Key Challenges and Limitations of AI in Healthcare

Abstract

Artificial Intelligence (AI) has become a critical catalyst in transforming healthcare through improved diagnostics, predictive analytics, and clinical decision support. However, the path toward effective integration of AI in clinical and organizational contexts is riddled with significant challenges and inherent limitations. This research critically examines the multifaceted barriers that shape AI’s implementation and reliability in healthcare systems. The study identifies key challenges across seven domains—data integrity and availability, technical and systemic integration, economic and resource constraints, human and ethical considerations, organizational and innovation barriers, regulatory and governance complexities, and emerging operational difficulties. Beyond these external challenges, it explores intrinsic limitations categorized as cognitive and algorithmic, ethical and humanistic, regulatory and socio-structural, and environmental and generative. Through this comprehensive analysis, the paper underscores that while AI holds immense potential to enhance the precision, efficiency, and accessibility of healthcare, its success ultimately depends on addressing these layered constraints through transparent design, inclusive policy frameworks, and sustained human oversight.

Introduction

Healthcare systems are becoming increasingly strained due to population growth, aging societies, and rising chronic disease burdens, creating the need for smarter, more efficient, and sustainable solutions. Artificial Intelligence (AI) has emerged as a transformative technology in healthcare, offering capabilities such as early disease detection, robotic-assisted surgery, personalized treatment, and improved drug development through machine learning, deep learning, and other advanced models. However, despite its strong potential, AI adoption in healthcare also introduces serious risks and limitations.

The major concern areas are categorized into key challenges and inherent limitations. Key challenges include data issues (fragmented, low-quality, and privacy-restricted healthcare data), technical barriers (integration with legacy hospital systems, lack of explainability, and poor standardization), and economic constraints (high costs, unequal access, and unclear return on investment). Additional challenges involve ethical and human factors such as bias, lack of trust, accountability issues, and reduced clinician autonomy; organizational barriers like poor leadership, resistance to change, and lack of collaboration; regulatory gaps due to inconsistent global policies; and emerging operational risks from advanced AI tools like generative models and real-time decision systems.

The text also highlights core limitations of AI, particularly in healthcare. These include algorithmic and cognitive limits, where AI relies on pattern recognition rather than true understanding, leading to issues like the black-box problem, algorithmic bias, and model drift. Ethical and societal concerns arise from lack of transparency, unequal performance across populations, and difficulties in ensuring fairness and informed consent. Socio-structural and environmental constraints further complicate adoption and sustainability.

Conclusion

Instead of existing as distinct issues, the difficulties and constraints posed by AI in healthcare are part of a continuum. Adoption is slowed by issues like data fragmentation, expenses, and legal restrictions, yet AI\'s inherent potential and bounds are defined by its limits. Realistic expectations and responsible use are made possible by acknowledging this continuum. For instance, explainability tools and hybrid human-AI decision models can help reduce black-box opacity, even though it cannot be completely removed. While algorithmic solutions cannot address AI\'s lack of empathy, human-AI cooperation can maintain clinical compassion. Every medical development calls for new ethical frameworks; technological advancement by itself does not guarantee ethical advancement. Resilient AI systems that value accountability, transparency, and human supervision should thus be given top priority in the healthcare industry. Adaptive governance, explainable algorithms, digital integrity, environmental sustainability, and inclusive international cooperation are necessary for future paths. The ultimate objective is a mutually beneficial collaboration in which AI enhances human judgment while remaining grounded in moral obligation and compassion, promoting medical treatment that upholds both technology advancement and human dignity.

References

[1] Zaidi SF, Shaikh A, Surani S. The pulse of AI: implementation of Artificial Intelligence in healthcare and its potential hazards. Open Respir Med J. 2024;18:e18743064289936. [2] Pagallo U, O\'Sullivan S, Nevejans N, et al. The underuse of AI in the health sector: opportunity costs, success stories, risks and recommendations. Health Technol (Berl). 2024;14(1):1-14. doi:10.1007/s12553-023-00806-7 [3] Siafakas, Nikolaos, and Eirini Vasarmidi. \"Risks of Artificial Intelligence (AI) in Medicine.\" Pneumon 37.3 (2024). [4] Omar Ali, Wiem Abdelbaki, Anup Shrestha, Ersin Elbasi, Mohammad Abdallah Ali Alryalat, Yogesh K Dwivedi,A systematic literature review of artificial intelligence in the healthcare sector: Benefits, challenges, methodologies, and functionalities,Journal of Innovation & Knowledge,Volume 8, Issue1,2023 [5] Topol, Eric. Deep medicine: how artificial intelligence can make healthcare human again. Hachette UK, 2019. [6] Onishchenko, G. G. \"Development of the risk analysis methodology given the current safety challenges for public health in the Russian Federation: vital issues and prospects.\" Health Risk Analysis 4 (2023): 4-18. [7] Healthcare, R. T. A. D. \"Bridging the gap: Overcoming data, technological, and human roadblocks to AI-driven healthcare transformation.\" J. Manag 8.1 (2021): 7-14. [8] Stanfill, Mary H., and David T. Marc. \"Health information management: implications of artificial intelligence on healthcare data and information management.\" Yearbook of medical informatics 28.01 (2019): 056-064. [9] Corrales Compagnucci, Marcelo, and Mark Fenwick. \"A Multidisciplinary Perspective on Cross-Border Health Data Transfers: Privacy, Risks and Solutions.\" International Transfers of Health Data: A Global Perspective. Singapore: Springer Nature Singapore, 2025. 1-15. [10] Dunn Cavelty M. Breaking the cyber-security dilemma: aligning security needs and removing vulnerabilities. Sci Eng Ethics. 2014;20(3):701-715. [11] Herrin J, Dempsey BJ 3rd. Web-enabled medical databases: a threat to security?. Methods Inf Med. 2000;39(4-5):298-302. [12] Acuña, Edwin Gerardo Acuña. \"Healthcare cybersecurity: Data poisoning in the age of ai.\" Journal of Comprehensive Business Administration Research (2024). [13] Sharkey A, Sharkey N. Granny and the robots: ethical issues in robot care for the elderly Ethics Inf Technol. 2012;14:27–40. [14] Korica, Petra, Neamat El Gayar, and Wei Pang. \"Explainable artificial intelligence in healthcare: Opportunities, gaps and challenges and a novel way to look at the problem space.\" International conference on intelligent data engineering and automated learning. Cham: Springer International Publishing, 2021. [15] Mohammadi, Samaneh, et al. \"Balancing privacy and performance in federated learning: A systematic literature review on methods and metrics.\" Journal of Parallel and Distributed Computing 192 (2024): 104918. [16] https://amzur.com/blog/ai-in-healthcare-transformation-how-to-calculate-ai-roi-in-healthcare/[last accessed on 10 October 2025] [17] Ueda D, Walston SL, Fujita S, Fushimi Y, Tsuboyama T, Kamagata K, Yamada A, Yanagawa M, Ito R, Fujima N, Kawamura M, Nakaura T, Matsui Y, Tatsugami F, Fujioka T, Nozaki T, Hirata K, Naganawa S. Climate change and artificial intelligence in healthcare: Review and recommendations towards a sustainable future. Diagn Interv Imaging. 2024 Nov;105(11):453-459. [18] Ahmed MI, Spooner B, Isherwood J, Lane M, Orrock E, Dennison A. A Systematic Review of the Barriers to the Implementation of Artificial Intelligence in Healthcare. Cureus. 2023 Oct 4;15(10):e46454. [19] https://studenttheses.uu.nl/bitstream/handle/20.500.12932/48841/20250221_Final%20thesis_FlorentineBeelaertsvanBlokland_publ.pdf?sequence=1[last accessed on 10 October 2025] [20] Petersson, L., Larsson, I., Nygren, J.M. et al. Challenges to implementing artificial intelligence in healthcare: a qualitative interview study with healthcare leaders in Sweden. BMC Health Serv Res 22, 850 (2022). [21] Li, Xianmiao & Abangbila, Linda. (2024). Resistance to medical artificial intelligence: Integrating AI awareness, AI risks, and displacement of responsibility. Journal of Infrastructure, Policy and Development. 8. 7923 [22] Mennella C, Maniscalco U, De Pietro G, Esposito M. Ethical and regulatory challenges of AI technologies in healthcare: A narrative review. Heliyon. 2024 Feb [23] Angus DC, Khera R, Lieu T, et al. AI, Health, and Health Care Today and Tomorrow: The JAMA Summit Report on Artificial Intelligence. JAMA. Published online October 13, 2025. doi:10.1001/jama.2025.18490 [24] Xu, H., & Shuttleworth, K. M. J. (2024). Medical artificial intelligence and the black box problem: a view based on the ethical principle of “do no harm”. Intelligent Medicine, 4, 52–57. [25] Chan, B. (2023). Black-box assisted medical decisions: AI power vs. ethical physician care. Medicine, Health Care and Philosophy, 26, 285–292. [26] Cross, J. L., Choma, M. A., & Onofrey, J. A. (2024). Bias in medical AI: Implications for clinical decision-making. PLOS Digital Health, 3(11), e0000651. [27] Mittermaier, M., Raza, M. M., & Kvedar, J. C. (2023). Bias in AI-based models for medical applications: challenges and mitigation strategies. npj Digital Medicine, 6(113). [28] Abdul Razak, M. S., Nirmala, C. R., Sreenivasa, B. R., Lahza, H., & Lahza, H. F. M. (2023). A survey on detecting healthcare concept drift in AI/ML models from a finance perspective. Frontiers in Artificial Intelligence, 5, 955314. [29] Guan, H., Bates, D., & Zhou, L. (2025). Keeping Medical AI Healthy: A Review of Detection and Correction Methods for System Degradation. arXiv preprint arXiv:2506.17442v1. [30] Montemayor, C., Halpern, J., & Fairweather, A. (2022). In principle obstacles for empathic AI: why we can’t replace human empathy in healthcare. AI & Society, 37(3), 1353–1359. [31] Pantanowitz, L., et al. (2024). Regulatory Aspects of Artificial Intelligence and Machine Learning. Modern Pathology, 37, 100609. [32] Vardas, E. P., Marketou, M., & Vardas, P. E. (2025). Medicine, Healthcare and the AI Act: Gaps, Challenges and Future Implications. European Heart Journal – Digital Health, 6(4), 833–839. [33] Strubell, E., Ganesh, A., & McCallum, A. (2024). Energy and Policy Considerations for Deep Learning in AI Applications. Communications of the ACM, 67(2), 35–43. [34] Winsta, J. (2025). The Hidden Costs of AI: A Review of Energy, E-Waste, and Inequality in Model Development. arXiv preprint arXiv:2507.09611v1. [35] National Academies of Sciences, Engineering, and Medicine. (2024). Generative Artificial Intelligence in Health Care: Opportunities, Risks, and Governance. Washington, DC: The National Academies Press.

Copyright

Copyright © 2026 Arpna Saxena, Shruti Jain, Aman Gupta. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.