Ethical, Legal, And Societal Implications Of AI In Medicine
Author: LawSphereX
Category: Technology, Cyber & Privacy
Published:
Updated:
ETHICAL, LEGAL, AND SOCIETAL IMPLICATIONS OF AI
IN MEDICINE
Kirti Upadhyaya, National Forensic Sciences University
ABSTRACT
Incorporating artificial intelligence (AI) into healthcare is a revolutionary paradigm change beyond technical progress to cover basic issues regarding medical practice, human connections, and societal values(Faiyazuddin et al. , 2025). This thorough study looks at artificial intelligence use in important medical areas, including pathology, mental health, drug discovery, robotic surgery, infectious disease control, radiology, genetic engineering, and clinical medicine(Ahmad et al. , 2021). The research indicates that although AI systems exhibit astounding powers in diagnostic accuracy, therapy optimization, and healthcare accessibility, their application produces intricate sociotechnical consequences that affect human agency, professional identity, legal accountability, and social standards.
The main research question revolves around the complex difficulties of ethical AI integration in medicine, where conventional systems have trouble fitting algorithmic decision-making methods that are usually opaque and probabilistic instead of deterministic. The study goals are identifying important ethical, legal, and social effects of medical artificial intelligence across several specialties; investigating the ongoing issue of algorithmic bias impacting healthcare equality; analysing the change of doctor-patient relationships through algorithmic intermediaries; and assessing legal structures in handling medical liability and accountability related to AI. One major knowledge gap concerns how to strike an equilibrium between the efficiency increases provided by artificial intelligence and the maintenance of compassionate, individualized care that defines conventional medical practice.
The research questions How does algorithmic bias in medical AI contribute to healthcare disparities across diverse populations? What legal frameworks are necessary to ensure clear accountability in AI-assisted medical decision- making? How is the traditional doctor-patient relationship being transformed by AI-mediated clinical interactions? What ethical safeguards are essential for the responsible use of AI in sensitive areas such as genomics and mental health? Analysing ethical systems and regulatory difficulties, the method uses a thorough qualitative study of artificial intelligence applications across several medical fields by looking at case studies, including historic legal precedents like Daubert v . Merrell Dow Pharmaceuticals(Solomon, and Hackett, 1996) and recent AI inventorship cases like Thaler v. Vidal.
Through methodical means of bias detection and reduction, thorough legal framework development, best human-AI partnership models, and global equity initiatives ensuring equal distribution of AI benefits, the value of this study resides in providing critical advice for ethical AI implementation in healthcare. This study is subject to certain limitations. First, the rapid pace of artificial intelligence development implies that some of the solutions presented for regulation or technology may soon become outdated. Second, the study could be restricted by its reliance on qualitative analysis and secondary sources. Third, the cross-jurisdictional legal scene is complicated and changing, therefore, the extrapolation of legal concepts across several nations presents ambiguities.
Finally, AI applications in healthcare frequently include proprietary algorithms that limit access to internal decision-making logic and call transparency of analysis into question. In conclusion, this research finds that successful medical AI integration calls on correcting algorithmic bias, developing thorough legal accountability systems, preserving human-centred care relationships, guaranteeing worldwide equity in AI access, and deploying ongoing monitoring systems to assess AI performance and unforeseen effects in healthcare delivery(Tariq, 2023).
1. Introduction
Incorporating artificial intelligence to healthcare signals a significant technical revolution in the annals of medicine. As IBM Watson's first oncology use shows and modern deep learning models capable of detecting diabetic retinopathy with superhuman accuracy attest, artificial intelligence systems are becoming vital to medical practice(Ahmad et al. , 2021). Lower diagnostic errors, individualized treatments, greater efficiency, and expanded access to highquality care are among the ongoing healthcare problems this shift seeks to address. However, medical AI implementation has complex sociotechnical implications that affect human agency, professional identity, legal accountability, and social standards besides purely technological problems(Jian et al. , 2021).
Other sectors in which artificial intelligence mostly influences business processes, medical artificial intelligence directly affects human health, life-and-death decisions, and personal components of human care. This study of artificial intelligence integration across essential medical domains reveals particular problems and opportunities needing dedicated approaches for responsible use in the areas of ethical governance, legal systems, and social adaptation.
2 Clinical Medicine
The Foundation of Medical AI
2.1 Applications
Including diagnostic support systems, treatment recommendation engines, and patient monitoring platforms, clinical medicine uses artificial intelligence most widely in healthcare(Mirbabaie, Stieglitz and Frick, 2021). Modern artificial intelligence systems can examine enormous volumes of patient data—including real-time physiological monitoring, imaging studies, laboratory results, and electronic health records—to spot patterns undetectable to human doctors and propose the best treatment plans. Examples include various clinical decision support systems helping doctors identify rare diseases or maximize medicine dosages, Google's sepsis prediction algorithm implemented in hospitals, and IBM Watson for Oncology's treatment suggestions.
These systems guarantee that treatment choices depend on the most current evidence and thorough data analysis while also helping to lower diagnostic errors, which afflict roughly 12 million Americans each year.
2.2 Ethical consequences
Ethical issues in clinical artificial intelligence centre on the basic issue of trust and the preservation of human agency in medical decision-making. When an AI system suggests a particular treatment, patients and physicians must wrestle with questions of reliability, transparency, and ultimate responsibility. The black box character of several machine learning techniques raises specific difficulties for informed consent since patients might not quite grasp how suggestions are produced(Starke et al. , 2021). When AI systems interpret data beyond human understanding, the patient's right to make informed decisions about their care—the principle of medical autonomy—becomes complex. How can patients really give informed consent for AI-assisted therapies when the logic underlying such treatments may be opaque even to their doctors?
This difficulty is made worse by the possibility that AI systems could perpetuate or amplify current prejudices in medical treatment, so particularly impacting vulnerable groups already facing discrimination in healthcare settings(Kiener, 2020). Still another important ethical aspect is trust. Although research indicates that in some diagnostic operations artificial intelligence systems can outperform human doctors, public faith in these systems is still low. Built on centuries of human contact and professional training, the link between physician and patient must now include algorithmic intermediaries whose decision-making procedures could be radically different from human thinking.
2.3 Legal Implications
The legal landscape for clinical AI remains largely uncharted, with conventional malpractice systems struggling to accommodate algorithmic decision-making. Liability determination becomes complex: Are physicians liable for following AI recommendations? Are healthcare systems responsible for flawed implementations? Should AI developers bear responsibility for algorithmic errors? Current malpractice law holds doctors accountable for medical errors below professional standards, requiring independent judgment regardless of AI output. However, proving causality in AI-related harm proves challenging since AI outcomes are probabilistic rather than deterministic, and AI models are often incomprehensible.
Existing legal frameworks assume human decision-makers are responsible for their choices. AI systems challenge this as nonhuman agents with probabilistic processes. New legal standards for AI accountability will likely require statutory action and extensive case law development. Regulatory approval faces additional complexities. While the FDA develops AI/ML medical device standards, technological advancement outpaces traditional timelines. Continuously learning AI systems raise questions about when regulatory approval should be renewed (Feng, Emerson & Simon, 2019).
AI-generated evidence admissibility in malpractice cases will likely face scrutiny under the Daubert vs Merrell Dow Pharmaceuticals (1993)(Solomon, Hackett, 1996) standard, which requires scientific evidence based on reliable, testable methods with known error rates and peer acceptance, challenging criteria for opaque AI systems.
2.4 Societal Results
Beyond personal patient-physician contacts, the social effects of clinical artificial intelligence include basic changes in healthcare delivery and professional identity. By algorithmic intermediaries, the classic model of medical care based on the doctor's clinical judgment and personal relationship with patients is changing. This change raises concerns about medical treatment dehumanization. Patients may feel their care is directed by machines rather than compassionate human specialists who understand their particular conditions and values. AI's promised efficiency benefits could come at the expense of the therapeutic relationship, long regarded as crucial for healing. Healthcare disparities represent another critical social issue.
Training AI systems on data from wealthy, educated populations with good healthcare access could cause underperformance for underprivileged groups, aggravating existing health inequalities. Ensuring AI systems function effectively across all communities requires deliberate effort and considerable resources that commercial developers may not prioritize.
3. Radiology and Medical Imaging
3.1 Technological Advancement and Applications
With profound learning algorithms showing amazing ability in picture recognition and interpretation, radiology has become the forefront of artificial intelligence adoption in medicine(Tiwari, Mishra, and Kuo, 2025). Today, artificial intelligence systems frequently help diagnose cancer in mammograms, locate fractures in X-rays, examine CT scans for stroke symptoms, and analyse MRI images for several pathologies. FDA-approved AI tools developed by companies like Aidoc, Zebra Medical Vision, and Google Health are aggressively used in clinical contexts globally. Artificial intelligence in radiology is alluring: vast volumes of medical imaging data are wellsuited to machine learning methods; pattern recognition jobs defining much of radiology align precisely with artificial intelligence skills.
Studies have demonstrated that in certain jobs, especially those requiring fast processing of great numbers of pictures, artificial intelligence (AI) systems can match or surpass human radiologist performance(Plesener et al. , 2023).
3.2 Ethical Considerations
A primary ethical concern involves training data bias and its impact on diagnostic precision across diverse patient populations. Many AI systems rely on datasets that disproportionately feature specific demographic groups—primarily affluent, white patients from leading academic hospitals—while inadequately representing minorities, rural populations, and individuals from developing nations(Jha et al. , 2023). This bias manifests in multiple ways: AI algorithms may demonstrate reduced effectiveness in identifying conditions in patients with darker skin pigmentation, fail to account for anatomical differences common among certain ethnic populations, or be developed using imaging technologies and protocols that differ from those available in resource-constrained settings. Such algorithmic bias risks perpetuating and intensifying existing healthcare inequalities(Nazer et al. , 2023).
The principle of diagnostic equity demands that medical technologies perform with equal effectiveness across all patient demographics. Meeting this standard requires diverse training datasets, comprehensive validation studies across various populations, and ongoing bias monitoring in clinical practice. Unfortunately, the commercial drivers behind AI development may conflict with these ethical imperatives, particularly when serving marginalized populations offers limited financial returns.
3.3 Legal Framework
and Responsibility Legal accountability in radiological AI presents unique challenges due to specialized image analysis and potential for small errors with significant consequences. Determining liability becomes complex when AI systems fail to detect cancer or generate false positives, as assigning fault requires understanding technical complexities exceeding traditional legal expertise. Standard care becomes ambiguous when AI assistance exists but isn't universally adopted. Should radiologists face heightened expectations when AI tools are accessible? When AI systems err, can radiologists justify reliance on technology as reasonable? These issues remain unresolved, requiring extensive legal precedent and regulatory clarification.
3.4 Professional Transformation and Societal Impact
Societal implications extend beyond individual patient care to fundamental changes within the medical profession. Radiology, traditionally a highly compensated specialty requiring extensive training, now faces questions about radiologists' future roles and medical education structures. AI systems handling routine interpretations present significant challenges for practicing radiologists. Job displacement represents substantial concern, especially for early-career professionals. While AI may not completely replace human radiologists, it will fundamentally reshape responsibilities toward complex cases, patient interaction, and AI oversight. Healthcare institutions implementing AI gain competitive advantages in accuracy and efficiency, pressuring others to adopt similar systems.
However, substantial implementation costs may prevent smaller hospitals from accessing advanced capabilities, potentially widening healthcare quality disparities.
4. Genetic Engineering and Precision Medicine
The convergence of artificial intelligence (AI), genetic engineering, and genomics marks a transformative frontier in personalized medicine. These technologies promise earlier disease detection, individualized treatments, and genetically tailored therapies, while raising profound ethical, legal, and societal concerns—particularly regarding heritable genetic modifications and "designer babies."
4.1 Artificial intelligence in genomic medicine
By allowing quick analysis of enormous genetic data, artificial intelligence systems transform genomics. Using artificial intelligence for polygenic risk scores forecasting disease susceptibility, pharmacogenomic analysis informing drug selection, and tumour profiling identifying targeted cancer treatments, platforms like 23andMe, Foundation Medicine, and Tempus These programs let practitioners identify faint genetic patterns, therefore enabling precision medicine on previously unattainable levels(Chen at el., 2025).
4.2 Ethical ramifications in genetic engineering and genomic artificial intelligence
A. Genomic Artificial Intelligence: Discrimination, Privacy, and Consent Critical moral questions arise from artificial intelligence in genomic medicine: Genetic Privacy: Sensitive information on people and family members is revealed by genetic data. Unintentionally, artificial intelligence systems could reveal nonpaternity, susceptibility to stigmatized diseases, or information on nonconsenting relatives. Informed Consent: Consent gets complex when data gathered for one goal—ancestry testing— is later applied for research or training in artificial intelligence. The changing character of artificial intelligence questions conventional onetime consent models(Wang et al. , 2016). Gaps still exist in life insurance, disability coverage, and employment fields outside of the scope of GINA (2008), the Genetic Information Non-discrimination Act. B.
Designer Babies and Genetic Engineering Targeted human embryo alterations are made possible by AI-enhanced gene editing tools like CRISPRCas9(Danti et al. , 2025). Designer babies—changing embryos for nonmedical qualities—raises issues even while current efforts focus on single gene disorders: • Embryos cannot give future autonomy and permission for irreversible genetic modifications influencing them and their offspring. Commodification of human life: Trait selection risks treating children as products, echoing eugenic ideas.
Social Stratification: Wealthy only access might result in a genetically privileged elite. • AI driven editing may cause unexpected mutations or multigenerational consequences. Genetic damage • Cultural Resistance: Some societies see intrusions as against natural or heavenly laws.
C. He Jiankui Scandal (China, 2018)(Liu, Shi, and Xu, 2024) Lacking peer supervision and ethical review, Dr. He Jiankui announced twin girls with modified CCR5 genes for HIV resistance. The case was condemned worldwide for ignoring consent procedures and exposing upcoming generations to unknown risks. He was sentenced three years in jail, first criminal conviction for illegal human gene editing. Worldwide discussion on this case sparked calls for global regulatory systems and moratoria on germline editing.
4.3 Consequences for society and equity of access
AI genomics helps to ensure that resources are distributed unfairly, therefore creating new health inequalities: Genomic databases include more European heritage samples, therefore lowering the precision of artificial intelligence technologies for underrepresented groups. Access Disparities: While sequencing expenses dropped, interpretative tools and clinical implementation stay high, therefore producing two-tier healthcare systems. Religious or cultural beliefs shape communities' genetic testing participation, hence culturally sensitive, trust based medical treatments are needed. Movements favouring local control over genomic data, opposing extractive research methods, and advancing equity in AI genomics research call for Indigenous Data Sovereignty.5. Pathology: Digital Transformation of Tissue Analysis
5. AI Applications in Pathology
Pathology, the medical field concentrating on diagnosis of disease by tissue, cellular, and fluid analysis, has embraced artificial intelligence as a strong tool for image analysis and pattern recognition. Digital pathology combined with AI enables automated analysis of histopathological images, therefore maybe increasing diagnostic accuracy, uniformity, and efficiency. Automated cancer detection and grading in tissue samples, quantification of biomarkers for therapy selection, analysis of cellular morphology, and prediction of treatment responses based on tissue properties are among AI applications in pathology. While research institutions keep creating focused algorithms for particular disease applications, corporations like Path AI, Paige, and Proscia have created FDA approved artificial intelligence technologies for many pathological purposes(Ahmad et al., 2021).
5.1 Ethical Considerations in Pathological Artificial Intelligence
The validity and reliability of AI diagnoses relative to human pathologists raises the main ethical issue in pathological AI. Pathological diagnosis requires subtle analysis of tissue architecture, cell morphology, and environmental elements that AI systems may struggle to fully grasp. The possibility of false positives or negatives in AI-assisted pathology could have major repercussions for patient care. Human supervision becomes crucial in pathological AI. Should AI be used only as a screening tool with human confirmation, or can AI systems make independent diagnostic decisions in specific settings? Establishing appropriate standards requires thorough consideration of patient safety and professional responsibilities, depending on the particular application, AI system accuracy, and availability of human expertise. Further ethical issues come from quality assurance and validation.
To guarantee uniform performance, pathological artificial intelligence systems have to be verified across various tissue kinds, colouring techniques, and imaging conditions. The commercial demands pushing artificial intelligence growth, nevertheless, might drive businesses to concentrate on popular uses instead of uncommon disorders or difficult diagnostic situations where artificial intelligence support may be most helpful.
5.2 Legal and Intellectual Property Problems
Complex issues on intellectual property rights and data ownership abound in the legal scene for pathological AI. Pathological images used to train artificial intelligence systems frequently reflect large institutional investments in tissue collection, processing, and expert annotation. Questions about equitable compensation and ownership of the resulting intellectual property arise when these pictures are employed to create commercial artificial intelligence products(Jackson et al. , 2021). Using patient tissue specimens for AI research creates privacy and consent concerns. Although pathological study is routine medical care, using tissue images for commercial AI development may exceed typical patient approval limits. Creating suitable permission systems that enable beneficial AI development while upholding patient rights requires careful balance.
Professional responsibility in pathological AI poses particular difficulties due to specialized diagnosis nature and potential for small errors with major repercussions. Establishing requirements for AI validation, quality control, and responsible use requires coordination between pathologists, AI developers, and regulatory bodies.
5.3 Society Influence on Pathology Practice
Incorporating AI into pathology has far-reaching effects for both the profession and healthcare delivery. Particularly in developing nations and remote regions lacking skilled pathologists, AI systems could alleviate shortages by enabling efficient utilization of existing expertise and extending services to underserved areas. AI deployment raises questions about professional displacement and pathologist role alterations. Pathologists may need new skills in AI supervision, challenging case review, and integrating AI insights with clinical context, requiring changes in professional development and medical education. Resource allocation becomes critical as AI technologies require significant investments in digital infrastructure, software licensing, and staff training. Smaller hospitals may struggle with AI adoption, causing variations in diagnostic capabilities.
6 Mental Health and Psychiatry
The Human Touch in Digital Care
6.1 AI Applications in Mental Healthcare
Mental health represents one of the most sensitive and complex applications of AI in medicine. AI systems are being developed for chatbots for therapy and counselling, mood monitoring through smartphone apps and wearable devices, natural language processing for suicide risk assessment, and predictive modelling for treatment response in psychiatric conditions(Poudel et al., 2025). Companies like Woebot, Wysa, and Ginger have developed AI-powered mental health platforms that provide automated therapy sessions, mood tracking, and crisis intervention. Research applications include analysis of social media posts for depression screening, voice analysis for psychiatric diagnosis, and brain imaging analysis for treatment selection in conditions like depression and schizophrenia(Poudel et al., 2025).
6.2 Ethical Challenges in Mental Health AI
Ethical implications of AI in mental health are particularly complex due to patient vulnerability and the deeply personal nature of psychological treatment. Consent becomes complicated when patients experience impaired judgment from mental health conditions, and long-term implications of AI-assisted care may not be fully understood. Emotional manipulation represents a significant ethical concern. AI systems designed for therapeutic conversations may use persuasive techniques considered manipulative, particularly when users develop emotional attachments to AI systems. The potential for exploiting psychological vulnerabilities raises questions about the ethics of automated emotional engagement in mental healthcare settings. AI's inability to grasp human nuance presents another critical challenge.
While AI systems may recognize patterns in language or behaviour indicating mental health risks, they may lack the empathy, cultural understanding, and contextual awareness essential for effective mental health treatment(Poudel et al. , 2025). The risk is that AI systems may provide inappropriate advice or miss critical nuances that human therapists would recognize. Privacy and confidentiality take on heightened importance in mental health AI. Mental health information is among the most sensitive personal data, and patients may be reluctant to seek care if they believe their information could be misused. AI systems that continuously monitor behaviour through smartphones raise particular concerns about surveillance and privacy.
6.3 Legal Implications and Responsibility
Legal responsibility in AI-assisted mental health care presents unprecedented challenges. When an AI system fails to recognize suicide risk or provides advice that contributes to harm, determining liability involves complex questions about the standard of care, professional responsibilities, and AI system limitations. The licensing and regulation of mental health AI systems remains largely undefined. Traditional mental health practice is regulated through professional licensing boards and established ethical codes, but AI systems operating independently may fall outside these regulatory frameworks. Developing appropriate oversight mechanisms requires collaboration between technology developers, mental health professionals, and regulatory agencies(Coghlan et al., 2023).
6.4 Societal Implications and Digital Divides
Societal implications include promises of increased access and concerns about quality and equity. AI-powered mental health tools could support underserved populations lacking access to traditional services, potentially addressing severe shortages of mental health professionals. However, the digital divide may limit access for populations needing support most. Individuals without reliable internet, modern smartphones, or digital literacy skills may be excluded from AI-assisted care. Cultural and linguistic barriers may limit AI effectiveness for diverse populations. Concerns exist about AI tools replacing rather than supplementing human connection. While AI provides convenient, accessible support, it cannot replace fundamental human needs for genuine relationships and community support essential for mental health and wellbeing.
7 Medical Research and Drug Discovery
Innovation and Intellectual Property
7.1 AI in Pharmaceutical Research
AI has transformed pharmaceutical research and drug discovery by accelerating the identification of promising drug candidates, optimizing clinical trial design, and predicting drug interactions and side effects. Machine learning algorithms can analyse vast molecular databases to identify potential therapeutic compounds, predict their properties, and design clinical trials that are more likely to succeed(Abbas et al. , 2024). AI systems can reduce the time required for drug discovery from years to months by virtually screening millions of molecular compounds, predicting their biological activity, and identifying the most promising candidates for further testing. In clinical trial matching, AI can identify patients who are most likely to benefit from experimental treatments, reducing recruitment time and improving trial success rates (Abbas et al. , 2024).
Advanced AI applications include de novo drug design, where algorithms create entirely new molecular structures optimized for specific therapeutic targets, and drug repurposing, where AI identifies new applications for existing medications. These approaches promise to significantly reduce the cost and time required to bring new treatments to market.
7.2 Ethical Implications in AI-Driven Research
The legal landscape for clinical AI remains largely uncharted, with conventional malpractice systems struggling to accommodate algorithmic decision-making. Liability determination becomes complex: Are physicians liable for following AI recommendations? Are healthcare systems responsible for implementing flawed systems? Should AI developers bear responsibility for algorithmic errors? Current malpractice law holds doctors accountable for medical errors below professional standards, requiring independent judgment regardless of AI output. However, proving causality in AI-related harm proves challenging since AI outcomes are probabilistic rather than deterministic, and AI models are often incomprehensible. Existing legal frameworks assume human decision-makers are responsible for their choices. AI systems challenge this as nonhuman agents with probabilistic processes.
New legal standards will likely require statutory action and extensive case law development. Regulatory approval faces complexities as technological advancement outpaces traditional FDA timelines. AI-generated evidence admissibility will face scrutiny under Daubert standards, requiring reliable methods—challenging criteria for opaque AI systems.
7.3 Legal and Intellectual Property Challenges
Legal Structures for AI Generated Drug Discovery AI-generated drug discovery creates novel intellectual property challenges. Traditional patent law assumes human inventors, making inventorship and ownership determination difficult when AI systems create molecular structures or identify medicinal uses. Classic Cases of Intellectual Property Diamond v. Chakrabarty (1980) established that genetically modified life forms can be patented, paving the way for new biological innovations. However, this case predates AI involvement in drug development and raises questions about how AI-generated compounds should be treated under patent law (Diamond v. Chakrabarty, 447 U. S. 303 (1980)). Thaler v. Vidal (2022) addressed whether AI systems can be listed as inventors on patent applications.
The Federal Circuit Court of Appeals ruled that under US patent law, only natural persons can be inventors, though questions remain about human inventors who use AI tools extensively in their discovery process(Thaler v. Vidal, 43 F. 4th 1207 (Fed. Cir. 2022)). Cases of AI Inventorship The Dabus decisions across several jurisdictions (US, UK, EU) demonstrated that while AI systems cannot be inventors, the level of human involvement required for patentability remains largely unclear. In Thaler v. Hirshfeld (2021), the USPTO maintained that inventors must be human beings, but the rights of AI developers versus researchers using AI tools remain complex(Thaler v. Hirshfeld, 558 F. Supp. 3d 238 (E. D. Va. 2021)).
Recent court rulings have clarified that AI systems cannot be named as inventors on patent applications, though issues persist regarding inventorship assignment when human researchers use AI tools extensively. The extent of human involvement needed for patent eligibility and the rights of AI creators versus researchers remain largely unresolved. Trade Secrets and Transparency As companies seek to protect their AI algorithms and training databases, trade secret law may become increasingly important in AI-driven drug development. However, using trade secrets in pharmaceutical research raises questions about the transparency and reproducibility required for scientific advancement and regulatory approval. Evolving Regulatory Framework As the FDA develops guidelines for AI-designed drugs, the Chevron U. S. A. , Inc. v.
Natural Resources Defence Council (1984) doctrine of judicial deference to administrative agencies becomes relevant. The recent Loper Bright Enterprises v. Raimondo (2024) ruling limiting Chevron deference may influence how courts review FDA decisions about AI-generated pharmaceutical products(Blanco-Gonzalez et al. , 2022). Approval processes must adapt for AI-designed medications. Regulatory agencies need frameworks for evaluating safety and efficacy of AI-discovered compounds, particularly when discovery processes may be difficult to document or replicate conventionally.
7.4 Society's influence on healthcare access
The primary social effects of AI in drug discovery centre on access, equity, and AI's potential to either reduce or exacerbate healthcare disparities. AI systems that accelerate drug research could lower development costs, potentially expanding medication availability. However, if AIdriven discovery primarily targets profitable markets and diseases, benefits may not be equally distributed. The global nature of AI drug discovery raises questions about benefit distribution across nations and populations. If AI systems are trained on data from multiple countries but resulting medications are predominantly sold in wealthy nations, concerns arise about equitable compensation and access for contributing communities(Silva, 2024). Economic incentives driving drug research may influence equity in AI-developed therapies.
AI systems may Favor profitable markets rather than rare diseases affecting low-income populations. Policy interventions or alternative funding mechanisms may be necessary to ensure AI benefits diverse communities. AI-driven discovery speed may create new disparities, with some groups accessing treatments faster than others. Managing global distribution requires international collaboration and new frameworks for technology transfer.
8. Robotic Surgery and Support Technologies
8. 1 The Transparency and Trust Challenge
Including artificial intelligence in robotic surgery presents formerly unheard-of moral issues regarding the black box problem. Both patients and surgeons have difficulties in grasping the logic behind important operations when artificial intelligence systems direct surgical decisions via murky algorithms. This opacity challenges the fundamental tenet of informed consent— how can patients really agree to treatments when even their doctors may not fully grasp the process by which the artificial intelligence reaches decisions? In surgical situations where quick decisions can affect patient results, the preservation of human agency becomes especially challenging. Surgeons have to balance AI recommendations with their clinical judgment while still bearing entire accountability for patient care.
This produces an ethical conflict between using the computational benefits of artificial intelligence and protecting human knowledge and intuition that have traditionally directed surgical approach.
8.2 Legal Duty in the Operating Room
The legal ramifications of AI-assisted surgery bring fresh difficulties for conventional malpractice systems. Liability distribution gets really difficult when artificial intelligence involvement causes surgical errors. The legal system has to answer whether surgeons are accountable for following incorrect AI recommendations, whether healthcare institutions are responsible for using poorly tested systems, or whether AI creators should be held responsible for algorithm failures. Present legal requirements stipulate that doctors must preserve independent clinical judgment whatever the aid given by artificial intelligence. Proving causation in AI related surgical damages is difficult, nevertheless, as AI results are often probabilistic rather than deterministic.
Likely applying to AI-generated surgical data, the Daubert standard for admissibility of scientific evidence will call for established standards for reliability, testability, peer review, known error rates, and scientific approval of AI systems.
8.3 Transformation of Surgical Care by Society
Artificial intelligence assisted robotic surgery challenges to radically change the doctor-patient relationship that has defined medical treatment for generations. Patients can think their treatment is increasingly influenced by equipment rather than caring human experts who appreciate their individual situation. The gains in efficiency from artificial intelligence could come at the expense of losing the therapeutic relationship itself. Since artificial surgical systems trained mostly on well-resourced people may perform poorly for underserved areas, health inequities present a serious issue. This could worsen pre-existing inequalities in surgical results since wealthy medical centres may only be able to provide sophisticated AI-assisted procedures.
9 Microbiology and Infectious Diseases
AI in Public Health
9.1 Moral Aspects of Disease Monitoring
Applications of artificial intelligence in microbiology and infectious disease management bring up serious moral issues about privacy, monitoring, and public health power. AI-powered illness monitoring systems can analyse enormous volumes of health data to forecast outbreaks and follow pathogen propagation, but this ability generates questions about personal privacy and possible overreach by public health agencies. The COVID19 epidemic showed both the promise and danger of artificial intelligence in infectious disease control. Although AI systems provided benefits for contact tracing, exposure notification, and resource distribution, they also brought questions about government monitoring and the erosion of civil rights under the guise of public health(Surianarayanan, and Chelliah, 2021).
9.2 Legal Systems Governing Aid-Driven Public Health
The legal environment for artificial intelligence in infectious disease management is mostly uncharted territory. Traditional public health law has to accommodate AI systems capable of processing data and recommending beyond human knowledge. Emergency use permissions for AI diagnostic tools during health crises create particular difficulties since regulatory agencies must balance speed of deployment with safety and effectiveness demands(Akhtar, 2025).
The worldwide character of infectious disease epidemics calls for coordinated AI answers that current legal systems may not fully enable. International collaboration in AI-driven disease monitoring raises difficult legal issues about data sovereignty, cross border information exchange, and harmonized regulatory standards.
9.3 AI Disease Management's Social Impact
AI in microbiology might help to lessen or worsen worldwide health inequalities. Although AI systems could democratize access to complex diagnostic tools in resource-limited environments, they could also foster new kinds of technological dependence and exacerbate inequalities between countries with varying AI capabilities . The fight against antimicrobial resistance is a perfect illustration of these difficulties. Although fair global access to these technologies remains unknown, artificial intelligence systems able to quickly identify resistant pathogens and recommend alternative treatments could dramatically improve patient outcomes. Aids discovered antimicrobial chemicals raises questions regarding intellectual property rights and the equilibrium between global health needs and incentives for invention(Lastra et al., 2024).
10. Conclusion
The introduction of artificial intelligence into medicine marks a paradigm change that goes beyond mere technological progress and addresses crucial issues pertaining to healthcare, human interactions, and cultural values. Despite legal precedents such Daubert v. Merrell Dow Pharmaceuticals (1993) starting to address algorithmic decision-making, algorithmic bias is the most persistent problem across all medical AI applications, ranging from clinical decision support tools that perpetuate healthcare disparities to radiology systems that perform poorly on underrepresented populations. Establishing liability and accountability in AI-assisted healthcare is another universal issue since traditional legal frameworks struggle to account for probabilistic and frequently opaque AI decision-making processes.
AI's impact on healthcare relationships is its most significant consequence because the relationship between doctor and patient, which has historically been founded on trust and empathy, must now accommodate algorithmic intermediaries with fundamentally different reasoning processes. This creates a tension between efficiency and humanity that is present in all areas where AI offers improvements in diagnostic accuracy and accessibility, but perhaps at the expense of the personal, empathetic care that characterizes traditional medical practice.
As the future of medical AI depends on addressing these multifaceted challenges while utilizing the transformative potential of technology, success depends on systematic bias detection and mitigation, the creation of a comprehensive legal framework via legislative action and the evolution of case law, the development of optimal human AI collaboration models, global equity initiatives to ensure equitable distribution of AI benefits, and continuous monitoring of the performance and unforeseen effects of AI systems.
REFERENCES
Abbas, M. K. G. (2024). The Role of AI in Drug Discovery. ChemBioChem. https://doi. org/10. 1002/cbic. 202300816 Ahmad, Z. (2021). Artificial intelligence in medicine, current applications and future role with emphasis on pathology. Diagnostic Pathology. https://doi. org/10. 1186/s13000-021-01085-4 Bin Akhtar, Z. (2025). Artificial intelligence within medical diagnostics: A multi-disease perspective. Artificial Intelligence in Health. https://doi. org/10. 36922/aih. 5173 Blanco-Gonzalez, A. (2022). The Role of AI in Drug Discovery: Challenges, Opportunities, and Strategies. arXiv. https://doi. org/10. 48550/ARXIV. 2212. 08104 Coghlan, S. (2023). To chat or bot to chat: Ethical issues with using chatbots in mental health. Digital Health. https://doi. org/10. 1177/20552076231183542 da Silva, R. G. L. (2024).
The advancement of artificial intelligence in biomedical research and health innovation in emerging economies. Globalization and Health. https://doi. org/10. 1186/s12992-024-01049-5 Danti, L. (2025). CRISPR/Cas9-mediated activation of NR5A1 steers human embryonic stem cells towards steroidogenic fate. Journal of Ovarian Research. https://doi. org/10. 1186/s13048- 025-01655-w de la Lastra, J. M. P. et al. (2024). : Leveraging AI and Machine Learning in Combating Antimicrobial Resistance – a Comprehensive Review. Journal of Medical Systems, 48(1). https://doi. org/10. 1007/s10916-024-02089-5 Faiyazuddin, M. (2025). The Impact of Artificial Intelligence on Healthcare: Comprehensive Review of Advancements in Diagnostics and Treatment. Health Science Reports. https://doi. org/10. 1002/hsr2. 70312 Feng, J. (2019).
Approval policies for modifications to Machine Learning-Based Software as Medical Device. arXiv. https://doi. org/10. 48550/ARXIV. 1912. 12413 Jackson, B. R. (2021). The Ethics of Artificial Intelligence in Pathology and Laboratory Medicine. Academic Pathology. https://doi. org/10. 1177/2374289521990784 Jha, D. (2023). A Conceptual Algorithm for Applying Ethical Principles of AI to Medical Practice. arXiv. https://doi. org/10. 48550/ARXIV. 2304. 11530 Jiang, L. (2021). Opportunities and challenges of artificial intelligence in the medical field. Journal of International Medical Research. https://doi. org/10. 1177/03000605211000157 Kiener, M. (2020). Artificial intelligence in medicine and the disclosure of risks. AI & SOCIETY. https://doi. org/10. 1007/s00146-020-01085-w Liu, Z. (2024). He Jiankui's unprecedented offense: how CRISPR-babies scandal reshaped legal governance in China.
Journal of Responsible Innovation. https://doi. org/10. 1080/23299460. 2024. 2372116 Mirbabaie, M. (2021). Artificial intelligence in disease diagnostics: Critical review and classification. Health and Technology. https://doi. org/10. 1007/s12553-021-00555-5 Nazer, L. H. (2023). Bias in artificial intelligence algorithms and recommendations for mitigation. PLOS Digital Health. https://doi. org/10. 1371/journal. pdig. 0000278 Plesner, L. L. (2023). Autonomous Chest Radiograph Reporting Using AI: Estimation of Clinical Impact. Radiology. https://doi. org/10. 1148/radiol. 222268 Poudel, U. (2025). AI in Mental Health: Review of Technological Advancements and Ethical Issues in Psychiatry. Issues in Mental Health Nursing. https://doi. org/10. 1080/01612840. 2025. 2502943 Solomon, S. M. (1996). Setting Boundaries between Science and Law: Lessons from Daubert v. Merrell Dow Pharmaceuticals.
Science, Technology, & Human Values. https://doi. org/10. 1177/016224399602100201 Surianarayanan, C. (2021). Leveraging Artificial Intelligence Capabilities for COVID-19 Containment. New Generation Computing. https://doi. org/10. 1007/s00354-021-00128-0 Tiwari, A. (2025). Current AI technologies in cancer diagnostics and treatment. Molecular Cancer. https://doi. org/10. 1186/s12943-025-02369-9 Wang, S. (2016). Genome privacy: challenges, technical approaches, and ethical considerations in the United States. Annals of the New York Academy of Sciences. https://doi. org/10. 1111/nyas. 13259 Daubert v. Merrell Dow Pharmaceuticals, Inc. , 509 U. S. 579 (1993) Diamond v. Chakrabarty, 447 U. S. 303 (1980) Chevron U. S. A. , Inc. v. Natural Resources Defence Council, 467 U. S. 837 (1984) Loper Bright Enterprises v. Raimondo, 601 U. S. ___ (2024) Tariq, Z.
(2023). Integrating AI and Humanities in Healthcare. arXiv. https://doi.org/10.48550/ARXIV.2302.07081 Thaler v. Vidal, 43 F.4th 1207 (Fed. Cir. 2022) Thaler v. Hirshfeld, 558 F. Supp. 3d 238 (E.D. Va. 2021) Thaler v. Commissioner of Patents, [2021] FCA 879 (Australia) Thaler v. Comptroller-General of Patents, [2021] EWCA Civ 1374 (UK) Stephen Thaler v. European Patent Office, T 1870/18 and T 1871/18 (EPO) Genetic Information Non-discrimination Act, Pub. L. No. 110-233 (2008) People v. He Jiankui, Shenzhen Nanshan District People's Court (China, 2019)