This paper presents the design, development, and evaluation of a chatbot system powered by generative artificial intelligence (AI) to automate and improve administrative services for university students. The system utilizes large language models (LLMs) to deliver context-aware responses to common inquiries about enrollment, certifications, and academic records. Developed with usability and accessibility in mind, the chatbot was deployed in a university environment and evaluated by 100 students. After interacting with the system, participants completed the System Usability Scale (SUS) to assess its effectiveness. The chatbot received an average SUS score of 82.3, indicating a high level of usability and satisfaction. Users valued its availability outside office hours and its ability to streamline procedural interactions. The system architecture incorporates supervised learning and safeguards to mitigate misinformation and support ethical compliance. This study highlights the potential of generative AI to enhance student services and reduce administrative workload in higher education. It also emphasizes the importance of context-specific training, human oversight, and continuous evaluation to ensure the responsible deployment of such technologies in academic environments.