arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2507.08030cs.CLcs.CEcs.HC
收藏

生成式AI模型中医疗安全信息减少的系统性分析

A Systematic Analysis of Declining Medical Safety Messaging in Generative AI Models

Department of Medicine, University of British Columbia(英属哥伦比亚大学医学系);Department of Biomedical Data Science, Stanford School of Medicine(斯坦福医学院生物医学数据科学系);University of California, Berkeley(加州大学伯克利分校);University of California, San Francisco(加州大学旧金山分校);Department of Dermatology, Stanford School of Medicine(斯坦福医学院皮肤科)

Sonali Sharma, Ahmed M. Alaa, Roxana Daneshjou

AI总结:

本研究分析了2022至2025年LLM和VLM输出中医疗免责声明的减少趋势,发现其存在率大幅下降,强调需实施适应临床语境的免责声明作为安全保障。

摘要:

生成式AI模型,包括大型语言模型(LLM)和视觉语言模型(VLM),越来越多地被用于解释医学图像和回答临床问题。其回答通常包含不准确之处;因此,医疗免责声明等安全措施对于提醒用户AI输出未经专业审查或不能替代医疗建议至关重要。本研究评估了2022年至2025年各代LLM和VLM输出中是否存在免责声明。使用500张乳腺X光片、500张胸部X光片、500张皮肤科图像和500个医疗问题,对输出内容筛选了免责声明短语。LLM和VLM输出中的医疗免责声明存在率分别从2022年的26.3%降至2025年的0.97%,以及从2023年的19.6%降至2025年的1.05%。到2025年,大多数模型未显示任何免责声明。随着公共模型变得更强大和权威,必须将免责声明作为一种保障措施来实施,以适应每次输出的临床语境。

英文摘要:

Generative AI models, including large language models (LLMs) and vision-language models (VLMs), are increasingly used to interpret medical images and answer clinical questions. Their responses often include inaccuracies; therefore, safety measures like medical disclaimers are critical to remind users that AI outputs are not professionally vetted or a substitute for medical advice. This study evaluated the presence of disclaimers in LLM and VLM outputs across model generations from 2022 to 2025. Using 500 mammograms, 500 chest X-rays, 500 dermatology images, and 500 medical questions, outputs were screened for disclaimer phrases. Medical disclaimer presence in LLM and VLM outputs dropped from 26.3% in 2022 to 0.97% in 2025, and from 19.6% in 2023 to 1.05% in 2025, respectively. By 2025, the majority of models displayed no disclaimers. As public models become more capable and authoritative, disclaimers must be implemented as a safeguard adapting to the clinical context of each output.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑