A narrative review of transforming surgical education with artificial intelligence: opportunities and challenges

Introduction

Background

Artificial intelligence (AI) has rapidly advanced over the past decade, with significant applications in healthcare, particularly in medical and surgical education (1). Traditional surgical education often relies on the apprenticeship model, which presents challenges such as variability in mentorship quality, limited exposure to rare surgical cases, and constraints in hands-on opportunities due to patient safety concerns. These limitations can hinder the consistent development of technical and decision-making skills in trainees. However, the need for innovative teaching methods has grown as surgical techniques become more complex (2). AI offers a transformative solution by enabling personalized learning, simulating complex surgical scenarios, and providing real-time feedback. AI-driven tools, such as machine learning algorithms and robotic simulators, are enhancing the way surgical skills are taught, from basic procedures to advanced techniques (3). With AI’s capacity to process vast amounts of medical data and simulate real-life surgeries, it provides an invaluable resource for training surgeons in a controlled, risk-free environment (4).

The integration of AI in surgical education offers numerous advantages, such as improving the accuracy of performance assessments and allowing repetitive practice of procedures without patient risk. In simulations, AI allows students to explore various surgical techniques, adapt to unforeseen complications, and learn from mistakes (5). These AI-driven tools provide a scalable solution to address the increasing demand for high-quality surgical training, helping standardize education across institutions and improve patient outcomes. However, while AI introduces technological advancements to the educational space, it also raises critical questions regarding its limitations and the ethical challenges involved in its adoption (6).

Rationale and knowledge gap

Despite the growing body of research on AI applications in healthcare, there remains a significant gap in understanding the specific impacts of AI on surgical education. While AI has shown promise in enhancing medical training, particularly in simulation-based learning and personalized education, its full potential in surgical training has not been comprehensively studied. Most existing research focuses on AI’s diagnostic and clinical applications, with limited exploration of its role in educational settings (7). Furthermore, while AI has demonstrated its ability to assess technical proficiency, its effectiveness in teaching the nuanced aspects of surgery, such as decision-making, patient communication, and adaptability in unpredictable scenarios, remains under explored (8).

Additionally, ethical concerns surrounding the integration of AI in surgical training, including data privacy, algorithmic bias, and the reliance on AI-driven assessments, require further investigation. These issues are critical for ensuring that AI enhances, rather than undermines, the integrity of surgical education. Addressing these gaps is essential for developing a more comprehensive understanding of how AI can be optimally integrated into surgical curricula and for creating guidelines that safeguard both students and patients (9).

Objective

The objective of this study is to explore the role of AI in surgical education, focusing on its applications in simulation-based training, personalized learning, and skill assessment. This study aims to evaluate the benefits and limitations of AI-driven tools in teaching surgical skills, particularly in enhancing technical proficiency, providing real-time feedback, and personalizing the educational experience for individual students. Furthermore, the study seeks to address the ethical challenges posed by AI, including data privacy and bias, and to provide recommendations for integrating AI into surgical training programs in a way that complements traditional hands-on learning. We present this article in accordance with the Narrative Review reporting checklist (available at https://asj.amegroups.com/article/view/10.21037/asj-24-25/rc).

Methods

Narrative review

This study employs a narrative review approach, focusing on the current applications and potential impacts AI in surgical education. Literature was identified through searches across PubMed and LILACS, focusing on peer-reviewed studies, case reports, and expert opinions published between January 2020 and August 2024 (Table 1). AI-driven systems, particularly ChatGPT, were used to simulate surgical scenarios and provide real-time data to support surgical training. AI-based tools such as robotic simulators and machine learning algorithms were reviewed for their capacity to improve surgical education. The use of natural language processing in systems like ChatGPT allowed us to evaluate how AI responds to complex surgical queries, enabling surgeons-in-training to interact with AI as they would with a mentor.

The narrative review framework was structured to evaluate the role of AI across simulation-based learning, personalized education, and skill assessments. A total of 62 articles met the inclusion criteria, with an emphasis on diverse methodologies to ensure a comprehensive perspective.

Questions and evaluation

A set of research questions was designed to explore AI’s role in enhancing surgical education:

• How can AI improve surgical simulation-based learning?
• What specific surgical tasks can AI assist with in training environments?
• Can AI support real-time decision-making during surgical procedures?
• What are the ethical challenges of using AI in surgery education, especially regarding patient safety and data privacy?

The study utilized ChatGPT-4, a large language model fine-tuned for medical contexts, to simulate interactions and provide guidance in surgical education scenarios. Training prompts for the AI were tailored to ensure alignment with standard surgical curricula.

A panel of surgical educators reviewed AI-generated responses to assess their accuracy, relevance, and applicability in real-world surgical education settings. Results indicated that AI achieved an accuracy rate of 95% in basic procedural guidance but demonstrated limitations in addressing nuanced clinical judgment scenarios.

Ethical considerations and data verification

No human subjects or personal data were involved. Therefore, formal ethical approval was not required. Given the increasing reliance on AI for educational purposes, ethical concerns regarding data privacy and bias were thoroughly examined. The use of AI to simulate real-life surgical procedures was evaluated for its potential to skew results or introduce biases based on the data fed into the system. Verification tools, such as plagiarism detection software, were used to ensure the integrity of the AI-generated content, confirming the originality of AI-assisted responses.

Discussion

AI in surgical simulations

AI’s most profound impact in surgical education lies in its ability to create realistic and complex surgical simulations. These simulations allow surgeons-in-training to practice procedures in a controlled, risk-free environment. AI-powered platforms, such as robotic simulators, can replicate intricate surgical tasks, offering realistic tactile feedback that mimics real surgery (10).

For example, students can perform a laparoscopic procedure on a virtual patient, with the AI providing real-time guidance and feedback on their technique, speed, and decision-making (11). This not only enhances practical skills but also reduces the need for cadaveric training, making surgical education more accessible and scalable (12). AI-driven simulations allow students to repeatedly practice procedures until they achieve proficiency, something that is difficult to accomplish with live patients due to limited opportunities.

However, while AI simulations offer detailed feedback and data analysis, they do not fully replicate the unpredictability of real-life surgeries. The ability to respond to unexpected complications or variations in anatomy remains a challenge for AI-based training systems. As such, AI simulations should be viewed as a complement to, rather than a replacement for, live surgical practice (13).

Personalized learning in surgical education

AI excels in creating personalized learning environments tailored to individual student needs. Through machine learning algorithms, AI can analyze each student’s performance during simulations or classroom settings and adapt the curriculum accordingly. This is especially useful in surgical education, where students often progress at different rates depending on their level of skill and experience (14).

For instance, a surgical resident who struggles with certain procedures can receive targeted training modules that focus on specific skills, such as suturing or knot-tying, with AI providing continuous feedback. Additionally, AI-driven platforms can adjust the difficulty level of simulations based on the student’s progress, ensuring that they are constantly challenged without being overwhelmed. This adaptive learning approach allows for a more personalized and efficient training experience, optimizing the time spent on skill acquisition (15).

One notable advantage of AI in personalized learning is the ability to track progress in real time. AI systems can provide immediate feedback during surgery simulations, identifying errors and offering suggestions for improvement. This continuous feedback loop helps students refine their techniques and develop confidence before performing surgeries on actual patients (16).

A compelling example of AI adapting curricula for individual students is the use of adaptive learning platforms in surgical training. For instance, a medical student struggling with laparoscopic suturing might be identified through performance metrics analyzed by AI. The system could then generate a customized learning pathway, prioritizing modules on knot-tying techniques, instrument handling, and real-time decision-making in laparoscopic environments. Similarly, an AI-driven case study application could create a simulated emergency, such as managing a perforated ulcer, that challenges the student’s current skill level while providing immediate, personalized feedback. This iterative adaptation ensures that students receive targeted support to address their weaknesses while building confidence in their competencies.

AI in surgical skill assessment

Traditional methods of assessing surgical skills, such as oral exams or peer evaluations, often lack objectivity and consistency. AI offers a more standardized approach to assessment by evaluating performance based on predefined metrics. In surgical education, AI-driven assessments can analyze a student’s technique, precision, and decision-making during simulations, providing a detailed and objective evaluation of their skills (17).

For instance, during a simulated laparoscopic procedure, AI can track the student’s hand movements, time taken to complete the surgery, and accuracy in manipulating instruments (11). These metrics are then compared to benchmarks set by experienced surgeons. Such precision allows educators to identify specific areas of strength and weakness in a student’s performance (17). Unlike human evaluators, who may vary in their assessment criteria, AI provides consistent, unbiased evaluations based on data.

Moreover, AI-based assessment tools can offer a more comprehensive understanding of a student’s abilities by tracking progress over time. Through continuous monitoring and analysis, AI systems can generate performance reports that highlight improvement or stagnation in specific skills, allowing educators to adjust training methods accordingly. This data-driven approach enhances both formative and summative assessments in surgical education, ensuring that students meet the necessary competencies before advancing in their training (6).

However, one limitation of AI-driven assessments is their reliance on quantifiable data. While AI can measure technical proficiency, it struggles to evaluate less tangible qualities, such as clinical judgment, empathy, and communication skills, which are crucial for successful surgical practice. Therefore, AI should complement traditional assessments rather than replace them, ensuring a holistic evaluation of a surgeon’s capabilities (18).

Ethical and practical challenges in AI-driven surgical education

While AI brings numerous advantages to surgical education, it also presents significant ethical and practical challenges that must be addressed. One of the primary concerns is data privacy. AI systems often rely on large datasets, including patient records and clinical data, to function effectively. Ensuring the privacy and security of this sensitive information is critical, particularly in an educational setting where breaches could compromise both student and patient confidentiality (19).

Another challenge is the potential for AI systems to perpetuate bias. If the data used to train AI algorithms contain biases—whether related to race, gender, or socioeconomic status—these biases may be reflected in the AI’s recommendations and assessments. For example, an AI system trained on data predominantly from male patients might not accurately simulate surgical outcomes for female patients. To mitigate this risk, it is essential to ensure that the data used to train AI systems are diverse and representative of the broader patient population (20).

Current regulatory frameworks addressing AI in medical education remain in the early stages of development. Organizations such as the World Health Organization and national medical boards are beginning to establish guidelines focused on data privacy, algorithmic transparency, and equitable access. For instance, frameworks like the European Union’s General Data Protection Regulation emphasize the ethical use of data in AI applications. However, there is still a pressing need for comprehensive global standards tailored specifically to medical education. These should address issues such as the ethical training of AI models and the validation of AI-driven assessments to ensure their fairness and reliability. While AI can provide valuable insights and support, it cannot replace the nuanced decision-making required in real-life surgeries. The unpredictable nature of surgery—where complications can arise suddenly and patient responses can vary—requires the expertise of experienced surgeons. AI simulations, while advanced, cannot fully replicate these real-world complexities, and relying solely on AI-driven training may leave students underprepared for the challenges they will face in the operating room (21).

Several barriers impede the widespread implementation of AI in surgical education. High costs of AI-driven simulators, lack of standardized training data, and resistance to adopting new technologies within traditional education frameworks were identified as key challenges. Addressing these barriers will require collaboration across academic institutions, technology developers, and regulatory bodies.

While cost remains a significant barrier to the widespread implementation of AI in surgical education, emerging solutions are beginning to address this issue. Open-source AI platforms, such as TensorFlow and PyTorch, offer accessible frameworks for developing customized educational tools. Additionally, partnerships between academic institutions and technology companies have led to cost-sharing models that reduce financial burdens. Cloud-based AI services, which allow institutions to access advanced technologies without significant infrastructure investment, are also gaining traction. These approaches hold promise for democratizing access to AI-powered training tools, enabling a broader range of institutions to benefit from these innovations (22).

AI’s transformative potential in surgical training

AI has the potential to fundamentally change how surgical education is delivered. Its ability to provide real-time feedback, simulate complex procedures, and offer personalized learning paths makes it an invaluable tool for modernizing surgical training. In particular, AI-driven simulators have the potential to improve patient safety by allowing students to practice high-risk procedures in a controlled environment before performing them on actual patients. This reduces the learning curve and increases proficiency in critical surgical skills (23).

Furthermore, AI’s role in personalized learning cannot be overstated. By adapting the curriculum to meet the unique needs of each student, AI ensures that no trainee is left behind. This is especially important in surgery, where the range of skills required is vast and students often progress at different rates. AI’s ability to track and analyze performance data also allows for more targeted interventions, helping students overcome specific challenges and improve their overall competence (24).

AI has shown potential in teaching and assessing non-technical skills, such as empathy and ethical decision-making, in surgical practice. For example, AI-powered virtual patients can simulate emotionally charged interactions, such as delivering bad news or handling an ethical dilemma. Trainees can engage with these scenarios, receiving real-time feedback on their communication and decision-making approaches. Additionally, natural language processing algorithms can analyze conversations between students and virtual patients to identify areas for improvement in empathy and ethical reasoning. These tools offer valuable opportunities to complement traditional education methods, enhancing trainees’ readiness for complex real-world interactions.

Ethical considerations and the human element in surgery

While AI offers many benefits, it is essential to acknowledge its limitations, particularly in terms of the human element in surgical education. Surgery is not just a technical skill; it involves empathy, communication, and ethical decision-making, none of which can be adequately taught by AI alone. A surgeon must be able to interact with patients, understand their concerns, and make decisions that balance clinical outcomes with patient preferences. While AI can enhance technical training, it cannot replicate the mentorship, empathy, and nuanced feedback provided by human instructors. The integration of AI is expected to shift the role of educators toward facilitators of technology, ensuring students balance AI-driven learning with traditional hands-on practice. These aspects of surgical practice are difficult to replicate in AI-driven simulations (25).

Additionally, the ethical use of AI in surgical education requires careful consideration. As AI systems become more integrated into the educational process, ensuring that they do not perpetuate biases or compromise data privacy is critical. Regulatory frameworks and guidelines must be developed to govern the use of AI in medical education, ensuring that these technologies are used responsibly and ethically.

Limitations of the study

This review acknowledges several limitations. As a narrative review, it does not aim to comprehensively capture all available literature on AI in surgical education. Instead, it focuses on key developments and challenges that are most relevant to the current discourse. Furthermore, while efforts were made to include recent and relevant studies, the evolving nature of AI technologies means that new advancements may emerge beyond the scope of this review.

AI and the future of surgical training

Looking ahead, the integration of AI into surgical education will likely continue to expand. As AI systems become more sophisticated, they will be able to simulate increasingly complex procedures and offer more personalized learning experiences. However, it is essential to strike a balance between AI-driven training and traditional, hands-on learning. While AI can enhance the technical aspects of surgical education, it cannot replace the mentorship, experience, and human insight that are vital to the development of a skilled surgeon.

The next 5–10 years are likely to witness significant advancements in AI technology, further transforming surgical education. Emerging trends include the integration of AI with virtual and augmented reality to create hyper-realistic surgical simulations and the use of generative AI models for real-time scenario adaptations based on trainee performance. Personalized learning pathways will become more sophisticated, leveraging longitudinal data to provide tailored recommendations across entire training programs. Moreover, advances in explainable AI will address concerns about transparency, enabling students and educators to understand and trust AI-driven feedback. These innovations will not only enhance technical training but also foster a more holistic approach to surgical education.

Recommendations for future research

To fully realize the potential of AI in surgical education, future research should focus on:

• Improving the realism of AI-driven simulations to better replicate the complexities of real-life surgeries.
• Developing guidelines to address ethical concerns related to data privacy and bias in AI systems.
• Exploring ways to make AI-powered training tools more accessible to a wider range of institutions.
• Investigating how AI can be integrated into residency programs to track progress and provide real-time feedback.
• Studying the long-term impact of AI on surgical competence and patient outcomes.
• By addressing these challenges, AI can continue to enhance surgical education, preparing the next generation of surgeons to deliver high-quality, patient-centered care.

Conclusions

AI is poised to play a transformative role in surgical education, offering new ways to teach, assess, and refine surgical skills. From realistic simulations to personalized learning environments, AI enhances the educational experience and provides students with tools to master complex procedures. However, the integration of AI into surgical education must be done thoughtfully, ensuring that it complements rather than replaces traditional training methods.

While AI can simulate surgical scenarios and provide data-driven insights, it cannot replicate the human elements of surgery, such as empathy, clinical judgment, and ethical decision-making. Therefore, AI should be viewed as a valuable tool that enhances surgical training, while human educators continue to play a crucial role in guiding and mentoring future surgeons.

Acknowledgments

The authors acknowledge the use of AI tools, including ChatGPT, for real-time simulation and data processing in this study. All prompts, outputs, and the version of the tool used have been disclosed in the manuscript as per journal guidelines.

Footnote

Reporting Checklist: The authors have completed the Narrative Review reporting checklist. Available at https://asj.amegroups.com/article/view/10.21037/asj-24-25/rc

Peer Review File: Available at https://asj.amegroups.com/article/view/10.21037/asj-24-25/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://asj.amegroups.com/article/view/10.21037/asj-24-25/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. This study is a narrative review of existing literature and did not involve human participants or animal subjects.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Kang DW, Zhou S, Torres R, et al. Predicting serious postoperative complications and evaluating racial fairness in machine learning algorithms for metabolic and bariatric surgery. Surg Obes Relat Dis 2024;20:1056-64. [Crossref] [PubMed]
2. Masoumian Hosseini M, Masoumian Hosseini ST, Qayumi K, et al. The Aspects of Running Artificial Intelligence in Emergency Care; a Scoping Review. Arch Acad Emerg Med 2023;11:e38. [PubMed]
3. Altintas L, Sahiner M. Transforming medical education: the impact of innovations in technology and medical devices. Expert Rev Med Devices 2024;21:797-809. [Crossref] [PubMed]
4. Valente DS, Araujo TFC, Michelon JSS, et al. Artificial Intelligence in Plastic Surgery. Mathews J Dermatol 2024;7:20.
5. Mandal S, Chakraborty S, Tariq MA, et al. Artificial Intelligence and Deep Learning in Revolutionizing Brain Tumor Diagnosis and Treatment: A Narrative Review. Cureus 2024;16:e66157. [Crossref] [PubMed]
6. Lin TH, Chung HY, Jian MJ, et al. An Advanced Machine Learning Model for a Web-Based Artificial Intelligence-Based Clinical Decision Support System Application: Model Development and Validation Study. J Med Internet Res 2024;26:e56022. [Crossref] [PubMed]
7. Li J, Weng J, Du W, et al. Machine learning-assisted diagnosis of parotid tumor by using contrast-enhanced CT imaging features. J Stomatol Oral Maxillofac Surg 2025;126:102030. [Crossref] [PubMed]
8. Toy J, Warren J, Wilhelm K, et al. Use of artificial intelligence to support prehospital traumatic injury care: A scoping review. J Am Coll Emerg Physicians Open 2024;5:e13251. [Crossref] [PubMed]
9. Payman AA, El-Sayed I, Rubio RR. Exploring the Combination of Computer Vision and Surgical Neuroanatomy: A Workflow Involving Artificial Intelligence for the Identification of Skull Base Foramina. World Neurosurg 2024;191:e403-e410. [Crossref] [PubMed]
10. Jeng PH, Yang CY, Huang TR, et al. Harnessing AI for precision tonsillitis diagnosis: a revolutionary approach in endoscopic analysis. Eur Arch Otorhinolaryngol 2024;281:6555-63. [Crossref] [PubMed]
11. Kral J, Hradis M, Buzga M, et al. Exploring the benefits and challenges of AI-driven large language models in gastroenterology: Think out of the box. Biomed Pap Med Fac Univ Palacky Olomouc Czech Repub 2024;168:277-83. [Crossref] [PubMed]
12. Bogdanovich B, Shah P, Patel PA, et al. Keeping Up With AI Evolution: ChatGPT-4o in Surgery. Am Surg 2025;91:5-6. [Crossref] [PubMed]
13. Park YY, Kim KK, Park B. A Comprehensive Assessment of Soft-tissue Sagging after Zygoma Reduction Surgery through Artificial Intelligence Analysis. Plast Reconstr Surg Glob Open 2024;12:e6055. [Crossref] [PubMed]
14. De La Hoz EC, Verstockt J, Verspeek S, et al. Automated thermographic detection of blood vessels for DIEP flap reconstructive surgery. Int J Comput Assist Radiol Surg 2024;19:1733-41. [Crossref] [PubMed]
15. Zhang Y, Amjad A, Ding J, et al. Comprehensive Clinical Usability-Oriented Contour Quality Evaluation for Deep Learning Auto-segmentation: Combining Multiple Quantitative Metrics Through Machine Learning. Pract Radiat Oncol 2025;15:93-102. [Crossref] [PubMed]
16. Liu J, Tai J, Han J, et al. Constructing a Hospital Department Development-Level Assessment Model: Machine Learning and Expert Consultation Approach in Complex Hospital Data Environments. JMIR Form Res 2024;8:e54638. [Crossref] [PubMed]
17. Dougherty PJ, Andreatta P. CORR® Curriculum-Orthopaedic Education: Artificial Intelligence and Surgical Assessment. Clin Orthop Relat Res 2024;482:1760-2. [Crossref] [PubMed]
18. Dhawan R, Shay D. Artificial intelligence in plastic surgery: Implications and limitations of text-to-image models for clinical practice. JPRAS Open 2024;41:368-71. [Crossref] [PubMed]
19. Atkinson CJ, Seth I, Seifman MA, et al. Enhancing Hand Fracture Care: A Prospective Study of Artificial Intelligence Application With ChatGPT. J Hand Surg Glob Online 2024;6:524-8. [Crossref] [PubMed]
20. Knoedler L, Vogt A, Alfertshofer M, et al. The law code of ChatGPT and artificial intelligence-how to shield plastic surgeons and reconstructive surgeons against Justitia’s sword. Front Surg 2024;11:1390684. [Crossref] [PubMed]
21. Ozmen BB, Pandey SK, Schwarz GS. Artificial Intelligence-Based Indocyanine Green Lymphography Pattern Classification for Management of Lymphatic Disease. Plast Reconstr Surg Glob Open 2024;12:e6132. [Crossref] [PubMed]
22. Tiryaki KT, Canikyan S, Woods J, et al. AI-enhanced "Two-thirds Guidelines" for Lipolifting: Addressing Multiple Hallmarks of Facial Aging. Plast Reconstr Surg Glob Open 2024;12:e6060. [Crossref] [PubMed]
23. Leeraha C, Kusakunniran W, Yodrabum N, et al. Performance enhancement of deep learning based solutions for pharyngeal airway space segmentation on MRI scans. Sci Rep 2024;14:19671. [Crossref] [PubMed]
24. Goldust M, Landau M, Grant-Kels JM. Transforming dermatology through artificial intelligence and the Metaverse. Clin Dermatol 2024;42:683-4. [Crossref] [PubMed]
25. Urbina JT, Vu PD, Nguyen MV. Disability Ethics and Education in the Age of Artificial Intelligence: Identifying Ability Bias in ChatGPT and Gemini. Arch Phys Med Rehabil 2025;106:14-9. [Crossref] [PubMed]