Beyond Behaviorist Representational Harms: A Plan for Measurement and Mitigation
This work addresses representational harms in AI systems, particularly for vulnerable groups affected by large language models, though it appears incremental in building on existing fairness research.
The paper tackles the problem of representational harms in algorithms by expanding definitions beyond behavioral aspects to include cognitive and affective states, and proposes a framework for measurement and mitigation with a case study illustration.
Algorithmic harms are commonly categorized as either allocative or representational. This study specifically addresses the latter, focusing on an examination of current definitions of representational harms to discern what is included and what is not. This analysis motivates our expansion beyond behavioral definitions to encompass harms to cognitive and affective states. The paper outlines high-level requirements for measurement: identifying the necessary expertise to implement this approach and illustrating it through a case study. Our work highlights the unique vulnerabilities of large language models to perpetrating representational harms, particularly when these harms go unmeasured and unmitigated. The work concludes by presenting proposed mitigations and delineating when to employ them. The overarching aim of this research is to establish a framework for broadening the definition of representational harms and to translate insights from fairness research into practical measurement and mitigation praxis.