
Sudan / Digital Systems / Interaction Design
Automating Bias
Statement of Relevance
Benjamin's analysis applies directly to my project. The content moderation systems that suppress Sudanese conflict documentation , removing footage as "graphic violence," deprioritising Arabic language testimony , operate exactly as she describes: not through explicit bias, but through default settings built on skewed training data. The system appears neutral while systematically erasing certain voices. Technology is never a neutral tool; it encodes the priorities and blind spots of whoever built it. To avoid replicating this in my own work, I will be explicit about whose perspectives I am centring, transparent about the limitations of my sources, and honest about my own position as an outsider designing around a conflict I am not inside. I will assess my project by its outcomes ,whose voices does it actually amplify , rather than by my intentions alone.
FIGURE 7 + FIGURE 8 : Assignment 4
FIGURE 9 + FIGURE 10: week 4 Independent Further Work
References:
Benjamin, R. (2019) Race After Technology: Abolitionist Tools for the New Jim Code. Cambridge: Polity.
Forensic Architecture (2021) Environmental Racism in Death Alley, Louisiana. Available at: forensic-architecture.org (Accessed: 5 May 2026).
Noble, S.U. (2018) Algorithms of Oppression. New York: NYU Press.
Reunion (n.d.) EJI | The Transatlantic Slave Trade. Equal Justice Initiative. Available at: reunion.xyz/tst (Accessed: 5 May 2026).
Canva link Below















