Review Article

AI and Metahuman-Based Educational Application in Healthcare: Opportunities and Challenges

Bok Sil Hong¹, Myoung-Rye Kim²

Abstract

Background/Objectives: Artificial intelligence (AI) and Metahuman technologies are emerging as transformative innovations in healthcare education. Traditional approaches such as lectures and conventional simulation-based training face limitations in scalability, cost, and learner engagement. AI-driven Metahuman systems combine cognitive intelligence with hyper-realistic embodiment, creating opportunities for more personalized, immersive, and globally accessible training. This review synthesizes current applications, opportunities, and limitations of AI and Metahuman-based education, with an emphasis on implications for medical and nursing training.

Methods: A structured literature search was conducted using PubMed, Scopus, and Web of Science for studies published between 2019 and 2025. Keywords included “AI professor,” Metahuman,” “digital human,” “AI tutor,” and “healthcare education.” Twenty-three peer-reviewed articles, reports, and case studies were selected.

Results: Applications were identified in three domains: (1) medical and nursing education, using virtual patients and AR-based tutors for clinical skills; (2) public health and team training, applying AI lecturers and Metahumans in crisis and interprofessional simulations; and (3) continuing professional development, offering adaptive and multilingual AI tutors. Opportunities included enhanced immersion, individualized learning, cost efficiency, and improved accessibility. Limitations involved technical instability, ethical risks such as bias and privacy concerns, faculty and student skepticism, and a lack of rigorous evidence on clinical outcomes.

Conclusions: AI and Metahuman-based education represents a paradigm shift in healthcare training. Rather than replacing human instructors, these systems should complement them, combining scalability and personalization with empathy and mentorship. Future progress will depend on robust clinical outcome studies, technological refinement, ethical safeguards, and hybrid pedagogical models. If developed responsibly, AI and Metahuman-based education can democratize access to high-quality healthcare training worldwide.

Keywords

Metahuman, Artificial Intelligence, Digital Human, AI Professor, Healthcare Education, Immersive Learning

Introduction

The rapid advancement of artificial intelligence (AI) and immersive digital technologies has begun to reshape the landscape of higher education. Among these, AI and Metahuman-based education technologies-the convergence of AI-driven cognitive functions with hyper-realistic digital human embodiment—have emerged as a disruptive force. While traditional e-learning platforms rely heavily on text or video-based content delivery, Metahuman technology enables the creation of highly realistic virtual educators capable of speech, gesture, and emotional interaction, providing learners with an unprecedented sense of presence and engagement [1,2].

The necessity for innovation is particularly acute in medical and healthcare education. Medical schools and nursing programs face severe challenges such as faculty shortages, rapidly increasing student numbers, and the growing demand for flexible, scalable, and cost-effective training environments [3,4]. Traditional lecture-based methods, while foundational, often lack interactivity and learner engagement. Simulation-based education, though effective, is constrained by high costs, limited availability of mannequins, and logistical barriers [5]. These constraints have accelerated the search for novel approaches capable of delivering both knowledge transfer and experiential training.

Within this context, AI and Metahuman-based education present a unique opportunity. By integrating large language models (LLMs), speech synthesis, and emotional recognition with hyper-realistic 3D avatars developed in platforms such as Unreal Engine, they can serve as virtual professors, digital tutors, or simulated patients [6,7]. Unlike conventional avatars or chatbots, these technologies combine cognitive intelligence and embodied realism, enabling personalized, interactive, and scalable education in medicine and health sciences.

The research background for this review lies in two converging needs: (1) the educational demand for innovative tools to bridge skill gaps and enhance learner motivation, and (2) the technological advances that make AI and Metahuman-based applications feasible and cost-effective. Recent pilot implementations in higher education worldwide suggest a paradigm shift is underway, but their implications for healthcare education remain underexplored [8,9]. Therefore, the objective of this review is to (a) clarify the conceptual foundations of AI and Metahuman-based education, (b) examine its applications in healthcare learning environments, and (c) critically evaluate its opportunities and challenges to inform future research and policy.

 Background and Technological Framework

The development of AI and Metahuman-based education must be considered from two perspectives: (a) the conceptual framework that traces the evolution of human-like digital entities, and (b) the technological foundations that enable their integration into educational environments. This dual perspective clarifies both the definitions and the technical underpinnings required to understand their role in healthcare education.

Conceptual Framework

Virtual Human

Virtual Humans were the earliest generation of anthropomorphic agents, typically designed as 3D animated avatars or conversational characters with scripted responses. They were primarily developed for structured information delivery and entertainment, offering limited realism and no adaptive intelligence [6,7]. In education, they appeared as basic e-learning avatars or customer service bots, confined to pre-programmed interactions without dynamic feedback or contextual understanding.

Digital Human

Digital Humans represented a step forward in realism and interaction. By integrating motion capture, neural rendering, and basic natural language processing, they achieved more photorealistic facial and body movements [8]. These entities found applications as digital tutors, training assistants, and service agents, enhancing user engagement and trust compared to simple Virtual Humans. In education, Digital Humans provided an improved sense of social presence, though their intelligence remained restricted to semi-scripted interactions [9].

AI Human

AI Humans shifted the focus from appearance to cognitive intelligence. Leveraging large language models (LLMs), natural language understanding, and retrieval-augmented generation (RAG), AI Humans can act as autonomous lecturers, capable of generating course content, simulating teaching styles, and adapting explanations to learner input [10]. Case examples include AI-University (AI-U), which aligned AI-generated responses with course materials, and HKUST’s AI lecturer, deployed in postgraduate programs [11].

Metahuman

Metahumans, popularized by Epic Games’ Unreal Engine MetaHuman Creator, marked a breakthrough in hyper-realistic embodiment. They feature detailed skin textures, micro-expressions, and dynamic gestures, making them nearly indistinguishable from real humans in digital environments [12]. Compared to earlier Digital Humans, Metahumans drastically reduce production costs and time while significantly increasing realism. In healthcare education, Metahumans have been applied as virtual patients and AR-based training tutors, allowing students to practice complex procedures safely and repeatedly [13].

AI and Metahuman-based Education

When combined, AI provides the mind—knowledge reasoning, adaptive dialogue, and domain-specific expertise—while Metahumans provide the body—voice, gestures, and emotional presence. This synergy creates interactive, adaptive, and socially engaging digital educators or simulated patients [14]. Unlike earlier technologies, this model supports not only knowledge transfer but also emotional resonance, motivating learners and enhancing realism in medical and nursing education.

Technological Foundations

AI Technologies

The cognitive engine of AI and Metahuman-based education is built upon several recent AI advancements. LLMs generate domain-specific explanations and adjust dynamically to learner queries [10]. Neural text-to-speech systems produce natural prosody and intonation, allowing Metahumans to deliver lectures and clinical feedback with human-like speech [8]. Additionally, emotion recognition and affective computing enable adaptive instruction by detecting learner stress, confusion, or disengagement, thereby enhancing motivation and reducing cognitive load [9].

Metahuman Platforms

The Unreal Engine MetaHuman Creator has become the most widely used platform for building hyper-realistic avatars. It allows rapid design of characters with cinematic quality and real-time animation [12]. These avatars can be integrated into AR and VR environments, enabling immersive simulations such as hospital ward rounds, surgical training, or pandemic response exercises. For example, in nursing education, Metahuman tutors have been used to demonstrate procedures such as venipuncture or cardiopulmonary assessments, offering realistic practice without risk to patients [13].

Convergence Structures

The hallmark of AI and Metahuman-based education is the fusion of cognition and embodiment. In this convergence: AI provides adaptive reasoning, domain knowledge, and conversational interaction, while Metahumans deliver realism, emotional cues, and presence.

This integrated model has been applied in: 1) Clinical tutoring – AI-driven Metahumans simulating patient–student interactions, enabling diagnostic reasoning and communication training 2) AR/VR-based procedural training – step-by-step guidance within immersive simulations of complex tasks; 3) University lecturing – AI lecturers embodied as Metahumans, ensuring scalable and consistent delivery of course content [11].

The convergence framework demonstrates how cognitive intelligence and embodied realism combine to move beyond static simulation toward scalable, interactive, and emotionally engaging healthcare education.

Table 1. Conceptual Comparison of Virtual Human, Digital Human, AI Human, and Metahuman

Applications in Healthcare Education

The integration of AI and Metahuman-based education has begun to transform healthcare education at multiple levels. From medical and nursing curricula to public health team training and continuing professional development (CPD), Metahuman applications provide scalable, immersive, and adaptive experiences. Unlike conventional mannequins or video-based resources, Metahumans can replicate both the cognitive complexity and the emotional dynamics of real-world healthcare interactions.

Medical and Nursing Training: Virtual Patient Simulation, Skills, and Interaction Training

Medical and nursing education traditionally relies on lectures, mannequin-based simulation, and supervised clinical practice. However, mannequin simulation is expensive, resource-limited, and often lacks the dynamic interactivity required to prepare learners for complex patient encounters. AI and Metahuman-based applications address these limitations by offering virtual patients and interactive tutors.

For instance, Metahuman avatars have been designed as virtual patients with hyper-realistic facial expressions and responsive dialogue systems, enabling students to practice history-taking, diagnostic reasoning, and patient-centered communication in a safe environment [15]. Nursing students have used AR-based Metahuman tutors to practice procedures such as venipuncture, catheterization, or wound dressing, receiving real-time AI-driven feedback on accuracy and safety. Studies indicate that learners trained with Metahuman tutors report higher engagement, greater confidence, and improved skill retention compared to traditional mannequin-based sessions [16].

Health Science and Team-based Training: Crisis Response and Multi-student Training

Public health emergencies, such as the COVID-19 pandemic, have highlighted the urgent need for scalable crisis response training. Traditional drills require significant resources and cannot be easily replicated across diverse settings. In contrast, AI and Metahuman-based platforms allow team-based crisis simulations that are repeatable, cost-effective, and globally accessible.

In XR environments, Metahumans can represent infectious patients, emergency responders, or population groups, while AI lecturers guide learners through the simulation [17]. These platforms enable multi-student collaboration, where medical, nursing, and public health students interact within the same digital scenario. Research shows that such immersive simulations enhance team coordination, leadership, and decision-making skills under stress [18].

Moreover, because the digital environment can be scaled across institutions, the same training module can be delivered simultaneously to students in different regions, promoting equity in preparedness education.

 Lifelong Learning (CPD): Metahuman Application Cases 

Healthcare professionals require continuous education to maintain competency and adapt to evolving clinical practices. Traditional CPD courses are often text-heavy, asynchronous, and difficult to personalize. AI and Metahuman-based applications provide interactive and adaptive lifelong learning opportunities tailored to individual needs. For example, Metahuman professors have been piloted in postgraduate nursing and medical programs, providing personalized lectures, interactive case studies, and simulated patient encounters [19]. In some cyber universities, AI lecturers embodied as Metahumans have reduced average study time by nearly 30%, while maintaining learner satisfaction [20]. CPD modules using Metahuman avatars also allow multilingual delivery, ensuring accessibility for global learners.

These applications not only improve knowledge retention but also promote reflective practice by offering real-time transcripts of learner–Metahuman interactions, which can be used for feedback and self-assessment.

Table 2. Applications of AI and Metahuman-based Education in Healthcare

Case Studies of AI Professors in Higher Education

The adoption of AI and Metahuman-based education in higher education has evolved along three main pathways: (1) the introduction of AI Professors who act as autonomous lecturers within degree programs, (2) the deployment of AI-centered lecture platforms that integrate instructional content generation and course management, and (3) the implementation of AI Tutors that complement human teaching with personalized support for students. These cases, reported in diverse geographic and institutional contexts, provide valuable insights into the potential and limitations of AI-driven education.

AI Professors (Real Lecturers) 

The highest level of AI integration occurs when artificial humans take on the role of full professors in actual courses.

Hong Kong University of Science and Technology (HKUST): HKUST pioneered the first AI digital human lecturers in Asia. Developed using generative AI and embodied in hyper-realistic digital avatars, these AI lecturers were introduced into a graduate-level course. Each lecturer was designed with distinctive personas, capable of delivering lectures with voice, gestures, and expressions that closely mimicked human teaching [6]. Student surveys emphasized the novelty, naturalness, and accessibility of AI professors, but also revealed concerns regarding their limited emotional depth, responsiveness to unexpected queries, and ability to form mentorship-like relationships. Nevertheless, this case remains a landmark in demonstrating that AI and Metahuman-based systems can operate as real lecturers in higher education.

Nanyang Technological University (NTU), Singapore: NTU implemented the AI professor known as “Professor Leodar”, an AI chatbot lecturer trained with retrieval-augmented generation (RAG). This system was deployed in formal courses, offering contextually relevant lectures and real-time feedback [21]. Evaluation data revealed striking results: 97.1% of participating students expressed positive perceptions, citing continuous availability, personalized responses, and the ability to access knowledge outside normal class hours. Students viewed the AI professor as a valuable complement to traditional teaching, although they stressed the importance of balancing AI instruction with human mentorship.

Cyber Universities, South Korea: Cyber universities in South Korea introduced AI professors directly into their accredited degree programs. These AI lecturers, built on digital human avatars with AI-driven voice synthesis, delivered entire online courses, effectively functioning as professors for remote learners [9]. Findings showed that students experienced shortened study time, stable levels of satisfaction, and greater flexibility in learning. Unlike pilot projects, this case demonstrates large-scale, sustainable deployment of AI professors within a national higher education system, making it a critical reference for the institutionalization of AI-led teaching.

AI-Centered Lecture Platforms 

While not embodied as professors, several universities have adopted AI-centered platforms that perform core teaching functions, such as content creation, assessment design, and instructional alignment.

AI-University (AI-U), University of Southern California (USC): AI-U represents a cutting-edge LLM-based system designed to replicate the instructional role of professors. By fine-tuning language models with domain-specific materials and using RAG, AI-U generated lecture scripts, aligned Q&A, and adaptive explanations tailored to each course [7]. In pilot testing within scientific and engineering classrooms, AI-U assumed the professor’s role in knowledge delivery, demonstrating the potential for AI platforms to reduce faculty workload while ensuring academic rigor.

Kudu AI, University of California, Los Angeles (UCLA): UCLA integrated Kudu AI as an advanced instructional platform that produces AI-generated textbooks, assignments, and teaching assistant (TA) resources [22]. Kudu AI has been applied across courses to provide consistent teaching materials, real-time study aids, and even lecture outlines. Students reported improved accessibility of resources, although they acknowledged that human guidance remained essential for nuanced discussions and mentorship. This case illustrates how AI platforms can operate as “quasi-professors,” handling much of the instructional workload while leaving relational and evaluative functions to human faculty.

AI Tutors 

The third model of AI integration involves intelligent tutoring systems, which act as personalized study companions rather than full professors or lecture platforms.

Syntea AI Tutor, IU International University of Applied Sciences (Germany): IU International deployed the Syntea AI Tutor to support individualized learning. Unlike AI professors, Syntea is designed to complement existing teaching by offering adaptive exercises, contextual explanations, and real-time guidance [19]. Empirical data showed that Syntea reduced average student learning time by 27%, while maintaining high satisfaction levels. This suggests that AI tutors can substantially improve learning efficiency, particularly in self-paced or distance education settings. However, because they are not intended to fully replace professors, their role remains that of an augmentative tool rather than a standalone lecturer.

Table 3. Summary of AI Professors, Platforms, and Tutors in Higher Education

Opportunities of AI and Metahuman-based Education in Healthcare

The integration of AI and Metahuman-based education technologies offers a number of unique opportunities for advancing healthcare training. These opportunities can be categorized into four primary domains: (1) customized learning, (2) enhanced immersion and motivation, (3) cost efficiency and scalability, and (4) improved accessibility.

Customized Learning 

AI-driven Metahumans have the capacity to deliver personalized education by adapting instructional content, pacing, and feedback to each learner’s needs. In medical and nursing schools, AI lecturers can analyze student performance data to provide targeted remediation for struggling students or advanced exercises for high performers [23]. For example, adaptive feedback in procedural skills training allows students to repeat complex tasks until proficiency is achieved, while receiving step-by-step guidance from a Metahuman tutor [24]. This customized approach aligns with competency-based medical education, ensuring that learners acquire essential skills at their own pace rather than progressing uniformly through standardized curricula.

Enhancement of Immersion and Learning Motivation

Traditional e-learning systems often suffer from low engagement and high dropout rates. Metahumans, however, provide visual realism, emotional expression, and interactive presence, which can significantly enhance immersion. For instance, students interacting with a Metahuman patient report higher motivation and stronger feelings of “being there” compared to text or video-based learning modules [25]. By simulating authentic patient encounters—complete with facial expressions, tone of voice, and emotional responses—Metahumans help learners build empathy and communication skills, reinforcing both technical and affective aspects of clinical practice [26]. This immersive quality is especially valuable in nursing and psychology training, where human interaction and empathy are critical learning outcomes.

Cost Efficiency and Scalability

Healthcare simulation with mannequins and physical labs is expensive and difficult to scale. In contrast, once developed, AI and Metahuman-based simulations can be replicated indefinitely with minimal additional cost [27]. Universities can deploy the same virtual patient or AI professor across thousands of students simultaneously, eliminating scheduling conflicts and reducing faculty burden. Evidence from Cyber universities shows that integrating AI professors not only shortened learning time but also reduced overall program costs without compromising student satisfaction [9]. This indicates the potential of AI and Metahuman-based tools to make high-quality healthcare education more financially sustainable.

Improvement of Educational Accessibility

AI and Metahuman-based education also enhances accessibility. Courses can be offered in multiple languages simultaneously, with AI generating real-time translations and Metahumans delivering culturally adapted expressions [28]. This has profound implications for global health training, enabling medical students in low-resource countries to access the same advanced training as those in developed settings. Additionally, 24/7 availability of AI professors ensures that learners can study at any time, accommodating different schedules and learning paces. Such flexibility democratizes education by reaching non-traditional learners, working professionals, and geographically isolated populations.

Table 4. Educational Opportunities of AI and Metahuman-based Education in Healthcare

Challenges and Limitations

Despite the many opportunities, the adoption of AI and Metahuman-based education faces significant challenges and limitations. These barriers must be acknowledged and addressed to ensure sustainable and ethical implementation.

Technical Limitations and Stability Issues

While LLMs and Metahuman platforms have advanced rapidly, technical challenges remain. Issues include lip-sync accuracy, gesture naturalness, and system latency, which can break the sense of immersion [29]. In addition, AI responses may occasionally produce factual errors or “hallucinations,” posing risks in clinical education where accuracy is essential [30]. Ensuring system reliability, security, and maintenance requires ongoing technological refinement and substantial investment.

Ethical Considerations

Ethical concerns represent one of the most debated limitations:

AI bias: LLMs trained on biased datasets may propagate discriminatory assumptions, which is particularly dangerous in healthcare education [31].

Privacy protection: Patient data used in training must comply with strict data protection laws. Without careful oversight, sensitive health information could be compromised [32].

Patient similarity: Using Metahumans that resemble real patients without consent raises ethical dilemmas regarding identity rights and dignity [33].

These considerations require the establishment of clear ethical guidelines and compliance with international standards before large-scale adoption.  

Professor and Student Acceptance 

Adoption of AI professors depends heavily on acceptance from both faculty and students. Some educators express concern that AI systems could devalue their role or diminish human mentorship in medical training [39]. Students, while often enthusiastic, may still perceive AI lecturers as lacking empathy, critical judgment, or nuanced clinical reasoning [34]. Acceptance therefore requires careful change management, where AI is positioned as a complement to human educators rather than a replacement. Faculty training and transparency in AI functioning can also improve trust.

Lack of Learning Effect Verification

Perhaps the most critical limitation is the lack of rigorous, long-term studies proving that AI and Metahuman-based education improves clinical competence. While pilot projects report high satisfaction and engagement, there is limited evidence on measurable outcomes such as exam performance, clinical decision-making, or patient care quality [35]. Without robust randomized controlled trials (RCTs), policymakers and institutions may hesitate to invest in widespread deployment.

Table 5. Main Challenges of AI and Metahuman-based Education in Healthcare

Future Directions

While current cases and pilot projects highlight the promise of AI and Metahuman-based education, significant work remains to ensure sustainable, ethical, and effective integration into healthcare curricula. Future efforts can be organized into five key directions.

Research on Learning Effectiveness

Rigorous studies are required to establish the true impact of AI and Metahuman-based education. Future research should employ randomized controlled trials (RCTs), longitudinal designs, and multi-institutional comparisons to evaluate outcomes such as knowledge retention, procedural skill mastery, diagnostic reasoning, and patient safety [36,37]. Evidence of effectiveness will be critical for policymaker and institutional buy-in.

Technological Advancement and Integration

Future development should enhance stability, naturalism, and adaptability. Improvements in multimodal AI (voice, gesture, facial expression, emotion recognition) will allow more authentic interactions. Integration with XR and metaverse technologies could provide full-scale simulations of hospitals, operating theaters, or disaster response settings, extending beyond current AR/VR applications [38].

Ethical and Regulatory Frameworks

The deployment of AI professors requires internationally accepted ethical and legal frameworks. These should address: 1) Mitigation of bias in healthcare training datasets, 2) Privacy and data protection compliance, 3) Protocols for using patient-like avatars with informed consent, 4) Clear disclosure of AI’s role in teaching and assessment [39].

Future policy must ensure that AI systems remain transparent, equitable, and accountable, preventing misuse or inequitable access.  

Institutional and Pedagogical Adaptation

Universities must prepare for pedagogical shifts. Faculty development programs should support educators in blending AI systems into their teaching while maintaining mentorship and critical evaluation roles [40]. Institutions should promote hybrid models, where AI handles scalable delivery and assessment, while human educators provide empathy, professional judgment, and mentorship.

Global Accessibility and Equity

AI and Metahuman-based education holds potential for democratizing access to medical training worldwide. Future initiatives should focus on: Multilingual delivery to serve diverse populations, Open-access AI curricula for low-resource regions, Partnerships between universities, governments, and NGOs to ensure equitable distribution [33].

Without deliberate attention to accessibility, AI-based education risks widening rather than narrowing the global health education gap.

Table 6. Future Directions for AI and Metahuman-based Education

Discussion

The integration of AI and Metahuman-based education in healthcare is no longer speculative but a demonstrated reality, as evidenced by deployments across Hong Kong, Singapore, Korea, the United States, and Germany. These cases, together with the identified opportunities and challenges, provide a critical lens through which to evaluate the current state and future potential of this paradigm shift.

Comparative Analysis: Traditional vs. AI/Metahuman Approaches

Traditional healthcare education relies heavily on didactic lectures, mannequin-based simulation, and supervised clinical practice. While effective to an extent, these methods are limited in scalability, cost, and ability to provide individualized learning experiences. By contrast, AI professors, Metahuman tutors, and AI-centered platforms offer personalized instruction, immersive simulations, and scalable content delivery [24–27]. For example, mannequin-based clinical training requires physical labs and costly equipment, often resulting in restricted access. In contrast, AR-based Metahuman tutors allow repeatable, on-demand practice in risk-free digital environments. Similarly, while traditional professors cannot be available 24/7, AI lecturers provide continuous access, enabling learners to study at their own pace.

However, traditional approaches retain strengths that AI systems cannot yet replicate—most notably, the empathy, mentorship, and nuanced judgment of human educators. Thus, the evidence suggests not a binary replacement but the emergence of hybrid models, where human and AI educators complement one another.

Educational Impact and Implications 

The opportunities identified—customized learning, enhanced immersion, scalability, and accessibility—have profound implications for healthcare education. They suggest that AI and Metahuman-based education could bridge skill gaps, particularly in resource-limited settings, by offering affordable, high-quality training to a broader population. This democratization of education aligns with global health priorities, especially in addressing shortages of trained healthcare professionals [28–30]. At the same time, the limitations—technical instability, ethical dilemmas, mixed acceptance, and lack of validated outcomes—signal caution. Without rigorous evaluation, institutions risk adopting technologies that may increase engagement without demonstrable gains in clinical competence. Moreover, ethical concerns such as AI bias and privacy violations could undermine trust if not addressed proactively [36–38]. The implication is clear: the adoption of AI and Metahuman-based education must be deliberate, evidence-based, and ethically guided. Institutions must resist the temptation of novelty-driven adoption and instead ground implementation in robust pedagogy and empirical validation.

Policy and Institutional Considerations 

The reviewed cases highlight the need for policy-level frameworks to support sustainable integration. Universities must establish guidelines for the ethical use of AI professors, ensure transparency in AI decision-making, and provide faculty with training to integrate AI systems effectively [34,35]. At the same time, governments and international bodies should develop standards for data security, consent, and equity of access. For example, while cyber universities in Korea demonstrated cost and time efficiency, their success depended on clear institutional policies and student readiness to engage with AI professors [26]. Similarly, NTU’s positive student evaluations of Professor Leodar reflect not only technological sophistication but also institutional efforts to align the AI system with curricular goals [25]. Thus, the discussion suggests that technological innovation must be matched by pedagogical and policy innovation if AI professors are to achieve meaningful and sustainable impact in healthcare education.  

Synthesis: Toward a Hybrid Model 

Across all cases, the evidence converges on a central insight: AI and Metahuman-based education is most effective when used as a complement to, rather than a replacement for, human educators. Human professors provide empathy, mentorship, and professional modeling, while AI professors contribute scalability, personalization, and accessibility. This hybrid approach addresses both sides of the equation: expanding access and efficiency while safeguarding the relational and ethical dimensions of medical education. The future of healthcare training will likely depend on the strategic blending of human and AI capacities, maximizing their respective strengths while minimizing their limitations.

Table 7. Comparative Summary of Traditional vs. AI/Metahuman-Based Healthcare Education.

Conclusion

 The review demonstrates that AI and Metahuman-based education is moving rapidly from experimental pilots to practical applications in healthcare. Case studies from Hong Kong, Singapore, Korea, the United States, and Germany show that AI professors, AI-centered platforms, and AI tutors can enhance engagement, scalability, and accessibility, while also reducing costs. Opportunities include personalized learning, immersive simulations, and global accessibility, which could democratize high-quality healthcare training across diverse contexts. Nevertheless, challenges persist. Technical limitations, ethical concerns, and questions of acceptance and educational validity remain unresolved. The evidence suggests that AI and Metahuman-based education is best positioned as a complement rather than a substitute for traditional teaching. Human educators provide empathy, mentorship, and professional modeling, while AI systems contribute efficiency and personalization.

The central implication is that the future of healthcare education will depend on hybrid models, balancing human and AI contributions. With continued research, technological refinement, ethical safeguards, and policy support, AI and Metahuman-based education can play a pivotal role in transforming how the next generation of healthcare professionals are trained.

Conflict of Interest

The author has no conflicts of interest to declare and agreed to the published version of the manuscript.

Author Contributions

BSH conceptualized, designed, and wrote the manuscript. MRK performed the literature search and critical analysis.

References

1. Zhang J, Fenton SH. Preparing healthcare education for an AI-augmented future. NPJ Health Syst. 2024;1(1):1-4
2. Duan S, Liu C, Rong T, Zhao Y, Liu B. Integrating AI in medical education: a comprehensive study of medical students’ attitudes, concerns, and behavioral intentions. BMC Med Educ. 2025;25:1-7
3. Breckner K, Neumayr T, Mara M, Streit M, Augstein M. The changing nature of human-AI relations: a scoping review on terminology and evolvement in the scientific literature. Int J Hum-Comput Interact. 2025;6:https://doi.org/10.1080/ 10447318.2025.2482742
4. Giuliana GT. What is so special about contemporary CG faces? Semiotics of Metahumans. Topoi. 2022;41:821-834
5. Collins E, Venaruzzo L, Komoder A. Metahuman: bridging the skills gap through immersive technology. ASCILITE Publications. 2024; https://doi.org/10.14742/apubs.2024.1117
6. Pang CC, Zhao Y, Yin Z, Sun J. Artificial human lecturers: initial findings from Asia’s first AI lecturers in class to promote innovation in education. arXiv. 2024; https://doi.org/10.48550/arXiv.2410.03525
7. Shojaei MF, Gulati R, Jasperson BA, Wang S, Cimolato S, Cao D, Neiswanger W, Garikipati K. AI-University: an LLM-based platform for instructional alignment to scientific classrooms. arXiv. 2025: https://doi.org/10.48550/arXiv.2504.08846
8. Pour ME, Gieber C, Schmitt J, Bruck R. Evaluation of Metahumans as digital tutors in AR medical training. Cur Dir Biomed Eng. 2023;9(1):41-44  
9. Lee J, Jang MR, Suh HJ. Revolutionizing Higher Education: A Case Study of AI professors in a Cyber University. J Intell Inform Syst. 2024;30(4):351-367
10. Bae S, Jung T, Cho J, Kwon O. Effects of Metahuman characteristics on user acceptance: from the perspective of uncanny valley theory. Behav Inf Technol. 2025;44(4):731-748
11. You M. Development of a digital human production education program using Metahuman. J Korea Soc Media Arts. 2022;20(6):37-50
12. Sun L, Yin C, XuQ, Zhao W. Artificial intelligence for healthcare and medical education: a systematic overview. Am J Trans Res. 2023;15(7):4820-4828
13. Gordon M, Baniel M, Ajiboye A, Uraiby H, Xu NY, Bartlett R, Hanson J, Haas M, Spadafore M, Grafton-Clark C, Gasiea RY, Michie C, Corral J, Kwon B, Dolmans D, Thammasitboon S. A scoping review of artificial intelligence in medical education: BEME Guide No. 84. Med Tech. 2024;46(4):446-470
14. Feigerlova E, Hani H, Hothersall-Davies E. A systematic review of the impact of artificial intelligence on education on educational outcomes in health professions education. BMC Med Educ. 2025;24(129):1-18
15. Shevtsova D, Ahmed A, Boot IWA, Sanges C, Hudecek M, Jacobs JJL, Hort S, Vrijhoef HJM. Trust in and acceptance of Artificial Intelligence applications in medicine: mixed methods study. JMIR Hum Factors. 2024;11:e47031
16. Yang D, Sun M, Zhou J, Lu Y, Song Z, Chen Z, Yang D, Wu Z, Ge H, Zhang Y, Gao C, Xuan J, Li X, Yin J, Zhu Z, Liu J, xin H, Jiang W, Wang N, Wang Y, xie L, Zheng Y, Powell CA, Thuemmler C, Chavannes NH, Wu L, Zhang H, He Y, Song Y, Bai C. Expert consensus on the “digital human” of metaverse in medicine. Clinical eHealth. 2023;6(7944):159-163
17. Knopp M, Warm EJ, Weber D, Kelleher M, Kinnear B, Schumacher DJ, Santen SA, Mendonca Ek, Turner L. AI-Enabled Medical Education: Threads of Change, Promising Futures, and Risky Realities Across Four Potential Future Worlds. JMIR Med Educ. 2023;6:159-163
18. Sami A, Tanveer F, Sajwani K, Kiran N, Javed MA, Ozsahin DU, Muhammad K, Waheed Y. Medical students’ attitudes toward AI in education:perception, effectiveness, and its credibility. BMC Med Educ. 2025;25(82):1-10
19. Moller M, Nirmal G, Fabietti D, Stierstor Q, Zakhavatkin M, Sommerfeld H, Schutt S. Revolutionizing Distance Learning: A Comparative Study of Learning Progress with AI-Driven Tutoring. arXiv. 2024; https://doi.org/10.48550/arXiv.2403.14642
20. Lambert SI, Madi M, Sopka S, Lenes A, Stange H, Buszello CP, Stephan A. An integrative review on the acceptance of artificial intelligence among healthcare professionals in hospitals. NPJ Digit Med. 2023;6(111)1-14
21. Thway M, Recatala-Gomez J, Lim FS, Hippalgaonkar K, Ng LWT. Battling Botpoop using GenAI for Higher Education: A Study of a Retrieval Augmented Generation Chatbots Impact on Learning. arXiv. 2024:https://doi.org/10.48550/arXiv.2406.07796
22. University of California, Los Angeles. Kudu AI: AI-generated textbooks, assignments, and TA resources for higher education. Los Angeles (CA): UCLA; 2025. Implementation report. https://kudu.com/index.html
23. Cook DA, Brydges R, Hamstra SJ, Zendejas B, Szostek JH, Wang AT, Erwin PJ, Hatala R. Comparative effectiveness of technology-enhanced simulation in health professions education: a systematic review and meta-analysis. Simul Healthc. 2012;7(5):309-320
24. Issenberg SB, McGaghie WC, Petrusa ER, Gordon DL, Scalese RJ. Features and uses of high-fidelity medical simulations that lead to effective learning: a BEME systematic review. Med Teach. 2005;27(1):10-28
25. Barsom EZ, Graafland M, Schijven MP. Systematic review on the effectiveness of augmented reality applications in medical training. Surg Endosc. 2016;30(4):4174-4183
26. Ji Z, Lee N, Frieske R, Yu T, Su D, Xu Y, Ishii E, Bang YJ, Madotto A, Fung P. . Survey of hallucination in natural language generation. ACM Comput Surv. 2023;55(12):1-38
27. Prajwal KR, Mukhopadhyay R, Namboodiri V, Jawahar CV. A Lip Sync Expert Is All You Need for Speech to Lip Generation In The Wild. arXiv. 2020; https://doi.org/10.48550/arXiv. 2008.10010
28. UNESCO. Recommendation on Open Educational Resources (OER). Paris: United Nations Educational, Scientific and Cultural Organization; 2019.
29. Obermeyer Z, Powers B, Vogeli C, Mullainathan S. Dissecting racial bias in an algorithm used to manage the health of populations. Science. 2019;366(6464):447-453
30. World Health Organization. Global strategy on digital health 2020–2025. Geneva: WHO; 2020.
31. European Union. Regulation (EU) 2016/679 (General Data Protection Regulation). Off J Eur Union. 2016;L119:1-88.
32. U.S. Department of Health and Human Services. HIPAA Privacy Rule (45 CFR Parts 160 and 164). Washington (DC): HHS; 2000 (as amended).
33. Mittelstadt BD, Allo P, Taddeo M, Wachter S, Floridi L. The ethics of artificial intelligence: mapping the debate. Big Data Soc. 2016;1-21
34. Steinert Y, Mann K, Centeno A, Dolmans D, Spencer J, Gelula M, Prideaux D. A systematic review of faculty development initiatives designed to improve teaching effectiveness in medical education. Med Teach. 2006;28(6):497-526
35. Chan KS, Zary N. Applications and challenges of implementing artificial intelligence in medical education: integrative review. JMIR Med Educ. 2019;5(1)
36. Liu X, Rivera SC, Moher D, Calvert MJ, Denniston AK, The SPIRIT-AI and CONSORT-AI working group. Reporting guidelines for clinical trial reports for interventions involving artificial intelligence: the CONSORT-AI extension. Nat Med. 2025;26:1364-1374
37. Rivera SC, Liu X, Chan AW, Denniston AK, Calvert MJ, The SPIRIT-AI and CONSORT-AI working group, SPIRIT-AI and CONSORT steering group, SPIRIT-AI and CONSORT-AI consensus group. Guidelines for clinical trial protocols for interventions involving artificial intelligence: the SPIRIT-AI extension. Nat Med. 2020;16:1351-1363
38. Zhu E, Masiello I, Hadadgar A, Zary N. Augmented reality in healthcare education: an integrative review. Peer J. 2014;2:e469
39. World Health Organization. Ethics and governance of artificial intelligence for health: WHO guidance. Geneva: WHO; 2021.
40. European Commission High-Level Expert Group on AI. Ethics guidelines for trustworthy AI. Brussels: European Commission; 2019.