📄 中文摘要
近年来，随着生物人工智能库的不断涌现，这些库承诺能够自动化基因组分析、药物发现、医学影像或临床数据解读。然而，单靠基准测试无法充分评估这些技术的可靠性和安全性。STEM-AI 的建立旨在对这些医疗人工智能库进行审计，以确保其在临床应用中的信任度。通过系统性的方法，STEM-AI 不仅关注技术的性能指标，还考虑了伦理、透明度和可解释性等重要因素，为医疗领域的人工智能应用提供了更全面的评估标准。
医疗人工智能库需要的不仅仅是基准测试，我们建立了 STEM-AI 来审计信任
📄 English Summary
Medical AI Repositories Need More Than Benchmarks. We Built STEM-AI to Audit Trust
In recent years, the emergence of bio-AI repositories has promised to automate genomic analysis, drug discovery, medical imaging, and clinical data interpretation. However, relying solely on benchmarks is insufficient to assess the reliability and safety of these technologies. The establishment of STEM-AI aims to audit these medical AI repositories to ensure their trustworthiness in clinical applications. Through a systematic approach, STEM-AI focuses not only on performance metrics but also considers critical factors such as ethics, transparency, and interpretability, providing a more comprehensive evaluation standard for the application of AI in the medical field.