EDuBot: a sustainable and integrated AI-driven chatbot for advising in higher education

Purpose The adoption of ChatGPT in higher education (HE) has not only highlighted its potential but also exposed critical limitations, including hallucination risks and poor institutional integration. The purpose of this study is to examine the transformative impact of EDuBot, an AI-driven chatbot integrated into a collaboration and communication platform (CCP), by assessing its role in pedagogical support and curriculum advising. The work aligns with UN Agenda 2030, contributing to SDGs 4, 8, 9 and 12 (United Nations, 2023). Design/methodology/approach A two-phase evaluation method was conducted. In the first phase, an exploratory case study illustrated efficiency gains when the chatbot was applied to study plan advising and curriculum guidance. In the second phase, a survey of 112 students measured service quality (SQ), information quality (IQ) and satisfaction in pedagogical advising. Findings Survey results confirmed high satisfaction (mean = 3.86/4) and validated that SQ and IQ together explain more than 50% of the variance in student satisfaction. These operational outcomes serve as proxy indicators supporting the estimation of EDuBot’s contribution to SDG-aligned objectives, including improved access to advising, a 20.25-h reduction in administrative workload per semester, scalable digital infrastructure for academic services and energy reductions of up to 60% for comparable query–response tasks. Research limitations/implications This study is limited to two institutions, within a single discipline (Industrial Engineering) and over a short evaluation period. Sustainability impacts are inferred from self-reported measures and literature-based benchmarks rather than direct system-level measurements. Future research should adopt longitudinal, cross-disciplinary designs and integrate direct operational and environmental metrics. Practical implications RAG based Chatbots can improve advising operations by reducing repetitive workload, enhancing the quality and consistency of guidance, increasing institutional efficiency, and improving working conditions for academic staff. Social implications By offering multilingual, 24/7 and unbiased access to advising services, EDuBot can widen access to academic support, promote educational inclusion, and reduce barriers faced by diverse student populations. Originality/value This study provides one of the first empirical evaluations of a RAG-based academic advising chatbot embedded in an institutional CCP, linking Information Systems theory with operational validation and proxy-based sustainability assessment.