From Apartheid to AI: The Challenges of Reconstructing Transnational Advocacy Networks

Jeanette Ruiz, Virginia Commonwealth University

Artificial intelligence (AI) is revolutionizing historical research, yet its limitations distort how we understand transnational advocacy networks (TANs). AI-driven archival analysis, network mapping, and sentiment tracking struggle to capture informal mobilization, moral framing, and historical bias—essential elements in advocacy movements. This paper compares the Anti-Apartheid Movement (AAM) and Fridays for Future (FFF) to reveal a fundamental divide: FFF actively integrates AI-generated personas and synthetic media into its activism, while AAM relied solely on human networks and informal organizing. AI tools like Natural Language Processing (NLP) and sentiment analysis privilege structured, digitized records while failing to reconstruct the adaptive, decentralized, and often deliberately hidden structures of activist networks. Many advocacy groups obscure, disguise, or encode their communications to evade surveillance, repression, or co-optation, making their informal organizing difficult for AI to trace. This contrast highlights a growing historical asymmetry: AI-powered movements like FFF may be overrepresented in digital archives, while human-led historical movements like AAM risk being misinterpreted or diminished. By critically assessing AI’s blind spots in reconstructing historical advocacy, this paper argues for an interdisciplinary approach that integrates AI with archival research, oral histories, and ethnography to prevent algorithmic distortions of global movements.

See paper

 Presented in Session 217. Beyond Models and Archives: Reconstructing Historical Complexity through Methodological Innovations