Retos éticos y metodológicos en la integración de inteligencia artificial en auditoría

Abstract

The study examines the ethical and methodological challenges of integrating artificial intelligence (AI) into financial auditing, a field that significantly transforms control processes in a digitalized world. The research highlights benefits such as automation and predictive analytics, but warns of ethical risks linked to a lack of algorithmic transparency, biases derived from historical data, and ambiguity in the assignment of responsibilities, factors that affect fairness and confidence in the results. It also identifies methodological challenges such as the dependence on reliable data, interpretive errors generated by algorithmic complexity, and vulnerability to cyberattacks, which compromise the reliability of audit procedures. The study, developed with a qualitative approach and a critical review of specialized literature, organized its analysis into six main categories. The findings highlight the need to implement transparency standards, more rigorous data validation processes, and ongoing training programs for auditors. The development of robust regulatory frameworks and the creation of internal ethics committees to ensure the responsible use of AI are also recommended. The study concludes that managing these challenges requires a balance between technological innovation, professional education, and clear regulations, which always maintains a human component in the oversight of automated audits. In Latin America, where adoption is still in its infancy, there is an opportunity to establish robust guidelines before mass implementation, which represents a key area for future research.