This paper examines the ethical challenges posed by invisible artificial intelligence (AI) interventions in academic library services. As libraries increasingly adopt AI to enhance user experience and efficiency, concerns arise around transparency, accountability, and bias within automated decision-making. Invisible AI-embedded frameworks functioning without explicit user awareness-raises issues regarding user privacy, data security, intellectual freedom, and the integrity of library services. The study highlights how silent algorithms can influence information access, research behavior, and potentially reinforce stereotypes, threatening the neutrality and inclusivity of library spaces. Key issues addressed include informed consent, user autonomy, and the need for explainability in AI-driven systems. Drawing on a survey of institutional practices and review of existing literature, the paper identifies recurring ethical challenges and policy gaps, especially when commercial AI solutions are adopted. To address these, the paper proposes guidelines for ethical AI use in libraries, emphasizing interdisciplinary collaboration and the protection of user rights. It concludes with a call for comprehensive institutional policies, staff training, and ongoing ethical audits to ensure AI supports, rather than undermines, the core values of academic librarianship.