La inteligencia artificial como herramienta de apoyo en la fase precontractual de la contratación estatal colombiana

Abstract

The implementation of artificial intelligence tools in public administration offers opportunities to optimize government procurement processes, yet it also raises questions regarding transparency, objective selection criteria, and the accountability of those involved in decision-making. This article analyzes the opportunities and challenges associated with using such technologies in planning, sectoral studies, the preparation of preliminary studies, and bid evaluation, while also examining implementation-related legal risks and documenting relevant regulations and practices. Furthermore, the study reviews regulatory provisions, institutional documents, legal doctrine, and specialized literature concerning AI, public procurement, and digital administration. Drawing on insights gained during an academic exchange in Brazil—including expert discussions on data protection and legal liability, as well as a comparative analysis of regulatory mechanisms—we examine the risks within the Latin American context, particularly in Colombia. We acknowledge that artificial intelligence can assist in processing vast amounts of information, reviewing documents, identifying past contractual precedents, structuring documentation, and evaluating bids, thereby reducing the time spent on repetitive tasks. However, these advantages do not imply that automated systems will replace the technical and legal judgment of public officials. The use of automated systems entails risks related to algorithmic opacity, bias, personal data protection, decision traceability, and liability. Consequently, we propose that the use of these tools be governed by principles of human oversight, transparency, verification, and accountability, ensuring that technological innovation enhances public management without undermining the social rule of law.