BACKGROUND With rapid expansion from functional application, artificial intelligence has entered relationship arenas that have historically been designated for human companionship and advice. Millions of people today receive ongoing emotional support from mental health chatbots and AI companions, which enable accessibility and stigma-free connection in situations where traditional counselling might not be available. OBJECTIVE This scoping review examines the ethical implications, hazards, and advantages of AI as a companion, confidant, and a counselor while taking cultural and generational setting into consideration. METHODS This scoping review was carried out on the literature in the fields of psychology, sociology, ethics, and human AI interaction from 2015 to 2025. In addition to searches in PubMed, PsycINFO, Scopus, Web of Science, and Google Scholar, Policy reports and gray literature were evaluated. Practice guidelines that were frequently cited, conceptual analyses, and empirical research were among the records that qualified. FINDINGS AI systems have the potential to improve therapeutic resources, democratize care, and lessen loneliness, especially for marginalized groups. Risks include commodification of intimacy, deterioration of genuine human ties, and over-reliance. Additionally, disparities arise, influenced by socioeconomic divisions, generational variations, and cultural norms. There are also gaps in longitudinal evaluation, cross-cultural research, and multidisciplinary integration despite growing scholarly interest. IMPLICATIONS The future of AI companionship depends on its ability to enhance human connection rather than replace it. Ethical integration necessitates research agendas that address long-term psychological and societal implications, design methodologies that incorporate authenticity and oversight and privacy and equitable safeguards. AI companionship can increase access to advice while maintaining the genuineness and tenacity of human ties with these safeguards.
Building similarity graph...
Analyzing shared references across papers
Loading...
Khadija Kamener
Technical University of Mombasa
Mount Kenya University
Building similarity graph...
Analyzing shared references across papers
Loading...
Khadija Kamener (Tue,) studied this question.
www.synapsesocial.com/papers/696719a7c0d1e3cfbfce8f86 — DOI: https://doi.org/10.70389/pjai.100020