Beyond the EU’s Cyber-Fortress: Decolonial Perspectives on AI and Migration

Abstract

Beyond the EU’s Cyber-Fortress: Decolonial Perspectives on AI and Migration AI is often framed as objective and neutral, but its application in social and political contexts reproduces harmful and discriminatory outcomes. By formalizing complex social issues into technical problems, algorithmic systems obscure the inherently political nature of tasks such as predicting ‘acceptable’ and ‘risky’ bodies. The management of human migration in the European Union exemplifies these dynamics. Indeed, migration is often framed as a security issue, portraying migrant people as threats linked to disease, crime, or risk, justifying expansive surveillance measures and predictive technologies that reinforce racial, gendered, and class-based inequalities. The EU’s move towards a ‘Cyber-fortress’ underscores a troubling detachment between technological systems and the human realities they govern. These systems often obscure the human realities behind migration, reinforcing discriminatory practices and ignoring systemic causes like colonialism, economic disparity, and global power imbalances. Moreover, prioritizing border control efficiency over human dignity has devastating impacts, dehumanizing migrant people and overlooking the inequalities embedded in these systems. Resisting this trajectory demands a critical rethinking of the ethical foundations of migration management and the technological apparatuses that sustain it. This involves moving beyond Western-centric and institutional paradigms toward decolonial approaches that center the voices of migrant people, those most impacted yet least represented in technological development and security research. It reclaims also a technology that must be removed from the exclusive domain of the powers that be and returned instead to democratic action. Even more urgently, it pushes us not to accept the normalization of the exceptional and to search, among data and algorithms, for what remains of the human.