Dice Question Streamline Icon: https://streamlinehq.com

Quantify underreporting for Airbag Deployment and Suspected Serious Injury+ outcomes in human benchmarks

Determine the extent of underreporting in police-reported human crash data for Airbag Deployment and K/A (Suspected Serious Injury+) outcomes to enable accurate benchmark comparisons with automated driving system crash rates.

Information Square Streamline Icon: https://streamlinehq.com

Background

The paper applies an underreporting correction for Any-Injury-Reported benchmarks but cannot correct Airbag Deployment and Suspected Serious Injury+ due to lack of data. This limits the precision of comparisons at higher-severity outcomes where accurate human benchmarks are critical.

Establishing reliable underreporting estimates for these outcomes would improve confidence in retrospective ADS safety assessments focused on serious injuries and crash severity proxies like airbag deployments.

References

No underreporting correction was applied to the Airbag Deployment and Suspected Serious Injury+ benchmarks, as no data is available to estimate the amount of underreporting in these outcome levels. There is reason to believe that the underreporting in human crashes in these outcomes are non-zero.

— Comparison of Waymo Rider-Only Crash Rates by Crash Type to Human Benchmarks at 56.7 Million Miles (2505.01515 - Kusano et al., 2 May 2025) in Methods – Outcome Levels