AI Mismatches: Identifying Potential Algorithmic Harms Before AI Development

TL;DR

This paper introduces an early-stage approach to identify and mitigate AI mismatches that can cause harm, using analysis of 774 cases and risk mapping matrices to improve AI safety and performance.

Contribution

It presents a novel AI Mismatch framework with matrices to predict and address potential harms before development progresses.

Findings

Analyzed 774 AI cases to identify critical risk factors.

Developed seven matrices mapping risk factors and high-risk areas.

Case studies show the approach reduces AI development risks.

Abstract

AI systems are often introduced with high expectations, yet many fail to deliver, resulting in unintended harm and missed opportunities for benefit. We frequently observe significant "AI Mismatches", where the system's actual performance falls short of what is needed to ensure safety and co-create value. These mismatches are particularly difficult to address once development is underway, highlighting the need for early-stage intervention. Navigating complex, multi-dimensional risk factors that contribute to AI Mismatches is a persistent challenge. To address it, we propose an AI Mismatch approach to anticipate and mitigate risks early on, focusing on the gap between realistic model performance and required task performance. Through an analysis of 774 AI cases, we extracted a set of critical factors, which informed the development of seven matrices that map the relationships between these factors and highlight high-risk areas. Through case studies, we demonstrate how our approach can help reduce risks in AI development.