Background: While Large Language Models (LLMs) have achieved widespread adoption, malicious prompt engineering specifically "jailbreak attacks" poses severe security risks by inducing models to bypass internal safety mechanisms. Current benchmarks predominantly focus on public safety and Western cultural norms, leaving a critical gap in evaluating the niche, high-risk domain of medical ethics within the Chinese context. Objective: To establish a specialized jailbreak evaluation framework for Chinese medical ethics and to systematically assess the defensive resilience and ethical alignment of seven prominent LLMs when subjected to sophisticated adversarial simulations. Methodology: We evaluated seven prominent models (e.g., GPT-5, Claude-Sonnet-4-Reasoning, DeepSeek-R1) using a "role-playing + scenario simulation + multi-turn dialogue" vector within the DeepInception framework. The testing focused on eight high-risk themes, including commercial surrogacy and organ trading, utilizing a hierarchical scoring matrix to quantify the Attack Success Rate (ASR) and ASR Gain. Results: A systemic collapse of defenses was observed, whereas models demonstrated high baseline compliance, the jailbreak ASR reached 82.1%, representing an ASR Gain of over 80 percentage points. Claude-Sonnet-4-Reasoning emerged as the most robust model, while five models including Gemini-2.5-Pro and GPT-4.1 exhibited near-total failure with ASRs between 96% and 100%. Conclusions: Current LLMs are highly vulnerable to contextual manipulation in medical ethics, often prioritizing "helpfulness" over safety constraints. To enhance security, we recommend a transition from outcome to process supervision, the implementation of multi-factor identity verification, and the establishment of cross-model "joint defense" mechanisms.
Six Interventions for the Responsible and Ethical Implementation of Medical AI Agents
Massachusetts General Hospital, Harvard Medical School(麻省总医院,哈佛医学院);Charité – Universitätsmedizin Berlin, corporate member of Freie Universität Berlin, Humboldt Universität zu Berlin, Institute of Medical Informatics(柏林夏里特医科大学,柏林自由大学成员,柏林洪堡大学,医学信息学研究所);Charité – Universitätsmedizin Berlin, corporate member of Freie Universität Berlin, Humboldt Universität zu Berlin, Institute for the History of Medicine and Ethics in Medicine(柏林夏里特医科大学,柏林自由大学成员,柏林洪堡大学,医学史与医学伦理学研究所);Else Kroener Fresenius Center for Digital Health, Faculty of Medicine, TUD Dresden University of Technology(德累斯顿工业大学 Else Kroener Fresenius 数字健康中心,医学院);Department of Medicine I, Faculty of Medicine, TUD Dresden University of Technology(德累斯顿工业大学医学院,第一内科系);Medical Oncology, National Center for Tumor Diseases (NCT), University Hospital Heidelberg(海德堡大学医院,国家肿瘤疾病中心 (NCT),医疗肿瘤科);Pathology & Data Analytics, Leeds Institute of Medical Research at St James's, University of Leeds(利兹大学,圣詹姆斯医学研究所,病理与数据分析系);Natera, Inc.(Natera公司)
An Ethical Trajectory Planning Algorithm for Autonomous Vehicles
Institute of Automotive Technology, Technical University of Munich(慕尼黑工业大学汽车技术研究所);Institute for Ethics in Artificial Intelligence, Technical University of Munich(慕尼黑工业大学人工智能伦理研究所)
Informed AI Regulation: Comparing the Ethical Frameworks of Leading LLM Chatbots Using an Ethics-Based Audit to Assess Moral Reasoning and Normative Values
Ethical Concerns of Generative AI and Mitigation Strategies: A Systematic Mapping Study
Faculty of Information Technology, Monash University(墨尔本大学信息科技学院);School of Science, Computing and Engineering Technologies, Swinburne University(斯威本大学科学与工程技术学院);School of Information Technology, Deakin University(德肯大学信息科技学院)
Ethical Considerations in Artificial Intelligence Courses
University of Kentucky(肯塔基大学);University of Southern California(南加州大学);University of Michigan(密歇根大学);IBM T.J. Watson Research Center(IBM T.J. 沃森研究中心);UNSW(新南威尔士大学);Data61/CSIRO(澳大利亚联邦科学与工业研究组织 Data61);TU Berlin(柏林工业大学)
ff4ERA: A new Fuzzy Framework for Ethical Risk Assessment in AI
Department of Computer Science(计算机科学系);Centro Interdipartimentale di Logica e Applicazioni (CILA)(逻辑与应用跨部门中心(CILA));Department of Life, Health and Environmental Sciences(生命、健康与环境科学系)