Utilization of Artificial Intelligence (AI) Tools for Health Information among Postgraduate Students: A Case of Daystar University, Kenya

Abstract

This study explored the utilization of Artificial Intelligence (AI) tools for health information among postgraduate students at Daystar University, Kenya. The research was guided by four objectives: to examine students’ perceptions of AI tools in accessing health information, to identify the kinds of health information sought using AI applications, to explore how students assess the credibility of AI-generated health information, and to determine the AI tools most preferred and frequently used by postgraduate students. Anchored on the Technology Acceptance Model (TAM) and the Health Belief Model (HBM), the study examined how perceived usefulness, perceived ease of use, attitudes, health beliefs, and self-efficacy influence postgraduate students’ engagement with AI-based health tools. The study adopted a qualitative approach using an exploratory research design to gain an in-depth understanding of students’ experiences and practices. Purposive sampling was used to select participants from the School of Communication at Daystar University, Nairobi Campus, the student enrolled during the 2024–2025 academic year. Data were collected through focus group discussions (FGDs) guided by semi-structured interview schedules. Supporting instruments included consent forms, demographic data sheets, observation checklists, and transcription templates to ensure a systematic and ethical data collection process. Expert review and pilot testing were used to establish instrument validity. Data were analyzed thematically and supported by content analysis to identify key patterns and insights. The findings revealed that postgraduate students perceive AI tools as useful, efficient, and easy to use for health information seeking. They primarily relied on AI for preventive, diagnostic, and mental wellness information, while emphasizing verification through professional consultation and cross-checking with authoritative sources. Students preferred tools such as ChatGPT, Google Bard (Gemini), and specialized fitness or health applications due to their accessibility, interactivity, and personalized feedback. However, concerns regarding information credibility, data privacy, and lack of localized health content were prominent. Participants emphasized the need for culturally relevant AI systems, transparency in information sourcing, and institutional oversight in guiding ethical use. The study concludes that while AI tools enhance access to health information and promote digital health literacy, their optimal use requires contextual adaptation, privacy safeguards, and user education. It recommends that universities integrate AI literacy into postgraduate curricula, establish partnerships with verified AI platforms, and develop ethical guidelines for AI-based health information seeking. Further research is suggested to explore longitudinal trends, cross-institutional comparisons, and the integration of localized data in AI health systems across sub-Saharan Africa.