Preparing ethical AI practitioners in pharma: challenges and strategies in higher education

The rapid integration of artificial intelligence (AI) into pharmaceutical sector has transformed the traditional paradigms of drug discovery, clinical trials, manufacturing and patient management (Patel et al., 2024; Singh et al., 2024; Dovzhuk et al., 2025). Using AI-driven approaches, we accelerate innovation, improve operational efficiency and increase the potential for personalized therapy (Bi et al., 2019). However, these advances simultaneously bring new ethical, technical and regulatory challenges, particularly in terms of data quality, patient privacy and compliance with evolving regulatory frameworks (Babu et al., 2021; Patil et al., 2023; Patel et al., 2024; Singh et al., 2024; Dovzhuk et al., 2025).

Since the pharmaceutical sector increasingly depends on interdisciplinary teams composed of Pharmacists, data scientists, engineers and regulators, higher education programmes face increasing pressure not only to equip graduates with advanced technical skills, but also to gain a strong understanding of ethical principles and regulatory requirements (Babu et al., 2021; Shinde et al., 2023; Ittenbach, 2023; Sharma et al., 2024). Recent literature has recognized the importance of embedding ethics, data governance and regulation literacy into the curriculum of the integrated artificial intelligence (IA) to prepare “industry-ready” professionals capable of managing the complex realities of implementing IA in pharmaceutical environments (Babu et al., 2021; Shinde et al., 2023; Patil et al., 2023; Ittenbach, 2023). Nevertheless, the existing educational model often lacks a comprehensive framework to address these multifaceted challenges in a systematic manner, and the comprehensive educational strategies explicitly designed in this context are unusually scarce.

Consequently, the purpose of the research is to address this gap and investigate how universities can develop education programmes to prepare ethical AI professionals for the pharmaceutical industry. The research focuses on identifying major industrial challenges (including data quality, ethics, regulatory complexity) and analysing educational and curriculum strategies aimed at ensuring that graduates are fully prepared to implement AI responsibly in the real pharmaceutical environment, such as explanation AI, data enhancement and regulatory literacy activities.

With the rapid expansion of the application of AI in the pharmaceutical industry, the demand for ethical, technically capable and industry-ready professionals has increased significantly. Designing a higher education programme that effectively prepares graduates for the ethical and regulatory complexity of the implementation of real AI in the pharmaceutical industry is a priority of emerging research and practice. The review of the literature aims to synthesize current research environments on the integration of AI, ethical frameworks and regulatory knowledge into medical education, and to pay special attention to pedagogical strategies and education interventions that are oriented to the urgent needs of the industry. Figure 1 presents the literature selection process.

Table 1 represents data of year 2023 about academia, industry and pharma companies:

• Academia: The adoption of AI in drug discovery is growing within academia, but its use is not yet universal. One-third (33%) have not adopted AI, while only 3% report extensive use. A gradual shift towards substantial use (27%) is evident (Coherent Solutions, 2025; Yadav et al., 2024).

• Industry: AI adoption is rapidly accelerating across the global pharmaceutical industry, transitioning from an exploratory tool to a core component of the research and development pipeline. Overall, 62% have adopted it, with 38% report occasional to extensive use. Extensive adoption stands at 12% (Coherent Solutions, 2025; Yadav et al., 2024).

• AI-first biotechs: The adoption of AI is fundamental and widespread in AI-first biotech companies, where it is the foundational technology for accelerating and de-risking drug discovery. With 75% using it substantially and 25% extensively. No organizations in this category report limited or no use (Coherent Solutions, 2025; Yadav et al., 2024).

• Pharmaceutical/biotech companies: Pharmaceutical and biotech companies are increasingly integrating AI technologies into their drug discovery pipelines (Dara et al., 2022; Sarkar et al., 2023). 58% are using AI drug discovery with 8% reportedly using extensively. Adoption is mostly occasional (27%) (Coherent Solutions, 2025; Yadav et al., 2024).

AI-first biotechs are leading in AI adoption. The academia and industry show moderate progress. The traditional pharmaceutical companies are slow to adopt AI.

Figure 2 provides a clearer representation of Table 1.

Relative Recency categorizes each contribution based on its publication year. Works older than ten years are labelled as oldest, those between six and ten years as older, and those between three and five years as mid/intermediate. Publications from the last two years are recent, while works from the previous year are very recent. Research works published in the current year are considered as most recent. This classification helps readers to quickly assess how the literature base has evolved. It also helps to understand which studies are most relevant to ongoing discussions. Table 2 represents the relative recency classification:

1. The date shows when the work was published, providing a temporary insight into how research evolved.

2. Author column shows the researcher who made the contribution.

3. Relative recency places each work in its oldest, oldest, intermediate, most recent, very recent, or most recent, based on the publication date. This allows readers to know the context and the applicability of the research in present-day trends.

4. The number of citations (low, medium, high) indicates the scholarly acceptance of each work. In addition, it tells us about work’s impact on subsequent research:

   • A work with fewer than 20 citations is considered under low citation volume.

   • Work with 20–100 citations presents a medium citation volume.

   • A paper that has over 100 citations is considered to have high academic prestige and great impact on subsequent studies. Thus presents a high citation volume.

Table 3 represents previous contributions by various authors. In brief, Williams (2014) is the earliest entry with a low number of references. This indicates the basic but less frequent references. On the other hand, Cotton et al. (2023) and Info (2023) are recent studies with high numbers of references, suggesting their modernity and influence. Singh and Huang (2025) is the most recent, but there are currently low citations typical of newly published works. This map allows readers to quickly assess what research is the fundamental, what field is currently being formed and what may become a future influencer.

Figure 3 presents a graphical overview of previous contributions with citation volumes categorized as high, medium and low.

Table 4 presents the ore focus by the lead authors and a Brief Comparison of Research Articles with key contributions and research gaps.

The research consistently identified several central challenges in the effective and responsible implementation of AI in the pharmaceutical sector:

Many sources highlight problems associated with data origin, non-standardized schemas in electronic health records (EHR), data leakage and the effects of data –omics in batches (Babu et al., 2021; Patel et al., 2024; Singh et al., 2024; Dovzhuk et al., 2025). The development of reliable AI systems is crucial to addressing these challenges, which requires a re-thinking of data strategy and integration specifically for AI applications (Dara et al., 2022).

Ethical issues are generally recognized, particularly regarding the secondary use of clinical studies and claims data and the risk of re-identification of patients (Babu et al., 2021; Patil et al., 2023; Dovzhuk et al., 2025). It is referred to the importance of technologies and frameworks for preserving privacy to ensure the safety and consent of patients (Xiang and Cai, 2021; Forcier et al., 2019).

The evolving regulatory landscapes of the FDA, EMA and ICH are often cited as obstacles to the application of AI in pharmaceuticals. The papers point out that professionals must map the life cycles of AI models to regulatory milestones and understand frameworks such as the FDA’s Guide for Good Machine Learning Practices (Patel et al., 2024; Singh et al., 2024; Dovzhuk et al., 2025). This includes understanding how AI impacts areas like pharmacovigilance and drug safety (Ball and Dal Pan, 2022; Bate and Stegmann, 2023).

Despite these challenges, only a minority of the literature addresses the direct translation of industry requirements into the design of higher education programmes:

Systematic Review and Framework Proposals (Babu et al., 2021) provide a systematic review focusing on AI in pharmaceutical management education in India, pointing out that there is no widely accepted curriculum that integrates AI and ethical guidelines (Sharma et al., 2024). discusses the development of a PhD-level curriculum on quantitative methods for the development of drugs, emphasizing the value of public-private partnerships and regulatory stakeholder participation, although it does not specifically focus on AI or ethics.

Case studies of curriculum integration (Shinde et al., 2023) provide cases of AI integration in the education of pharmaceutical engineering, reporting increased participation of students and the development of interdisciplinary skills through modules including case studies, interactive seminars and virtual laboratories.

The incorporation of emerging educational tools (Patil et al., 2023) emphasizes the use of AI-powered simulations and virtual reality laboratories in personalized medical training and the content of the curriculum on ethical and regulatory considerations.

Several works recommend broad strategies without providing detailed implementation frameworks or empirical evaluations:

• Interdisciplinary curriculum: There is consensus on the need for curricula that bridge AI, data science, life sciences and regulatory affairs (Shinde et al., 2023; Ittenbach, 2023; Sharma et al., 2024), covering everything from fundamental machine learning algorithms to their real-world applications in drug discovery (Dara et al., 2022; Sarker, 2021).

• Inclusion of ethical and regulatory content: Papers stress the value of teaching students about privacy-preserving technologies, ethical trade-offs and regulatory requirements (Babu et al., 2021; Patil et al., 2023; Dovzhuk et al., 2025). However, most suggestions remain at the conceptual level.

• Experiential learning and industry collaboration: Case-based learning, practical projects and stronger industry-academic partnerships are proposed as means to increase practical and ethical readiness (Shinde et al., 2023; Sharma et al., 2024). Also advocates for the use of virtual labs and future expansion of industry partnerships (Shinde et al., 2023).

The literature consistently reveals several important limitations:

Limited-operational courses: No identified papers have presented a fully developed, empirically validated course specifically designed to produce ethical AI professionals in the pharmaceutical industry. Most contributions are proposals at a high level or partial implementations (Babu et al., 2021; Shinde et al., 2023; Patil et al., 2023; Ittenbach, 2023; Sharma et al.,2024).

Insufficient empirical evaluation: There is little evidence of the effectiveness of education (Shinde et al., 2023). It reports an increase in the participation of students (participation, projects, patents), but there is a lack of controlled studies and an assessment of ethical/regulatory competences.

Under development of advanced educational approaches: Although tools such as virtual laboratories and AI simulations are mentioned (Patil et al., 2023), no study describes in detail critical elements such as pharmaceutical-specific ad hocAI laboratories, regulatory literacy hackathons or hands-on technology modules for improving privacy.

A review of current literature reveals a broad consensus on both the main challenges and the important educational priorities for preparing ethical AI professionals for the pharmaceutical industry.

First, there is a consistent recognition in the research that the pharmaceutical context presents distinctive challenges for the implementation of AI. Issues such as data quality and heterogeneity (including non-standardized electronic health records, data gaps and batch effects) are highlighted as the main technical hurdles to address through targeted educational interventions (Babu et al., 2021; Patel et al., 2024; Singh et al., 2024; Dovzhuk et al., 2025). Ethical considerations, especially the privacy of patients, the secondary use of health data and the application of privacy-enhancing technologies, have been repeatedly cited as a fundamental factor in the effective AI practice in pharmaceuticals (Babu et al., 2021; Patil et al., 2023; Dovzhuk et al., 2025). Simultaneously, the regulatory environment is noted for its complexity and rapid evolution, requiring professionals to understand and anticipate the demands of agencies such as the FDA and EMA (Patel et al., 2024; Singh et al., 2024; Dovzhuk et al., 2025), including the application of AI in areas like clinical trial design (Harrer et al., 2019).

Secondly, there is widespread agreement that current higher education programmes must become more interdisciplinary, blending technical AI skills with domain-specific knowledge of pharmaceutical sciences, ethical reasoning and regulatory literacy (Shinde et al., 2023; Ittenbach, 2023; Sharma et al., 2024). Some studies propose integrating AI modules into existing curriculums and emphasize the benefits of case studies, interactive laboratories, virtual simulations and industry partnerships to foster applied and ethical competencies (Babu et al., 2021; Shinde et al., 2023; Patil et al., 2023; Ittenbach, 2023; Sharma et al., 2024). Most of these recommendations, however, remain conceptual or high-level rather than providing fully developed or empirically tested curriculum models (Babu et al., 2021; Shinde et al., 2023; Patil et al., 2023; Ittenbach, 2023; Sharma et al., 2024). Thirdly, despite agreement on educational needs, the implementation and assessment of the proposed curriculum is lacking. Although some case studies have noted an increase in student engagement and involvement in interdisciplinary projects after the introduction of AI-focused modules (Shinde et al., 2023), there is little evidence to suggest that these interventions have achieved significant gains in ethical decision-making or regulatory compliance capabilities. There are no detailed, detailed, rigorously evaluated curricula that direct the identified industrial challenges to structured educational strategies (Babu et al., 2021; Shinde et al., 2023; Patil et al., 2023; Ittenbach, 2023; Sharma et al., 2024).

In summary, literature clearly converges on the need for higher education programmes to address the challenges of data, ethics and regulation in pharmaceutical AI (Kim et al., 2020; Wong et al., 2018). In addition, in the process of educating, there has long been an argument for interdisciplinary approaches, industry stakeholder engagement and active learning resource use. However, the issue of transforming these insights into rich, evidence-based curriculum frameworks showing efficacy remains an open issue and represents a main avenue for prospective inquiry and application in this area. The current research shows that the fundamental challenges faced by the implementation of ethical AI in the pharmaceutical industry are widely recognized and reinforces consensus on key elements of higher education. However, in this field there are no robust and empirically tested programme models and detailed pedagogical interventions that explicitly map educational content to industry-specific requirements. Reducing these gaps by operationalizing and rigorously evaluating specific educational strategies represents a critical opportunity for improving both the training and professional practice of pharmaceutical AI (Babu et al., 2021; Shinde et al., 2023; Patil et al., 2023; Ittenbach, 2023; Sharma et al., 2024; Dovzhuk et al., 2025).

This study uses an exhaustive literature review method in synthesizing existing knowledge and practices on designing higher education programmes in educating ethical AI experts for the pharmaceutical sector. The study is organized as follows:

Systematic literature search in peer-reviewed journals from 2013 to 2025 was done with an emphasis on where AI, pharmaceutical education, ethics and regulation overlap. The search strategy involved specifically chosen keywords like “artificial intelligence”, “pharmaceutical”, “education”, “curriculum”, “ethics” and “regulation” modified from well-documented search protocols (Babu et al., 2021):

• The SCOPUS database was used as the source of data, in accordance with best evidence in previous systematic reviews (Babu et al., 2021).

• Screening and selection criteria were guided by relevance to three core pillars (Babu et al., 2021): explicit discussion of higher-education curriculum/programme design for ethical AI in pharma (Shinde et al., 2023); identification of pharmaceutical industry challenges (data quality/heterogeneity, ethics/privacy, regulatory complexity); and (Patil et al., 2023) proposal or evaluation of pedagogical or curricular strategies linked to these challenges.

• The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) framework was applied to ensure transparency and reproducibility in the inclusion and exclusion of articles (Babu et al., 2021).

Data was gleaned from chosen articles about:

• Industry-specified challenges of interest for AI for pharma (e.g. data heterogeneity, ethics and patient privacy, regulation).

• Explanation of curricular interventions or learning structures especially crafted to combat such issues (e.g. incorporation of AI modules, training in ethics, courses in regulatory literacy, and experiential tools such as case studies or simulated laboratories).

• All existing information on the intervention’s success (e.g. measures of engagement, student responses, if any).

Narrative and thematic analyses were conducted to delineate patterns, areas of agreement and gaps in the literature (Patil et al., 2023; Ittenbach, 2023; Anonymous authors, 2025; Roy et al., 2025).

Where available, empirical studies and case examples were reviewed to evaluate reported outcomes associated with specific curricular models, including engagement rates and interdisciplinary competencies (Shinde et al., 2023). Commentaries, stakeholder proposals and public–private partnership models were also included to provide insight into collaborative approaches to curriculum design (Sharma et al., 2024).

This literature-driven approach guarantees a systematic, transparent and targeted synthesis of existing studies on educating ethical AI professionals for the pharmaceutical sector with special reference to identification of challenges specific to the sector and curricular approaches suggested to remedy them (Babu et al., 2021; Shinde et al., 2023; Patil et al., 2023; Ittenbach, 2023; Sharma et al., 2024; Singh et al., 2024; Anonymous authors, 2025; Dovzhuk et al., 2025; Roy et al., 2025).

Findings are summarised into two sections: Industry Challenges Identified and Educational and Curricular Approaches.

The following industry challenges have been identified.

There is strong agreement in the literature that addressing poor data quality and data heterogeneity is a critical challenge for AI implementation in the pharmaceutical sector. Non-standardized electronic health record schemas, missing data patterns and batch effects in omics data are repeatedly cited as obstacles that AI professionals in pharma must be trained to handle (Babu et al., 2021; Patel et al., 2024; Singh et al., 2024; Dovzhuk et al., 2025), underscoring the need for sophisticated data strategies (Aldoseri et al., 2023).

The reviewed works consistently highlight ethical issues as central to the pharmaceutical context, particularly regarding secondary use of clinical and claims data, risks of patient re-identification and compliance with privacy requirements (Babu et al., 2021; Patil et al., 2023; Dovzhuk et al., 2025). There is explicit mention of the need to understand and apply privacy-enhancing technologies within educational settings, although detailed pedagogical instructions are lacking (Xiang and Cai, 2021; Forcier et al., 2019).

The complexity and dynamic character of pharmaceutical regulation, such as FDA, EMA and ICH guidelines, is recognized as a significant challenge for industry and education. There is broad agreement that graduates need to be equipped to navigate and implement these regulatory guidelines in the development and deployment of AI systems (Patel et al., 2024; Singh et al., 2024; Dovzhuk et al., 2025) including understanding the role of AI in modern drug discovery and development (Sarkar et al., 2023) and pharmacovigilance (Ball and Dal Pan, 2022; Bate and Stegmann, 2023).

The following Educational and Curricular Approaches have been analysed.

Few studies explicitly give examples of incorporating the AI along with ethics and regulation content into pharmaceutical professionals’ higher education programmes:

• Babu et al. (2021) present a systematic review indicating the paucity of curricula in pharmaceutical management education that have comprehensive treatment of AI and ethics.

• Shinde et al. (2023) describe a case study of applying AI modules in pharmaceutical engineering curricula and finding benefits to student interest, interdisciplinarity and involvement in academic competitions and research projects.

• Patil et al. (2023) discuss the integration of AI in pharma education, with a focus on simulations and virtual reality (VR) labs, but without comprehensive operational models.

Wide-ranging pedagogical recommendations are the addition of interdisciplinary modules, case studies, virtual and interactive laboratories, and university–industry partnerships (Shinde et al., 2023; Patil et al., 2023; Ittenbach, 2023; Sharma et al., 2024). These should cover core AI concepts in drug discovery (Chen et al., 2023; Kim et al., 2020) and their applications.

Shinde et al. (2023) outline interactive learning processes and recommends greater industry collaborations and utilization of virtual labs for practical skill acquisition.

Patil et al. (2023) emphasize AI-driven simulation instruction and VR application, in addition to curricula which include ethical and regulatory considerations.

• While many articles invite the incorporation of ethics, data science and regulatory literacy into pharmaceutical AI education, few operationalize comprehensively developed or empirically validated models of education (Babu et al., 2021; Shinde et al., 2023; Patil et al., 2023; Ittenbach, 2023; Sharma et al., 2024).

• Pedagogical approaches like explainable AI, privacy-enhancing technologies workshops and regulatory “hackathons” are occasionally proposed but not comprehensively described and assessed.

• There is limited evidence for the success of implemented curricular interventions. Only Shinde et al. (2023) give outcome-related measures (e.g. higher participation rates), and none of the sources reviewed provide a formal evaluation of improvements in students’ ethical thinking or regulatory abilities.

• Everyone agrees that a successful curriculum has to be interdisciplinary, integrating technical AI skill sets, domain-specific pharmaceutical science knowledge, regulatory skills and firm foundation in ethics (Shinde et al., 2023; Ittenbach, 2023; Sharma et al., 2024).

• Hands-on industry engagement and experiential learning (e.g. via projects and labs) are commonly recognized as valuable but are not well-documented or empirically examined (Shinde et al., 2023; Sharma et al., 2024).

Generally, the research environment is marked by acknowledgment of urgent challenges and general, conceptual suggestions for curricular reform, but is missing concrete, evidence-based programme designs or educational interventions. Consequently, though there is settled agreement about interdisciplinary, ethics-focused and industry-relevant education in this field, operationalization and empirical testing to bridge the skills gap for ethical AI professionals in pharma is an ongoing need for future research and institutional practice (Babu et al., 2021; Shinde et al., 2023; Patil et al., 2023; Ittenbach, 2023; Sharma et al., 2024).

The literature indicates strong agreement on the core challenges (data, ethics, regulation) to AI in pharma and the significance of interdisciplinary, ethics-driven education and practically focused higher education programmes. There is, however, significant absence of detailed curricula, pedagogical instantiation and stringent evaluation of educational effectiveness in enabling ethical AI professionals for the pharmaceutical industry (Babu et al., 2021; Shinde et al., 2023; Patil et al., 2023; Ittenbach, 2023; Sharma et al., 2024).

The systematic review of new literature highlights both advances and continued challenges in constructing higher education programmes to prepare ethical AI workers for the pharmaceutical sector. The results identify high agreement about the main challenges data quality, ethical issues and regulatory sophistication but also disclose wide gaps in addressing these challenges through curriculum development and pedagogy.

The universal awareness of data heterogeneity, patient privacy threats and intricate regulatory landscapes accentuates the rising value of niche skill sets in pharma AI (Babu et al., 2021; Patil et al., 2023; Patel et al., 2024; Singh et al., 2024; Dovzhuk et al., 2025). It is certain that core knowledge of computer science or data science is inadequate; AI experts in the industry must also learn GxP-compatible data management, privacy-enhancing technologies (Xiang and Cai, 2021) and pharmaceutical regulatory complexities (Babu et al., 2021; Patil et al., 2023; Dovzhuk et al., 2025). Notwithstanding this, the explicit alignment of these issues to specific learning outcomes or instructional units is limited in the literature.

Few studies report early attempts at curriculum reform. For instance, Shinde et al. (2023) report improved student engagement and interdisciplinary skill-building following the integration of AI modules into pharmaceutical engineering programmes. Likewise, Patil et al. (2023) report the implementation of AI-based simulations and VR laboratories for personalized medicine education. Yet, these efforts are typically outlined at an overall level, and information on content, test methods or alignment with particular industry requirements is limited. Furthermore, although the majority of papers reviewed recommend the integration of ethics and regulatory understanding (for instance, through case study or inter-disciplinary modules) (Babu et al., 2021; Shinde et al., 2023; Patil et al., 2023; Ittenbach, 2023; Sharma et al., 2024), few present definite models for incorporating these aspects in a manner that expressly equips graduates for actual application of pharmaceutical AI, such as using AI for clinical trial design (Harrer et al., 2019) or pharmacology research (van der Lee and Swen, 2023).

A significant gap is the non-availability of empirically confirmed operationalized curricula that take all the combination of AI technical competencies, pharmaceutical background, ethics and regulatory needs comprehensively into account. While various references propose industry collaboration and learning by experience (e.g. project-based learning, virtual labs or simulation exercises) (Shinde et al., 2023; Patil et al., 2023; Sharma et al., 2024), there is no rigorous study of the efficacy of these measures on student outcomes.

The literature also indicates that regulatory landmarks and ethical trade-offs in AI model deployment which are necessary for FDA or EMA approval are not sufficiently covered in existing curricula. Use of methodologies like “regulatory hackathons”, privacy-enhancing-technology workshops and explainable AI labs specifically tied to pharmaceutical case studies is noted only in passing, without attendant detail or assessment.

These survey results hold significant value for both practice and research. For curriculum designers and teachers, the strong agreement on key matters requires a transformative change from the ad hoc combining of modules to the development of rich, cross-disciplinary programmes designed interactively with industry practitioners and ethics panels. For researchers, this study also draws attention to the fact that simply acknowledging the gap exists is not enough; the necessary next step is to move from abstract conceptual frameworks to empirical testing of the classroom-based interventions. Therefore, future research needs to focus on the design, delivery and long-term impact of specially designed pedagogic interventions – aids such as FDA/EMA guideline-informed regulatory hackathons or project-based learning with anonymized real-world pharmacokinetic data sets – to systematically study their ability to develop measurable competence in ethical reasoning, navigating regulation and responsible application of AI to the practice of pharmacy.

To direct this vital empirical inquiry, particular research directions become paramount. Initially, it is imperative to explore and substantiate innovative pedagogical and assessment frameworks that can function within an environment saturated with AI. This encompasses the formulation of AI-resistant assessments to maintain academic integrity. At the same time development of a dependable methods for evaluating competencies that cannot be replicated by AI, such as intricate ethical reasoning. Subsequently, research should concentrate on the human–AI interaction in educational contexts. It should emphasize interventions that promote self-regulation among students and assess how exposure to AI tools influences professional judgment and the acquisition of foundational knowledge. Ultimately, to ensure that educational materials remain at the cutting edge of technology, investigations must be conducted to examine the practical applications and constraints of advanced AI. Investigations must be done to validate large language models for specific pharmaceutical applications and tackling the computational and bias issues associated with AI-generated content.

This paper provides a foundational synthesis and a clear call to action, proposing key pillars for a future curriculum model that must now be developed and empirically tested. This study has searched the literature regarding the programme design of higher education aimed at educating ethical AI professionals for the pharmaceutical sector. The existing evidence highlights a general agreement on fundamental issues such as data quality assurance, ethical risks around patient privacy management, and dealing with sophisticated and dynamic regulatory environments that graduates should be able to manage (Babu et al., 2021; Patil et al., 2023; Patel et al., 2024; Singh et al., 2024; Dovzhuk et al., 2025).

While a number of studies support more interdisciplinary and industry-relevant curricula, and report first steps towards combining AI content and portions of ethics and regulation into educational curricula (Babu et al., 2021; Shinde et al., 2023; Patil et al., 2023; Ittenbach, 2023; Sharma et al., 2024), the literature demonstrates a significant absence of full, operationalized and empirically tested curricular models specifically designed for the pharmaceutical setting. Pedagogical approaches like case studies, virtual laboratories and partnerships with industry are frequently mentioned but lack further definition and are only sporadically tested for effectiveness.

In short, higher education initiatives to serve the requirements of ethical AI practice in the pharmaceutical industry should build on the proven consensus of industry problems but need to go beyond theoretical suggestions. The discipline would be served by the development and empirically validated testing of curriculum models that clearly connect technical education to practical ethical and regulatory requirements, preparing graduates to leverage AI tools effectively and responsibly (Patil et al., 2023; Kumar et al., 2023). Pursuing this agenda continues to be a critical mission for pharmaceutical science educators, researchers and industry professionals operating at the crossroads of AI and pharmaceutical sciences.