Generative AI Powered by Large Language Models for Leadership Development: Between Epistemic Responsibility and the Future of Innovation

This chapter examines language-based generative AI, particularly large language models (LLMs), as active epistemic agents in leadership development. Moving beyond their perceived neutrality, the authors argue that LLMs serve as mirrors of dominant assumptions, co-constructors of leadership meaning, and framing devices of legitimate knowledge. The sources highlight how these systems reinforce prevailing Euro-American managerial ideals, often marginalizing diverse leadership paradigms such as Indigenous, collectivist, or spiritual frameworks, thereby amplifying structural and cultural biases. Through epistemic feedback loops and personalization, LLMs can subtly shape leadership identity and constrain developmental horizons, leading to adaptive standardization or performative conformity. The chapter emphasizes the critical need for epistemic responsibility in innovation leadership, advocating for leaders to interrogate system framings, reflect on their influence, and cultivate design awareness. This involves fostering inclusive data practices, participatory design methodologies, and culturally responsive AI governance to ensure AI-mediated innovation ecosystems are just, inclusive, and reflect pluralistic worldviews. Ultimately, the chapter calls for leaders to reclaim framing power and engage AI dialogically to foster an innovation leadership that is reflexive, ethically grounded, and attuned to human complexities.