EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Proxy-Based Evaluation and the Limits of Assurance in AI Governance

Aman Sharma

ACM AI Letters · 2026

DOI: 10.1145/3828672 ↗

Kurz gefasst

Fachartikel in der Zeitschrift ACM AI Letters, 2026. Verfasst von Aman Sharma. Die Arbeit nennt 14 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

AI governance frameworks increasingly rely on proxy-based evaluation mechanisms, including performance metrics, benchmarks, transparency documentation, audit procedures, and robustness tools, to operationalize accountability and trust. While these proxies enable scalable oversight, they can also be mistaken for evidence that the underlying governance goal has been achieved. This Policy Letter synthesizes recent literature from AI governance, auditing, robustness, and deployment studies to examine how proxy satisfaction can diverge from operational performance in deployed systems. To make this mismatch concrete, the article develops explicit proxy-goal mappings and discusses two illustrative examples: transparency documentation as a proxy for accountable understanding, and pre-deployment validation as a proxy for sustained operational performance. The analysis shows that even formally satisfied evaluation mechanisms can still leave performance degradation, hidden harms, distribution shifts, and deployment-time failure modes insufficiently examined. Rather than proposing a new metric or regulatory framework, the article offers a governance-focused interpretation of the existing literature that clarifies why assurance claims should distinguish between proxy and deployment evidence.

Zitieren

APA

Sharma, A. (2026). Proxy-Based Evaluation and the Limits of Assurance in AI Governance. ACM AI Letters. https://doi.org/10.1145/3828672

Vancouver

Sharma A. Proxy-Based Evaluation and the Limits of Assurance in AI Governance. ACM AI Letters. 2026. doi:10.1145/3828672

BibTeX
@article{Sharma2026,
  author = {Sharma, Aman},
  title = {Proxy-Based Evaluation and the Limits of Assurance in AI Governance},
  journal = {ACM AI Letters},
  year = {2026},
  doi = {10.1145/3828672}
}
Kostenlos starten →