MARIA: A Framework for Marginal Risk Assessment without Ground Truth in AI Systems

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Chen, Jieshan, Ma, Suyu, Lu, Qinghua, Lee, Sung Une, Zhu, Liming
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866912679224934400
author Chen, Jieshan
Ma, Suyu
Lu, Qinghua
Lee, Sung Une
Zhu, Liming
author_facet Chen, Jieshan
Ma, Suyu
Lu, Qinghua
Lee, Sung Une
Zhu, Liming
contents Before deploying an AI system to replace an existing process, it must be compared with the incumbent to ensure improvement without added risk. Traditional evaluation relies on ground truth for both systems, but this is often unavailable due to delayed or unknowable outcomes, high costs, or incomplete data, especially for long-standing systems deemed safe by convention. The more practical solution is not to compute absolute risk but the difference between systems. We therefore propose a marginal risk assessment framework, that avoids dependence on ground truth or absolute risk. It emphasizes three kinds of relative evaluation methodology, including predictability, capability and interaction dominance. By shifting focus from absolute to relative evaluation, our approach equips software teams with actionable guidance: identifying where AI enhances outcomes, where it introduces new risks, and how to adopt such systems responsibly.
format Preprint
id arxiv_https___arxiv_org_abs_2510_27163
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle MARIA: A Framework for Marginal Risk Assessment without Ground Truth in AI Systems
Chen, Jieshan
Ma, Suyu
Lu, Qinghua
Lee, Sung Une
Zhu, Liming
Software Engineering
Artificial Intelligence
Human-Computer Interaction
D.2.8; D.2.9.m; I.2
Before deploying an AI system to replace an existing process, it must be compared with the incumbent to ensure improvement without added risk. Traditional evaluation relies on ground truth for both systems, but this is often unavailable due to delayed or unknowable outcomes, high costs, or incomplete data, especially for long-standing systems deemed safe by convention. The more practical solution is not to compute absolute risk but the difference between systems. We therefore propose a marginal risk assessment framework, that avoids dependence on ground truth or absolute risk. It emphasizes three kinds of relative evaluation methodology, including predictability, capability and interaction dominance. By shifting focus from absolute to relative evaluation, our approach equips software teams with actionable guidance: identifying where AI enhances outcomes, where it introduces new risks, and how to adopt such systems responsibly.
title MARIA: A Framework for Marginal Risk Assessment without Ground Truth in AI Systems
topic Software Engineering
Artificial Intelligence
Human-Computer Interaction
D.2.8; D.2.9.m; I.2
url https://arxiv.org/abs/2510.27163