# Credibility-Weighted Pricing of Autonomous Vehicle Liability Under Operational Design Domain Shift

arXiv: https://arxiv.org/abs/2606.17451 · HTML: https://mlsanity.com/paper/2606.17451/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 9 of its 23 references that are in the map.
- First posted on arXiv 16 Jun 2026.

## Builds on (most cited first, 9 of 9)

- [A Simple Framework for Contrastive Learning of Visual Representations](https://mlsanity.com/paper/2002.05709/) (2020)
- [Supervised Contrastive Learning](https://mlsanity.com/paper/2004.11362/) (2020)
- [Composable Effects for Flexible and Accelerated Probabilistic Programming in NumPyro](https://mlsanity.com/paper/1912.11554/) (2019)
- [Hamiltonian Monte Carlo for Hierarchical Models](https://mlsanity.com/paper/1312.0906/) (2013)
- [Comparison of Waymo Rider-Only Crash Data to Human Benchmarks at 7.1 Million Miles](https://mlsanity.com/paper/2312.12675/) (2023)
- [RAVE Checklist: Recommendations for Overcoming Challenges in Retrospective Safety Studies of Automated Driving Systems](https://mlsanity.com/paper/2408.07758/) (2024)
- [Benchmarks for Retrospective Automated Driving System Crash Rate Analysis Using Police-Reported Crash Data](https://mlsanity.com/paper/2312.13228/) (2023)
- [Initial Indications of Safety of Driverless Automated Driving Systems](https://mlsanity.com/paper/2403.14648/) (2024)
- [Dynamic Benchmarks: Spatial and Temporal Alignment for ADS Performance Evaluation](https://mlsanity.com/paper/2410.08903/) (2024)

## Abstract (arXiv)

Automated Driving System deployments create a foundational ratemaking challenge: sparse experience, shifting operational design domains, and non-stationary risk across software releases. We propose a hierarchical Bayesian credibility framework pooling across cities, software versions, and territories via a learned ODD-similarity kernel, nesting Buhlmann-Straub as a limiting case. Demonstrated on 648 verified-engaged Waymo crashes across four U.S. metros from the NHTSA Standing General Order database against 116 million matched miles, city-aggregate credibility weights are moderate (0.12-0.46), partial pooling decisively outperforms no pooling, and a power analysis shows the learned kernel's advantage becomes detectable at approximately twelve deployed cities.