AI-Augmented Design and the Expertise Bias in Subjective Evaluations of Creative Output

• 成果类型: Article
• 署名作者: Samet, Jordan A.; Williamson, Michael G.; Yip, Michael A.
• 署名单位: Indiana University System; IU Kelley School of Business; Indiana University Bloomington; University of Illinois System; University of Illinois Urbana-Champaign; University System of Georgia; University of Georgia
• 刊物名称: ACCOUNTING REVIEW
• ISSN/ISSBN: 0001-4826; 1558-7967
• DOI: 10.2308/TAR-2024-0592
• 发表日期: 2026-09
• 页码: 445-465
• 关键词: Artificial intelligence; Subjective evaluation; expertise bias; creativity; balanced scorecard; performance; MODEL

摘要

Results from multiple experiments demonstrate that evaluators more favorably evaluate creative output produced by designers with higher expertise, even when the underlying creativity of the output is held constant (hereafter, expertise bias). We find, however, that this bias is mitigated when evaluators know that Artificial Intelligence (AI) can augment creative design processes, because AI's capabilities reduce the perceived exclusivity of designers' domain expertise. We also show that designers can restore the perceived exclusivity of their expertise, reestablishing the expertise bias, by choosing not to use available AI tools. Although prior research focuses on AI's ability to enhance or inhibit the creativity of output, we highlight that it can enhance the creative process by mitigating a prevalent human bias in the subjective evaluation of this output. We also contribute to a better understanding of why some experienced designers refuse to utilize AI.