人工智能辅助诊断中的责任归属

随着大模型、生成式人工智能等新技术在医疗领域的广泛应用，人工智能医疗面临的法律风险显现出新的样态。人工智能医疗的算法歧视和数据安全问题催生了一般人格权和具体人格权侵权风险，医疗健康数据的处置与收益分配催生数据产权纠纷，人工智能技术深度嵌入后医疗损害责任亦存在归责不确定风险。基于人工智能医疗法律风险新的样态变化，需要建立相应的算法审查机制消弭算法歧视，完善数据全生命周期管理体系确保数据安全，确立数据产权分层确权授权规则化解产权纠纷，并针对医疗损害过错情形合理分配侵权责任。

目的 为精准高效监管医疗人工智能产业提供参考。 方法 通过总结医疗人工智能诊断产品主体责任困境，梳理国内外相关研究，构建医疗人工智能诊断产品主体责任体系。 结果 构建以医疗人工智能上市许可持有人为“第一责任人”的主体责任体系，提出算法透明和可解释，分类分级监管模式、社会共治监管模式三大保障措施。 结论 医疗人工智能诊断产品主体责任体系有助于落实主体责任，打造“负责任的、有益的”人工智能，实现“自律”“善治”“有序”。

当人工智能融入医疗决策时，技术自主权的模糊界定与责任分配的多主体博弈催生出法律权责争议与伦理治理难题。本文以临床诊断、治疗策略生成及预后评估为焦点，揭示算法“黑箱”引发的数据偏倚与决策可溯性缺陷，并从责任主体界定、技术缺陷标准化评估及跨学科协同三向度提出解决方案。在法律维度上，建议基于欧盟《人工智能法案》的“高风险系统”分类与产品责任相关法规的无过错追责原则，构建开发者-操作方-医疗机构三级责任框架，即辅助型系统由操作医师承担终局责任，决策型系统则强制开发者提供可验证算法并承担连带责任。在伦理层面上，建议通过动态审查机制（如强制公开决策树逻辑与数据均衡性指标）平衡技术效能与患者知情权。

“人工智能+”行动已经成为政府工作任务之一。“人工智能+医疗”在快速发展的同时也面临着诸多风险,包括内生和衍生的技术安全风险,人机双向价值对齐困难、不同伦理诉求难以平衡、医疗数字鸿沟以及医疗不公平的伦理风险,患者隐私和个人信息权益难以保障、医疗AI的侵权责任难以界定的法律风险等。探索基于患者赋权的“人工智能+医疗”的治理路径,需要形成前端赋权、中端监管与后端追责的治理机制,即前端赋予个体权利,建立健康医疗数据权利体系和医疗AI算法权利体系;中端强化医疗AI的运行监管体系,形成自我监管、专门监管、风险监管等多种监管模式;后端从医疗AI的法律客体地位、侵权责任规则和赔偿分担机制三个方面完善医疗AI的侵权追责体系。

随着人工智能算法在自动驾驶与医疗诊断领域的深度应用，算法决策的自主性与多主体参与特征对传统侵权责任体系构成挑战。本文以“责任主体认定—归责原则适用—因果关系判定—责任承担方式”为分析框架，采用案例分析法与规范分析法，结合2022-2025年国内最新判例与《民法典》相关条款，探索算法侵权责任的类型化路径。研究发现，算法侵权责任需构建三元主体网络（开发者、运营者、使用者），归责原则应区分场景适用无过错责任（自动驾驶）与过错责任（医疗AI），因果关系采用技术关联性与法律过错双重判定标准，责任承担需创新多元化形态与社会化分担机制。研究提出，应通过制定《人工智能算法责任条例》、建立算法透明度分级制度、完善社会化分担机制等路径，实现算法创新与风险防控的动态平衡。本文的理论价值在于为算法侵权责任提供统一分析框架，实践意义在于为司法裁判与立法完善提供参考，推动人工智能产业的负责任发展。

人工智能产品相较于普通产品最大区别在于其自主性，一方面体现在人工智能产品运作方式更具复杂性，它们能够自主决策与自主行为，处于“感知—分析—行动”模式，其运行过程为“黑箱”操作，难以为人类所探究; 另一方面体现在产品侵权的致害主体更具多元性，除了作为传统产品侵权主体的生产者与销售者外，还包括研发者，甚至人工智能产品本身。基于此，当前产品缺陷的司法认定存在困境。一是现有判断标准在人工智能产品视域下缺乏适应性，由于《产品质量法》中“不合理标准”的抽象性、复杂性与《人工智能标准化白皮书(2018版)》中“技术标准”的应急性、不确定性，现有标准的使用难以解决人工智能产品缺陷的司法认定问题; 二是我国现有立法未将产品缺陷进行类型化认定，而仅对产品进行笼统的缺陷判断，在人工智能这一多元责任主体与多样未知风险的领域，该种产品缺陷的司法认定模式在适用时更为捉俭见肘。为将人工智能产品缺陷的司法实践规范化，在产品责任框架下应将缺陷划分为设计缺陷、制造缺陷、警示缺陷及跟踪观察缺陷，这不仅有利于契合人工智能产品的归责原则，也与产品侵权责任各要素相对应。人工智能产品缺陷司法认定标准应与有关主体应履行的义务相结合。具体而言，人工智能产品设计缺陷的责任主体为设计研发者，其应当履行安全保障、性能平衡、全面测试等义务，该缺陷采用“风险—效用”的司法认定规则。制造缺陷的责任主体为制造者，其应当履行配合设计预期效果与按照设计方案进行制造的义务，该缺陷采用“对预期设计偏离”的司法认定标准。警示缺陷的责任主体为制造者、销售者，警示义务应当达到警示内容的充分性、警示时间的更迭性与警示语言的简明醒目性等要求，该缺陷采用“合理充分”的司法认定标准。跟踪观察缺陷的责任主体为设计研发者、制造者与销售者，跟踪观察不仅要求畅通产品的反馈渠道，也要积极定期对产品进行监察，更要及时对问题产品作出反应，该缺陷采用“个案认定，综合判断”的司法认定原则。

医疗领域正经历以多模态大模型为特征的智能化转型，然而技术赋能与伦理风险共生并存，医疗领域大模型的深度应用使伦理风险日益凸显，亟需聚焦医疗领域大模型，全面识别其伦理风险并探索有效的治理路径。本文深入探讨了医疗领域大模型应用过程中存在的数据隐私风险、算法决策风险、主体关系风险、社会公平风险等四大核心伦理风险，并结合典型案例加以解析：提出了基于“数据 ‒ 算法 ‒ 应用 ‒ 法律”四位一体的医疗领域大模型治理框架，涵盖构建数据治理体系、创新算法治理机制、建设临床应用规范、完善法律监管框架4个方面；分析了医疗领域大模型发展面临的关键技术挑战和政策挑战。最后，展望了医疗领域大模型的未来方向，包括探索基于区块链医疗数据确权、开发轻量化模型普惠基层医疗、构建“政产学研医”协同生态系统，以期为推动医疗领域大模型技术规范健康发展、保障患者权益及完善医疗伦理治理体系提供理论与实践支撑。

随着人工智能（AI）在医疗领域的快速发展，临床辅助决策支持系统（CDSS）逐步融入临床诊疗流程。然而，CDSS在提升诊疗效率与精准性的同时也引发了复杂的责任归属界定问题。本文围绕CDSS问责困境的国内外现状展开系统讨论，从责任主体模糊、算法“黑箱”特性、数据偏倚、法律与监管制度落后与部署环境异质性5个方面深入剖析了CDSS问责困境的形成机制，并提出以全生命周期治理为核心的协同治理框架：主张通过技术路径增强AI的可信度，通过法律与监管路径构建风险分级监管与无过错赔偿基金等创新机制，通过组织路径强化机构内控，通过伦理与教育路径重塑“人机协同”的职业新范式。破解CDSS问责困境既要保障患者安全与医疗公平，也要为身处技术变革中的医务人员提供清晰的责任边界与坚实的职业保障，从而推动医疗AI技术健康发展。

… attribution prior to the development of biomedical AI … the behavior of an AI doctor that is practicing medicine independently. … In summary, medical institutions should take fault liability or …

Good decision-making is a complex endeavor, and particularly so in a health context. The possibilities for day-to-day clinical practice opened up by AI-driven clinical decision support systems (AI-CDSS) give rise to fundamental questions around responsibility. In causal, moral and legal terms the application of AI-CDSS is challenging existing attributions of responsibility. In this context, responsibility gaps are often identified as main problem. Mapping out the changing dynamics and levels of attributing responsibility, we argue in this article that the application of AI-CDSS causes diffusions of responsibility with respect to a causal, moral, and legal dimension. Responsibility diffusion describes the situation where multiple options and several agents can be considered for attributing responsibility. Using the example of an AI-driven ‘digital tumor board’, we illustrate how clinical decision-making is changed and diffusions of responsibility take place. Not denying or attempting to bridge responsibility gaps, we argue that dynamics and ambivalences are inherent in responsibility, which is based on normative considerations such as avoiding experiences of disregard and vulnerability of human life, which are inherently accompanied by a moment of uncertainty, and is characterized by revision openness. Against this background and to avoid responsibility gaps, the article concludes with suggestions for managing responsibility diffusions in clinical decision-making with AI-CDSS.

Artificial intelligence (AI) in medicine is an increasingly studied and widespread phenomenon, applied in multiple clinical settings. Alongside its many potential advantages, such as easing clinicians’ workload and improving diagnostic accuracy, the use of AI raises ethical and legal concerns, to which there is still no unanimous response. A systematic literature review on medical professional liability related to the use of AI-based diagnostic algorithms was conducted using the public electronic database PubMed selecting studies published from 2020 to 2023. The systematic review was performed according to 2020 PRISMA guidelines. The literature review highlights how the issue of liability in case of AI-related error and patient’s damage has received growing attention in recent years. The application of AI and diagnostic algorithm moreover raises questions about the risks of using unrepresentative populations during the development and about the completeness of information given to the patient. Concerns about the impact on the fiduciary relationship between physician and patient and on the subject of empathy have also been raised. The use of AI in medical field and the application of diagnostic algorithms introduced a revolution in the doctor–patient relationship resulting in multiple possible medico-legal consequences. The regulatory framework on medical liability when AI is applied is therefore inadequate and requires urgent intervention, as there is no single and specific regulation governing the liability of various parties involved in the AI supply chain, nor on end-users. Greater attention should be paid to inherent risk in AI and the consequent need for regulations regarding product safety as well as the maintenance of minimum safety standards through appropriate updates.

… We propose to cut this Gordian knot by resting liability for malfunctions on medical AI squarely on the shoulders of the health systems that deploy them, with one caveat that we'll …

The aim of this literature review was to compose a narrative review supported by a systematic approach to critically identify and examine concerns about accountability and the allocation of responsibility and legal liability as applied to the clinician and the technologist as applied the use of opaque AI-powered systems in clinical decision making. This review questions (a) if it is permissible for a clinician to use an opaque AI system (AIS) in clinical decision making and (b) if a patient was harmed as a result of using a clinician using an AIS’s suggestion, how would responsibility and legal liability be allocated? Literature was systematically searched, retrieved, and reviewed from nine databases, which also included items from three clinical professional regulators, as well as relevant grey literature from governmental and non-governmental organisations. This literature was subjected to inclusion/exclusion criteria; those items found relevant to this review underwent data extraction. This review found that there are multiple concerns about opacity, accountability, responsibility and liability when considering the stakeholders of technologists and clinicians in the creation and use of AIS in clinical decision making. Accountability is challenged when the AIS used is opaque, and allocation of responsibility is somewhat unclear. Legal analysis would help stakeholders to understand their obligations and prepare should an undesirable scenario of patient harm eventuate when AIS were used.

The growing use of artificial intelligence (AI) in health care has raised questions about who should be held liable for medical errors that result from care delivered jointly by physicians and algorithms. In this survey study comparing views of physicians and the U.S. public, we find that the public is significantly more likely to believe that physicians should be held responsible when an error occurs during care delivered with medical AI, though the majority of both physicians and the public hold this view (66.0% vs 57.3%; P = .020). Physicians are more likely than the public to believe that vendors (43.8% vs 32.9%; P = .004) and healthcare organizations should be liable for AI-related medical errors (29.2% vs 22.6%; P = .05). Views of medical liability did not differ by clinical specialty. Among the general public, younger people are more likely to hold nearly all parties liable.

The integration of artificial intelligence (AI) into healthcare in Africa presents transformative opportunities but also raises profound legal challenges, especially concerning liability. As AI becomes more autonomous, determining who or what is responsible when things go wrong becomes ambiguous. This article aims to review the legal concepts relevant to the issue of liability for harm caused by AI in healthcare. While some suggest attributing legal personhood to AI as a potential solution, the feasibility of this remains controversial. The principal–agent relationship, where the physician is held responsible for AI decisions, risks reducing the adoption of AI tools due to potential liabilities. Similarly, using product law to establish liability is problematic because of the dynamic learning nature of AI, which deviates from static products. This fluidity complicates traditional definitions of product defects and, by extension, where responsibility lies. Exploring alternatives, risk-based determinations of liability, which focus on potential hazards rather than on specific fault assignments, emerges as a potential pathway. However, these, too, present challenges in assigning accountability. Strict liability has been proposed as another avenue. It can simplify the compensation process for victims by focusing on the harm rather than on the fault. Yet, concerns arise over the economic impact on stakeholders, the potential for unjust reputational damage, and the feasibility of a global application. Instead of approaches based on liability, reconciliation holds much promise to facilitate regulatory sandboxes. In conclusion, while the integration of AI systems into healthcare holds vast potential, it necessitates a re-evaluation of our legal frameworks. The central challenge is how to adapt traditional concepts of liability to the novel and unpredictable nature of AI—or to move away from liability towards reconciliation. Future discussions and research must navigate these complex waters and seek solutions that ensure both progress and protection.

… for malfunctions or failures stemming from medical AI. It is known that people show … liability risk from using health care artificial intelligence tools. The New England Journal of Medicine …

The adoption of advanced artificial intelligence (AI) systems in healthcare is transforming the healthcare-delivery landscape. Artificial intelligence may enhance patient safety and improve healthcare outcomes, but it presents notable ethical and legal dilemmas. Moreover, as AI streamlines the analysis of the multitude of factors relevant to malpractice claims, including informed consent, adherence to standards of care, and causation, the evaluation of professional liability might also benefit from its use. Beginning with an analysis of the basic steps in assessing professional liability, this article examines the potential new medical-legal issues that an expert witness may encounter when analyzing malpractice cases and the potential integration of AI in this context. These changes related to the use of integrated AI, will necessitate efforts on the part of judges, experts, and clinicians, and may require new legislative regulations. A new expert witness will be likely necessary in the evaluation of professional liability cases. On the one hand, artificial intelligence will support the expert witness; however, on the other hand, it will introduce specific elements into the activities of healthcare workers. These elements will necessitate an expert witness with a specialized cultural background. Examining the steps of professional liability assessment indicates that the likely path for AI in legal medicine involves its role as a collaborative and integrated tool. The combination of AI with human judgment in these assessments can enhance comprehensiveness and fairness. However, it is imperative to adopt a cautious and balanced approach to prevent complete automation in this field.

The application of Artificial Intelligence (AI) facilitates medical activities by automating routine tasks for healthcare professionals. AI augments but does not replace human decision-making, thus complicating the process of addressing legal responsibility. This study investigates the legal challenges associated with the medical use of AI in radiology, analyzing relevant case law and literature, with a specific focus on professional liability attribution. In the case of an error, the primary responsibility remains with the physician, with possible shared liability with developers according to the framework of medical device liability. If there is disagreement with the AI’s findings, the physician must not only pursue but also justify their choices according to prevailing professional standards. Regulations must balance the autonomy of AI systems with the need for responsible clinical practice. Effective use of AI-generated evaluations requires knowledge of data dynamics and metrics like sensitivity and specificity, even without a clear understanding of the underlying algorithms: the opacity (referred to as the “black box phenomenon”) of certain systems raises concerns about the interpretation and actual usability of results for both physicians and patients. AI is redefining healthcare, underscoring the imperative for robust liability frameworks, meticulous updates of systems, and transparent patient communication regarding AI involvement.

A discussion concerning whether to conceive Artificial Intelligence (AI) systems as responsible moral entities, also known as “artificial moral agents” (AMAs), has been going on for some time. In this regard, we argue that the notion of “moral agency” is to be attributed only to humans based on their autonomy and sentience, which AI systems lack. We analyze human responsibility in the presence of AI systems in terms of meaningful control and due diligence and argue against fully automated systems in medicine. With this perspective in mind, we focus on the use of AI-based diagnostic systems and shed light on the complex networks of persons, organizations and artifacts that come to be when AI systems are designed, developed, and used in medicine. We then discuss relational criteria of judgment in support of the attribution of responsibility to humans when adverse events are caused or induced by errors in AI systems.

The use of artificial intelligence (AI) systems in healthcare provides a compelling case for a re-examination of ‘gross negligence’ as the basis for criminal liability. AI is a smart agency, often using self-learning architectures, with the capacity to make autonomous decisions. Healthcare practitioners (HCPs) will remain responsible for validating AI recommendations but will have to contend with challenges such as automation bias, the unexplainable nature of AI decisions, and an epistemic dilemma when clinicians and systems disagree. AI decisions are the result of long chains of sociotechnical complexity with the capacity for undetectable errors to be baked into systems, which introduces a new dimension of moral luck. The ‘advisory’ nature of AI decisions constructs a legal fiction, which may leave HCPs unjustly exposed to the legal and moral consequences when systems fail. On balance, these novel challenges point towards a legal test of subjective recklessness as the better option: it is practically necessary; falls within the historic range of the offence; and offers clarity, coherence, and a welcome reconnection with ethics.

… We close with a few thoughts about liability for medical AI writ large. First, and broadest, this is a space in flux; we have laid out the workings of generally applicable law, but there …

The advent of artificial intelligence (“AI”) holds great potential to improve clinical diagnostics. At the same time, there are important questions of liability for harms arising from the use of this technology. Due to their complexity, opacity, and lack of foreseeability, AI systems are not easily accommodated by traditional liability frameworks. This difficulty is compounded in the health care space where various actors, namely physicians and health care organizations, are subject to distinct but interrelated legal duties regarding the use of health technology. Without a principled way to apportion responsibility among these actors, patients may find it difficult to recover for injuries. In this Article, I propose that physicians, manufacturers of clinical AI systems, and hospitals be considered a common enterprise for the purposes of liability. This proposed framework helps facilitate the apportioning of responsibility among disparate actors under a single legal theory. Such an approach responds to concerns about the responsibility gap engendered by clinical AI technology as it shifts away from individualistic notions of responsibility, embodied by negligence and products liability, toward a more distributed conception. In addition to favoring plaintiff recovery, a common enterprise strict liability approach would create strong incentives for the relevant actors to take care.

… The use of AI-enabled medical robots in healthcare settings poses a host of legal and ethical questions, especially regarding the attribution of liability for the injuries and deaths caused …

… concerning AI tort liability and, drawing on the fundamental principles of tort law, seeks to develop a framework that both accommodates the structural and technical characteristics of AI …

… This paper examines the challenges of establishing tort liability for AI-related … AI in both regions, including an analysis of the challenges of applying traditional tort liability principles to AI-…

Determining liability if a patient suffers an injury due to AI or robotic application use is complex. The study used comparative legal analysis and examined the current legal frameworks and whether applying current principles of medical liability is adequate to determine liability arising from AI and robotics-related medical malpractice. The study identifies the gaps in current legislation and proposes applying new legal doctrines such as vicarious liability, custodian liability, corporate liability, strict liability, and finally, applying the benefit and burden rule. AI and robotics will revolutionize medical care. However, new legislation is necessary to mitigate risk and ensure patient safety.

… gives an incorrect medical diagnosis, or AI that gives a patient an incorrect dose of medicine. … Finally, I will concentrate on tort liability, because in contractual settings liability is usually …

… by AIrelated torts. This article critiques the limitations of the product liability approach to AI torts and takes a holistic view of the legal difficulties introduced by emerging AI technologies. It …

… presented with an AI-generated diagnostic or prognostic result occupies a challenging position and must justify: (1) why they followed the diagnosis or treatment suggested by the AI; or (…

… Given the slow progress in reducing diagnostic errors, not adopting new tools … , AI poses challenges for applying tort principles. Because it is primarily plaintiffs who will struggle, liability …

Background: Artificial intelligence (AI) is applied in numerous areas of society and has also led to significant changes in the field of medicine. Medicine is a branch of science of exceptional importance, and it is therefore necessary to ensure a high level of patient protection. The quality of healthcare has significantly improved through the use of artificial intelligence in various stages of the medical process, from the analysis of medical data and diagnostics, through therapy planning, to patient monitoring and the management of healthcare systems. Objective: The aim of this paper is to analyze the civil law aspects of artificial intelligence in medicine, with a particular focus on questions of liability for damage resulting from the use of such systems. Methods: The paper will first present the basic characteristics and areas of application of artificial intelligence in medicine, and then examine potential sources of damage and the legal basis for the liability of various stakeholders, including AI system manufacturers, software developers and data providers, healthcare institutions, and healthcare professionals. Special attention will be given to the challenges of proving causation and allocating liability in situations where decisions are made or supported by autonomous algorithmic systems. Results and Discussion: However, at the same time, numerous legal issues arise, particularly in the field of civil liability in cases where the application of artificial intelligence results in harm to a patient. Given the great importance of medicine and the need to ensure a high level of patient protection, the application of artificial intelligence must be accompanied by appropriate legal protection. The paper gives answers to a number of questions, with particular emphasis on the question of who may be held liable for damage caused by the use of artificial intelligence in medicine, as well as under which regulations and in what manner such liability is determined. Conclusion: Artificial intelligence has the potential to significantly enhance medical practice, its application must be accompanied by appropriate legal mechanisms that ensure patient protection and clearly define the responsibility of all participants in the system. Future legal development in this area will likely focus on further adapting existing civil law institutions to the specificities of artificial intelligence, while simultaneously strengthening preventive risk management mechanisms and transparency of AI systems in medicine.

… an AI-system, the most likely avenue to establishing tort liability … diagnosis in terms of success rates. The standard of medical diagnostic care would then likely require reliance on an AI…

Background/Objectives: Artificial intelligence-based software as a medical device (AI-SaMD) refers to AI-powered software used for medical purposes without being embedded in physical devices. Despite increasing approvals over the past decade, research in this domain—spanning technology, healthcare, and national security—remains limited. This research aims to bridge the existing research gap in AI-SaMD by systematically reviewing the literature from the past decade, with the aim of classifying key findings, identifying critical challenges, and synthesizing insights related to technological, clinical, and regulatory aspects of AI-SaMD. Methods: A systematic literature review based on the PRISMA framework was performed to select the relevant AI-SaMD studies published between 2015 and 2024 in order to uncover key themes such as publication venues, geographical trends, key challenges, and research gaps. Results: Most studies focus on specialized clinical settings like radiology and ophthalmology rather than general clinical practice. Key challenges to implement AI-SaMD include regulatory issues (e.g., regulatory frameworks), AI malpractice (e.g., explainability and expert oversight), and data governance (e.g., privacy and data sharing). Existing research emphasizes the importance of (1) addressing the regulatory problems through the specific duties of regulatory authorities, (2) interdisciplinary collaboration, (3) clinician training, (4) the seamless integration of AI-SaMD with healthcare software systems (e.g., electronic health records), and (5) the rigorous validation of AI-SaMD models to ensure effective implementation. Conclusions: This study offers valuable insights for diverse stakeholders, emphasizing the need to move beyond theoretical analyses and prioritize practical, experimental research to advance the real-world application of AI-SaMDs. This study concludes by outlining future research directions and emphasizing the limitations of the predominantly theoretical approaches currently available.

… legal analysis of post-market provisions in the AIA and MDR, using a case study of a high-risk Class III AI SaMD … increase the operational burden and liability concerns. Similar concerns …

… SaMD; the individuation of the healthcare situation or condition the SaMD is intended for; and the description of the SaMD’… and legal scholars to rethink liability or to apply current liability …

This chapter illustrates how the US legal system, through tort law, may impose responsibility on software vendors and their users for possible harm that medical software might cause to …

… of typical SaMD are not well-suited to address technological advancements in AI-based SaMD … with an experienced attorney in the field to avoid potential legal liability in the future [41]. …

… strict liability and fault-based liability regime currently proposed in the revised Product Liability Directive (PLD) and the AI Liability … around legal liability for harms involving AI systems. …

… This lack of transparency raises concerns regarding legal liability in the event of incorrect outputs by AI/ML SaMD. In general, reviews of medical error-related lawsuits against AI …

Abstract Artificial intelligence (AI) could revolutionise health care, potentially improving clinician decision making and patient safety, and reducing the impact of workforce shortages. However, policymakers and regulators have concerns over whether AI and clinical decision support systems (CDSSs) are trusted by stakeholders, and indeed whether they are worthy of trust. Yet, what is meant by trust and trustworthiness is often implicit, and it may not be clear who or what is being trusted. We address these lacunae, focusing largely on the perspective(s) of clinicians on trust and trustworthiness in AI and CDSSs. Empirical studies suggest that clinicians’ concerns about their use include the accuracy of advice given and potential legal liability if harm to a patient occurs. Onora O’Neill’s conceptualisation of trust and trustworthiness provides the framework for our analysis, generating a productive understanding of clinicians’ reported trust issues. Through unpacking these concepts, we gain greater clarity over the meaning ascribed to them by stakeholders; delimit the extent to which stakeholders are talking at cross purposes; and promote the continued utility of trust and trustworthiness as useful concepts in current debates around the use of AI and CDSSs.

AI-enabled clinical decision support (CDS) is increasingly embedded in diagnosis and care pathways, yet liability remains unclear when recommendations cause harm. This paper offers a structured, actionable approach. We synthesize liability regimes across the United States, European Union, and Asia-Pacific and highlight gaps specific to adaptive, software-driven systems. We ground the analysis in ethical risks-accountability shifts, automation bias, and power asymmetries-that should shape how responsibility is shared. We then introduce a quantitative Tiered Liability Model that allocates total harm among developers, physicians, and hospitals using policy weights and three measurable indices: developer culpability, physician oversight, and AI explainability. We prove basic feasibility and monotonicity properties of the allocation, explain how to operationalize the indices with audits, EHR-based oversight signals, and logging/traceability, and provide a regulatory crosswalk that maps regional doctrines to these indices. Two scenario analyses-a pneumonia misdiagnosis and a hospital operations system that contributes to understaffing-illustrate how the model yields fair, incentive-aligned allocations and how improvements in explainability and oversight shift liability. To support adoption, we include a ready-to-use procurement scorecard with measurable evidence and pass/fail gates, together with an implementation roadmap and regulatory levers (metrics standardization, safe harbors, post-market surveillance). The result is a coherent framework that protects patients, sustains innovation, and gives regulators, providers, and vendors a common language for assigning responsibility.

INTRODUCTION Recent developments in the field of Artificial Intelligence (AI) applied to healthcare promise to solve many of the existing global issues in advancing human health and managing global health challenges. This comprehensive review aims not only to surface the underlying ethical and legal but also social implications (ELSI) that have been overlooked in recent reviews while deserving equal attention in the development stage, and certainly ahead of implementation in healthcare. It is intended to guide various stakeholders (eg. designers, engineers, clinicians) in addressing the ELSI of AI at the design stage using the Ethics by Design (EbD) approach. METHODS The authors followed a systematised scoping methodology and searched the following databases: Pubmed, Web of science, Ovid, Scopus, IEEE Xplore, EBSCO Search (Academic Search Premier, CINAHL, PSYCINFO, APA PsycArticles, ERIC) for the ELSI of AI in healthcare through January 2021. Data were charted and synthesised, and the authors conducted a descriptive and thematic analysis of the collected data. RESULTS After reviewing 1108 papers, 94 were included in the final analysis. Our results show a growing interest in the academic community for ELSI in the field of AI. The main issues of concern identified in our analysis fall into four main clusters of impact: AI algorithms, physicians, patients, and healthcare in general. The most prevalent issues are patient safety, algorithmic transparency, lack of proper regulation, liability & accountability, impact on patient-physician relationship and governance of AI empowered healthcare. CONCLUSIONS The results of our review confirm the potential of AI to significantly improve patient care, but the drawbacks to its implementation relate to complex ELSI that have yet to be addressed. Most ELSI refer to the impact on and extension of the reciprocal and fiduciary patient-physician relationship. With the integration of AIbased decision making tools, a bilateral patient-physician relationship may shift into a trilateral one.

Making good decisions in extremely complex and difficult processes and situations has always been both a key task as well as a challenge in the clinic and has led to a large amount of clinical, legal and ethical routines, protocols and reflections in order to guarantee fair, participatory and up-to-date pathways for clinical decision-making. Nevertheless, the complexity of processes and physical phenomena, time as well as economic constraints and not least further endeavours as well as achievements in medicine and healthcare continuously raise the need to evaluate and to improve clinical decision-making. This article scrutinises if and how clinical decision-making processes are challenged by the rise of so-called artificial intelligence-driven decision support systems (AI-DSS). In a first step, this article analyses how the rise of AI-DSS will affect and transform the modes of interaction between different agents in the clinic. In a second step, we point out how these changing modes of interaction also imply shifts in the conditions of trustworthiness, epistemic challenges regarding transparency, the underlying normative concepts of agency and its embedding into concrete contexts of deployment and, finally, the consequences for (possible) ascriptions of responsibility. Third, we draw first conclusions for further steps regarding a ‘meaningful human control’ of clinical AI-DSS.

Policy Points With increasing integration of artificial intelligence and machine learning in medicine, there are concerns that algorithm inaccuracy could lead to patient injury and medical liability. While prior work has focused on medical malpractice, the artificial intelligence ecosystem consists of multiple stakeholders beyond clinicians. Current liability frameworks are inadequate to encourage both safe clinical implementation and disruptive innovation of artificial intelligence. Several policy options could ensure a more balanced liability system, including altering the standard of care, insurance, indemnification, special/no-fault adjudication systems, and regulation. Such liability frameworks could facilitate safe and expedient implementation of artificial intelligence and machine learning in clinical care.

An increasing number of automated and artificial intelligence (AI) systems make medical treatment recommendations, including personalized recommendations, which can deviate from standard care. Legal scholars argue that following such nonstandard treatment recommendations will increase liability in medical malpractice, undermining the use of potentially beneficial medical AI.  However, such liability depends in part on lay judgments by jurors: when physicians use AI systems, in which circumstances would jurors hold physicians liable? Methods: To determine potential jurors’ judgments of liability, we conducted an online experimental study of a nationally representative sample of 2,000 U.S. adults. Each participant read 1 of 4 scenarios in which an AI system provides a treatment recommendation to a physician. The scenarios varied the AI recommendation (standard or nonstandard care) and the physician’s decision (to accept or reject that recommendation). Subsequently, the physician’s decision caused harm. Participants then assessed the physician’s liability. Results: Our results indicate that physicians who receive advice from an AI system to provide standard care can reduce the risk of liability by accepting, rather than rejecting, that advice, all else being equal. However, when an AI system recommends nonstandard care, there is no similar shielding effect of rejecting that advice and so providing standard care. Conclusion: The tort law system is unlikely to undermine the use of AI precision medicine tools and may even encourage the use of these tools.