Artificial Intelligence systems are increasingly integrated into critical decision-making contexts, raising concerns about their potential to perpetuate bias and inequality. Existing approaches to AI fairness have primarily focused on developer-led interventions, often neglecting the role of end-users in addressing these issues. This paper introduces a novel framework that positions end-users as active agents in mitigating bias. Through practical prompt engineering techniques, including prefix-based strategies, iterative refinement, reasoning-based prompting, and in-context learning, users can dynamically influence AI outputs without requiring access to system internals. The framework also addresses barriers such as moral disengagement, automation bias, and the complexity of implementation, offering solutions that enhance user engagement and foster collective efficacy. By reframing fairness as a shared responsibility, this work highlights the potential of participatory strategies to bridge the gap between technical advancements and equitable real-world applications, advancing the development of inclusive and accountable AI systems.