL) insurance pricing, systematically evaluating its potential to replace traditional actuarial models in modeling claim frequency and severity. Using two publicly available MTPL datasets, the authors conduct an end-to-end comparison of TabPFN against generalized linear models (GLMs) and XGBoost under a context learning paradigm, assessing generalization performance, computational efficiency, and sensitivity to context size. Results indicate that although TabPFN enables zero-shot inference without fine-tuning, it does not consistently outperform established baselines, exhibits substantially longer inference times, and is highly sensitive to the number of context examples. Consequently, TabPFN currently remains insufficient as a substitute for conventional actuarial approaches. This work establishes the first empirical benchmark for tabular foundation models in non-life actuarial applications.

Abstract

Modelling claim frequency and severity for non-life insurance pricing predominantly relies on generalised linear models, with gradient-boosted machines as the leading machine learning alternative. Tabular foundation models (TFMs) offer a fundamentally different paradigm. By pre-training on large collections of synthetic datasets, TFMs enable inference on new data through in-context learning, without any dataset-specific fitting or hyperparameter tuning. This paper presents a first empirical evaluation of TabPFN for motor insurance pricing, benchmarking it against GLM and XGBoost on two publicly available MTPL datasets. Our results show that TabPFN does not consistently outperform established baselines, exhibits substantially longer inference times, and is sensitive to the size of the in-context training set. While tabular foundation models represent a promising direction, particularly in data-scarce settings, their current formulation does not offer a viable replacement for established actuarial methods.

Problem

• Tabular Foundation Models
• Insurance Pricing
• Claim Frequency
• Claim Severity
• Non-life Insurance

Innovation

• Tabular Foundation Models
• In-Context Learning
• Insurance Pricing
• TabPFN
• Non-Life Insurance