AI 中文总结
本研究通过美国1,503名代表性样本的在线调查，探索公众对生成式AI错误的认知、态度与检查行为，发现高度认知且常通过外部资源核验，并提出披露指导。
AI 中文摘要
生成式AI输出可能包含错误，如幻觉、无响应结果，或其他不准确或潜在有害的内容。为了探索公众对这些错误的新兴理解、态度和行为，我们在美国开展了一项在线调查，共有1,503名受访者参与，样本具有人口代表性。我们报告了公众对生成式AI错误的高度认知。此外，许多受访者表示会检查生成式AI的输出，例如通过与其他在线资源进行比较。最后，我们提出了关于生成式AI错误的解释和产品内披露的指导建议。
英文摘要
Generative AI output can contain errors, such as hallucinations, non-responsive results, or otherwise inaccurate or potentially harmful content. To explore the public's emerging understanding, attitudes, and behavior regarding such mistakes, we ran an online survey in the United States with 1,503 respondents, with a representative sample of the population. We report high public awareness of generative AI mistakes. Further, many respondents report checking generative AI output, for example, by comparing results with other online resources. We conclude with guidance for explanations and in-product disclosures about generative AI mistakes.
Comments12 pages, 7 figures, 2 tables, ancillary appendix file
Journal refP.G. Kelley, D. Feldmann, R. Jana, C. Thompson-Kuhn, A. Woodruff. Why and How People Check Generative AI Output for Mistakes. Ninth AAAI/ACM Conference on AI, Ethics, and Society, AIES, October 2026
