When Compliance Data Masquerades as Evaluation: Measurement Validity for Deployed AI Systems

2026-09-11

AI Summary

论文讨论了使用合规数据进行AI系统评估的问题，提出需明确能力、结果、暴露机会等要素的一致性以确保评估有效性。

Abstract

We argue that a recurring failure in the evaluation of deployed AI systems occurs when data collected for operational monitoring or regulatory compliance are interpreted as if they were designed for comparative evaluation. Automated driving provides a concrete example of this problem. U.S. disengagement and crash-reporting regimes produce valuable operational evidence, but differences in reporting scope, exposure, deployment domain, event capture, and comparator construction limit the safety claims that can be supported from these measurements alone. We frame this issue as a measurement-validity problem in AI evaluation rather than as a transportation-specific data limitation. We argue that comparative claims about deployed AI systems require alignment between the intended capability, measured outcome, exposure opportunity, deployment domain, data-generation process, and evaluation comparator. Using automated-driving safety evaluation as a case study, we propose an evaluation contract that makes these assumptions explicit before operational data are interpreted as evidence of comparative performance. The broader implication is that data useful for monitoring deployed AI systems are not automatically valid benchmarks for evaluating them.

Problem

• AI evaluation
• measurement validity
• operational monitoring
• comparative performance
• data interpretation

Innovation

• Evaluation Contract
• Measurement Validity
• Deployed AI Systems