Many traditional risks and harms concerning AI—such as black box decision-making and bias—are foundationally a data problem. Similarly, many solutions to AI risks and harms—such as data security and accountability for decisions—rely on sound data practices for implementation. Data cannot be secured if they are not managed effectively. This chapter examines the intersection of civic technology and artificial intelligence (AI) governance, using the South African civic technology project Vulekamali as a case study. It highlights the significant contributions that civic technologists can make to the discussion on AI's future, drawing from their experience in social impact innovation. The chapter addresses the challenges of AI governance, including the pacing problem of law and the Collingridge Paradox. It proposes alternative solutions for entrenching responsible AI in contexts where governance lags behind technology development. The chapter concludes by stressing the importance of establishing a shared agenda for human flourishing in AI development and the critical role of the civic technology community in shaping future AI narratives.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Shaping a Responsible AI Future: Lessons from South Africa's Civic Technology Community

  • Gabriella Razzano

摘要

Many traditional risks and harms concerning AI—such as black box decision-making and bias—are foundationally a data problem. Similarly, many solutions to AI risks and harms—such as data security and accountability for decisions—rely on sound data practices for implementation. Data cannot be secured if they are not managed effectively. This chapter examines the intersection of civic technology and artificial intelligence (AI) governance, using the South African civic technology project Vulekamali as a case study. It highlights the significant contributions that civic technologists can make to the discussion on AI's future, drawing from their experience in social impact innovation. The chapter addresses the challenges of AI governance, including the pacing problem of law and the Collingridge Paradox. It proposes alternative solutions for entrenching responsible AI in contexts where governance lags behind technology development. The chapter concludes by stressing the importance of establishing a shared agenda for human flourishing in AI development and the critical role of the civic technology community in shaping future AI narratives.