Relacionamentos românticos Humano-Chatbot: uma revisão sistemática e mineração de texto

Abstract

The advancement of conversational artificial intelligence has expanded the possibilities of interaction between humans and computer systems, fostering the emergence of Human–Chatbot Relationships (HCRs), in which intelligent agents can assume roles such as friends, emotional companions, therapists, and romantic partners. In this context, this study aimed to investigate the emotional, social, and gender-related impacts of human–chatbot relationships involving artificial intelligence systems. To achieve this objective, a Systematic Literature Review based on the PRISMA protocol was conducted and complemented by text mining techniques, resulting in the analysis of 33 studies published between 2020 and 2025. The results revealed a significant growth in research on the topic, highlighting the predominance of social, emotional, and romantic relationships. The main motivations identified were the search for companionship, reduction of loneliness, emotional support, and the opportunity for self-expression without judgment. Among the positive impacts, reductions in loneliness, emotional support, and the continuous availability of systems were observed. On the other hand, risks related to emotional dependency, social isolation, unrealistic expectations, privacy concerns, and ethical issues were identified. The analysis of gender-related issues revealed the recurrence of phenomena such as sexualization, objectification, and the reinforcement of traditional gender roles, as well as the limited representation of non-binary identities in the literature. It is concluded that human–chatbot relationships constitute an emerging phenomenon with the potential to promote emotional well-being, while also presenting psychological, social, ethical, and gender-related challenges. These findings highlight the need for further research and the development of more responsible and inclusive systems.