La Inteligencia Artificial como Apoyo, No Sustituto: Análisis Multisistema de Escenarios Éticos en Recursos Humanos

Abstract

This study presents a comparative analysis of three advanced artificial intelligence systems (ChatGPT, Gemini, and Claude) in resolving complex ethical dilemmas within human resources management in industrial settings. The systems were evaluated through structured ethical scenarios simulating real-world HR decision-making situations, including conflicts of interest, workplace equity dilemmas, and hiring/termination decisions with moral implications. All three systems demonstrated significant analytical capabilities in identifying relevant ethical factors and structuring arguments. However, notable differences emerged in their approaches to contextual sensitivity and consideration of human nuances. ChatGPT excelled in systematic analysis, Gemini in integrating multiple perspectives, and Claude in considering emotional implications. The findings underscore that while AI systems offer valuable capabilities in information analysis and structuring, they cannot and should not replace human judgment in sensitive ethical decisions. Empathy, moral intuition, and deep contextual understanding remain exclusively human competencies. AI should function as a support tool that enriches human decision-making processes by providing comprehensive analysis and alternative considerations, while preserving moral authority and final responsibility in human hands. This study contributes to developing ethical frameworks for responsible AI implementation in human resources, emphasizing the importance of designing systems that enhance rather than substitute human ethical judgment capabilities. The research supports a human-centered approach to AI integration, where technology serves to amplify human wisdom rather than replace human conscience in morally sensitive organizational decisions. These findings have direct applications for HR practitioners and organizational leaders seeking to leverage AI benefits whionnel decisions.

Ähnliche Arbeiten