Whose Ground Truth? Embracing Ambiguity in Human-Centered AI

2026-10-07

AI Summary

This work addresses a fundamental limitation of traditional artificial intelligence, which relies on a single ground-truth paradigm and thereby overlooks the ambiguity and diversity inherent in human interpretation. To overcome this constraint, we propose a human-centered perspective that reconstructs the entire AI pipeline by modeling the space of plausible human judgments and introducing a mechanism to disentangle genuine ambiguity from annotation noise. This research establishes both a theoretical foundation and methodological guidance for designing AI systems that better align with the diversity of human perception. Ultimately, it advances a paradigm shift in artificial intelligence, moving away from the pursuit of a single canonical answer toward embracing multiple valid interpretations.

Abstract

As AI systems increasingly interact with people and make decisions about them, understanding human interpretations becomes an important part of developing human-centered AI. Conventional machine learning and AI systems are largely developed under the assumption that a single definitive ground truth exists, with variability in human annotations often resolved through aggregation or treated as noise. However, for many human-centered tasks, human interpretation is inherently ambiguous, and multiple interpretations of the same input may be simultaneously reasonable and valid. Reducing such ambiguity to a single target risks overlooking meaningful information about the diversity of human perception, judgment, and experience. In this position paper, we call for a shift towards modeling the interpretation space of plausible human judgments, while distinguishing meaningful ambiguity from annotation noise. We argue that this perspective should guide how AI systems are represented, learned, evaluated, deployed, and governed, supporting more human-centered AI that better reflects the diversity of human interpretation.

Problem

• Human-Centered AI
• Ground Truth
• Ambiguity
• Human Interpretation
• Annotation Noise

Innovation

• Human-Centered AI
• Ground Truth Ambiguity
• Interpretation Space
• Annotation Noise
• Human Diversity