The impact of artificial intelligence on the protection of human rights in vulnerable contexts: risks and opportunities in the identification of the disappeared and border management

Abstract

This chapter examines how artificial intelligence (AI) affects the protection of human rights in transitional justice and migration management, highlighting both its potential and its risks. It analyzes the use of tools such as machine learning, facial recognition, and natural language processing to identify victims and perpetrators, document mass violations, and optimize humanitarian assistance. While these technologies can enhance efficiency and support the reconstruction of democratic memory, they also pose serious legal and ethical challenges: algorithmic bias, model opacity, data gaps that perpetuate exclusion and threats to due process and privacy. In the migration sphere, the automation of border control and asylum assessment reproduces structural inequalities and may undermine fundamental principles such as non-refoulement. The text warns of existing regulatory gaps and calls for AI governance grounded in human rights, based on transparency, independent auditing, and shared accountability among states, companies and humanitarian actors. It concludes that only through robust ethical and legal frameworks can AI contribute to justice, memory and the protection of human dignity, preventing technological efficiency from becoming a new form of exclusion or impunity.