FemAI Encounters: South Africa Lab: A Reflection on Human Ethics in the Age of Artificial Intelligence
Written by: Aqeelah A. Bray
The FemAI Encounters: South Africa Lab brought together Commissioner Mfundo Nomvungu and a dynamic collective of academics, policymakers, activists, and technologists to explore what ethical AI could mean for gender justice. Discussions examined AI’s growing influence on law, policy, and governance, highlighting its dual capacity to promote inclusion while deepening inequality. Grounded in South Africa’s constitutional and policy frameworks, the event connected AI ethics to themes of equality, privacy, and accountability.
A powerful GBV Policy session linked the National Strategic Plan on GBVF (2020–2030) to digital ethics, emphasizing the importance of protecting survivors and rejecting the commodification of their experiences. The GRIT team presented innovations that centresurvivor agency and safety. The lab concluded with a call for community-owned data systems and a deeper commitment to data and digital sovereignty rooted in local contexts. Moreover, the Lab’s format, which was deeply collaborative and co-creative in its knowledge buildingallowed for discussions that were layered and complex, speaking to AI’s relationship with justice, privacy, gender, and access.
However, while the sessions examined gender bias, data sovereignty, and community ownership, there was little interrogation of what it means for AI to handle survivors’ data. Arguably, pathologizing and reducing lived experience that carries trauma, vulnerability, and agency to data is in and of itself telling of the ways survivors are seen in a system that values statistics over people. Moreover, the question of who holds that data, who benefits from it, and how it could be used against survivors remains deeply unresolved. Moreover, data centres are primarily controlled by the global north, further entrenching colonial power’s control over Africa’s digital sovereignty. As we rush to digitise justice, we must ask whether the tools we design are capable of holding such sacred, painful truths with the dignity and safety they demand.
Equally absent was a deeper conversation about the environmental impact of AI. A request made through ChatGPT consumes 10 times the electricity as other search engines. The UN Environment Programme (2025) reports that globally, the infrastructure required for AI may soon consume six times more water than Denmark, a country of 6 million, when a quarter of humanity already lacks access to clean water and sanitation. Additionally, the microchips that power AI require rare minerals which are mined in environmentally destructive ways and produce hazardous toxic waste (UN Environment Programme, 2025). The same technology that promises progress also consumes vast ecological resources, often extracted from the Global South. Eco-feminist viewpoints require that we see the full picture: social, spiritual, and environmental justice are intertwined, and we cannot attain gender justice without environmental justice.
References:
United Nations Environment Programme. (2024, September 21). AI has an environmental problem. Here’s what the world can do about that. UN Environment Programme. https://www.unep.org/news-and-stories/story/ai-has-environmental-problem-heres-what-world-can-do-about