Faculty Perspectives on AI Integration in Anatomy Education in the United Arab Emirates: Cross-Sectional Survey

Artificial intelligence (AI) is reshaping medical and health professions education; yet, adoption in anatomy remains uneven and often ad hoc. Anatomy's spatial and visualization demands make it a compelling domain for AI, but discipline-specific opportunities and risks are not well characterized in the United Arab Emirates.

This study examines United Arab Emirates anatomy educators' AI use, attitudes, perceived barriers and enablers, and strategic perspectives on AI integration using a design informed by the Unified Theory of Acceptance and Use of Technology 2 (UTAUT2).

A cross-sectional survey of anatomy faculty at United Arab Emirates medical and health sciences colleges used 5-point Likert items to assess educational technology proficiency, AI use patterns, AI attitudes, perceived barriers and facilitators, and professional development needs. Quantitative data were summarized descriptively and explored with nonparametric tests. Open-ended strengths, weaknesses, opportunities, and threats questions were analyzed using reflexive thematic analysis, organized within the strengths, weaknesses, opportunities, and threats framework, and interpreted through UTAUT2 constructs. Quantitative and qualitative strands were integrated at interpretation through triangulation.

In total, 30 anatomy faculty participated. Self-rated educational technology proficiency was high (mean 3.73 out of 5, SD 1.01), and overall attitudes toward AI in anatomy education were positive (mean 4.23, SD 0.73), with strong interest in AI-focused professional development (mean 4.50, SD 0.73). Most respondents reported using generative AI tools, predominantly ChatGPT, for content creation, quiz and examination item generation, summarization of complex material, and, to a lesser extent, visualization and workflow streamlining. Capacity-related barriers predominated: time and workload pressures (mean 3.27, SD 1.17) and training gaps (mean 3.13, SD 1.22) were rated as moderate obstacles, whereas budget or resource limitations (mean 2.63, SD 1.19) and academic integrity concerns (mean 2.80, SD 1.10) were minor obstacles. Student interest (mean 4.23, SD 0.86) and institutional encouragement (mean 4.00, SD 1.14) emerged as strong facilitators, with no statistically detectable differences by academic rank, age, or years of experience in this small, underpowered sample. Qualitatively, themes highlighted strong institutional support and digital readiness as strengths; training needs, workload, and policy gaps as weaknesses; visualization, personalization, and efficiency as opportunities; and overreliance, ethical risks, and erosion of hands-on anatomy pedagogy as threats. UTAUT2 interpretation indicated high performance expectancy and social influence (student and institutional support) but reduced effort expectancy and facilitating conditions due to time, training, and governance constraints, collectively tempering behavioral intention.

In this exploratory sample, United Arab Emirates anatomy educators were broadly receptive to generative AI and already experimenting and valuing the benefits for 3D visualization, adaptive practice, and feedback. However, workload, limited training, and unclear governance (disclosure, assessment integrity, and cadaveric or patient images) constrain uptake, underscoring the need for protected time, workflow-aligned training, and discipline-specific policies to enable sustainable, ethical integration.