Lowering the Barrier to Surgical Research: Can Artificial Intelligence (AI) Support Trainees? Authors: Michael Rouse, Yumna Kaleem, Justin M Yeung Published: 2026 Introduction Research is increasingly viewed as a core component of surgical training, yet for many trainees the pathway into academic work remains unclear and difficult to access. Within the Australasian training environment, where research productivity is increasingly expected for progression and competitiveness, reducing barriers to entry is particularly relevant. Despite widespread institutional recognition of the critical role of research in developing a surgical career, trainees continue to struggle to acquire the necessary skill sets and engage in significant research. These challenges include financial costs, time commitments, lack of protected time, limited mentorship opportunities, changes in clinical rotations, lengthy ethics approval pathways, and minimal undergraduate education, leading to a limited research skill set [1]. Artificial intelligence (AI) is increasingly being integrated into healthcare research [2]. Although it cannot replace the scientific foundations of research, its utilisation can be beneficial to facilitate various steps in research such as search strategies, data extraction, database development, manuscript set‐up and referencing. This perspective highlights the current ways that AI can support junior clinicians to develop research projects, whilst recognising its limitations and risks. Where Artificial Intelligence May Help AI tools are rapidly evolving and span multiple stages of research development, from idea generation and literature exploration through to manuscript preparation and journal selection as summarised in Table 1 and Figure 1. The integration of AI into research should be approached in a way that supports, rather than replaces, the development of core research skills. Trainees should use AI to assist with structuring ideas, refining workflows, and improving efficiency while maintaining active engagement with study design, critical appraisal, and data interpretation. TABLE 1Practical applications of artificial intelligence across the surgical research workflow.Research stageKey tasksAI functionExample AI toolsIdeaIdentify research questions, explore gapsLiterature summarisation, gap identificationElicit; Consensus; SciteLiteratureSearch strategy, identify relevant studiesSemantic search, mapping, screening supportSemantic Scholar; Perplexity; Research Rabbit; RayyanProtocolDevelop study design, draft protocolStructured drafting, methodological suggestionsClaude; SciSure; TrialKey; ChatGPTEthicsPatient information sheets, consent formsPlain‐language simplification, readability optimisationChatGPT; ClaudeDataData extraction, database creationAutomated extraction, data structuringSciSpace Data Extractor; Elicit; ChatGPT (REDCap design)AnalysisStatistical planning, data interpretationSuggest tests, generate code, explain outputsChatGPT (R/Python); JASP; IntellectusManuscriptDrafting, editing, formattingLanguage refinement, structure optimisationChatGPT; Writefull; GrammarlyJournalJournal selection, submission strategyJournal matching, scope alignmentElsevier Journal Finder; Wiley Journal Finder; Springer Journal Suggester Practical applications of artificial intelligence across the surgical research workflow. Conceptual framework for the use of artificial intelligence across the surgical research workflow. Topic Identification and Literature Exploration One of the earliest challenges in research is the identification of a relevant and feasible research question. This is where access to mentorship is essential. Trainees should seek supervisors through established institutional networks, trainee research collaboratives, and academic departments, using AI as a supplementary tool rather than a substitute for expert guidance. By using AI‐assisted academic search platforms, analysis of large bodies of literature can identify emerging themes and potential gaps in current clinical evidence [3]. These tools may help trainees gain an overview of a research field more rapidly than traditional search methods, which require a prior ‘expert’ understanding of the topic, helping with discussions with mentors and refinement of a research question. AI systems can also assist with focusing on literature search strategies, in particular with systematic reviews, by suggesting relevant keywords, synonyms, and indexing terms, where librarian services previously would be an essential aspect of guidance. Study Design and Protocol Development Developing a research protocol or clinical trial protocol requires familiarity with study design principles, governance, reporting guidelines, and clinical research methodology. AI tools may help generate examples of structured protocol templates or assist with drafting background sections and research aims [4]. Emerging platforms can also provide guidance on clinical trial design specific to the clinical area, including suggestions for eligibility criteria, outcome measures, and methodological frameworks [4]; providing trainees with limited prior experience a useful scaffold to follow. AI may also assist with drafting participant information sheets and plain‐language summaries, which are often challenging to write at a language level that participants easily understand. Emerging evidence suggests that AI‐assisted plain‐language summaries may improve readability and reduce drafting time compared with traditional approaches [5]. In this comparative study, AI‐generated plain‐language summaries were more readable and required less drafting time than those produced using traditional approaches [5]. Evidence Synthesis and Data Organisation Systematic reviews and evidence synthesis are common entry points into academic surgery but can be both complex and highly time‐intensive. AI‐assisted tools may help streamline parts of this process by supporting the organisation, extraction, and preliminary synthesis of research data, allowing investigators to work more efficiently through large volumes of literature [6]. While human appraisal remains critical to evaluating study quality and bias, AI can reduce the time required for preliminary data extraction. AI systems may also assist in designing electronic research databases, including generating suggested variable structures for platforms such as REDCap, which traditionally required expertise, time, and money to set up. Statistical Planning, Manuscript Preparation and Submission Statistical analysis can represent a significant challenge for trainees without formal training in statistics. AI‐assisted tools may provide preliminary analytic support by helping trainees structure workflows, organise outputs, and approach data interpretation in a more systematic way [7]. Such guidance can serve as a starting point for discussion with experienced statisticians, especially when timely access to statistical support is often limited. In a survey of radiology trainees in the UK, up to a third had limited statistical support access, which significantly hindered research outputs [8]. AI may also assist with manuscript preparation by generating structured outlines aligned with journal requirements, improving grammar and readability, and ensuring adherence to word limits and formatting guidelines. Automated referencing tools can further streamline citation formatting and reference list management. Finally, using large language models (LLMs) to aid submission to the right journal is often helpful, in particular screening those with minimal timelines for review, correct audience, and reduced costs or waivers for submission or open access publication. Limitations and Risks Despite its promise, AI has important limitations in the research environment. Generative AI systems may produce inaccurate or fabricated information, including references that do not exist [9]. Without careful verification, reliance on such outputs could compromise the integrity of scientific work. AI systems also lack contextual understanding of clinical nuance and novelty. Identifying meaningful research questions requires insight into patient care, surgical practice, and the existing evidence base—skills that cannot be replicated by AI. Additionally, the use of AI in research is also governed by an emerging as well as evolving ethical and professional standard. Current guidance emphasises transparency, verification of outputs and appropriate disclosure of AI use in academic writing. Trainees, therefore, need to ensure accuracy, appropriate, honest use and adherence to institutional and journal‐specific policies [9, 10]. Finally, excessive dependence on AI tools could limit the development of core research competencies. Study design, data interpretation, and scientific writing are essential skills that trainees must acquire through mentorship and experience [10]. The risk is not that AI will replace researchers, but that it may produce superficial work without true understanding of a topic. Conclusion Surgical research remains essential to evidence‐based practice and advancement of patient care. Yet for many trainees, the pathway into research remains difficult to navigate. Artificial intelligence has emerged as a potential tool that may help reduce some of these barriers. By streamlining tasks such as literature exploration, protocol drafting, data organisation, and manuscript preparation, AI may allow trainees to focus more on the intellectual aspects of research. However, AI cannot replace the core skills that define good research: critical thinking, methodological rigour, and mentorship. Its role should therefore be viewed not as a substitute for research training but as a scaffold that supports trainees navigating the early stages of academic work. Author Contributions Michael Rouse: writing – original draft, writing – review and editing. Yumna Kaleem: writing – original draft, writing – review and editing. Justin M. Yeung: conceptualization, writing – review and editing. Acknowledgments Open access publishing facilitated by The University of Melbourne, as part of the Wiley ‐ The University of Melbourne agreement via the Council of Australasian University Librarians. Data Availability Statement Data sharing not applicable to this article as no datasets were generated or analysed during the current study. References Associated Data Data Availability Statement Data sharing not applicable to this article as no datasets were generated or analysed during the current study.