Social sustainability is a fundamental pillar for the balanced and inclusive development of contemporary societies. It embodies the commitment to social equity, inclusion, respect for human rights, and the promotion of collective well-being. Artificial intelligence (AI), with its increasing deployment, is emerging as a key tool for addressing social challenges and promoting sustainability goals, which, to date, also involve education, following the provisions of Sustainable Development Goal 4 (SDGs 4) of the UN 2030 Agenda. The integration of artificial intelligence (AI) technology into education has introduced many possibilities. However, it also raises ethical issues that require careful consideration. This paper aims to highlight the importance of considering AI as a driver for educational inclusion, starting with an analysis of its potential. The study delves into the benefits and challenges of the introduction of a generative AI chatbot by Multiversity Group, Italy’s largest group operating in the Digital Education sector and Europe’s second largest group, which, driven by its vision of AI-enhanced education, collaborated with strategic consulting firm Bain & Company to build such a generative chatbot, leveraging OpenAI technology, to provide a better learning experience for students by fostering engagement and interaction with professors, and to combine didactic delivery and interactive assistance to learners. This paper aims to explore the emerging contribution of AI in education to promote innovative and inclusive educational delivery to reduce possible educational inequalities, considering the benefits, challenges and ethical implications of implementing the technology. AI can be an enabler for social progress, but it requires attention relative to the risks of discrimination and risks related to privacy and data security. The paper analyzes the high-impact challenges faced in developing the generative AI project to ensure that student data is securely managed in compliance with legal standards, such as GDPR 2016/679, the qualitative value of responses, the need for continuous monitoring, and overcoming any technological limitations.