Responsible Application of Artificial Intelligence in State

To answer the question: what algorithmic transparency is required for the responsible implementation of artificial intelligence (AI) in public administration, weighing fundamental rights of access, explainability and accountability without wasting the potential it entails? From a fundamental rights balancing approach, the first part of the chapter, based on a comparative review of international standards for reliable AI and argumentative doctrine, describes the global landscape of international law for a responsible implementation of this technology. In the second part, in order to understand this discussion at the country level, recommendations and principles of transparency incorporated in national legislation are contrasted. Finally, by means of a categorization, the progress and challenges of algorithmic transparency standards in Chile are described. The conclusions highlight the orientation taken by the legislator regarding the fact that the information of the AI systems of the public administration, according to Article 7 of Law 20.285 must be safeguarded by the body in charge of the right to public information and therefore to ensure compliance with the algorithmic transparency obligations of the automated systems with AI of the State.