Scholarlink

This chapter examines the development and application of AI in social services through stakeholder engagement within the AI FORA project, drawing on meetings held in May 2022 and 2023 with diverse participants, including marginalized communities, social workers, policymakers, technologists, and academics. Focusing on Spain as a case study, we identify key challenges and opportunities in AI-driven social service provision, emphasizing the importance of compositional (demographic and socioeconomic factors), contextual (local systems and biases), and collective (community trust and participation) dimensions. Our findings reveal that AI risks exacerbating inequities when these factors are overlooked. However, stakeholders pointed out that context-aware AI applications, designed with adaptability, transparency, and participatory oversight can mitigate these risks. The chapter highlights three key policy lessons: (1) AI systems must evolve through continuous auditing and community input, (2) predictive tools should account for local realities to avoid bias, and (3) participatory governance is essential to ensure equity.

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

Policy Learnings and Policy Change for AI-Based Social Services

  • 章节

  • 出版模式：

    开放获取

  • 发表日期：

    2026年

查看全文
  • Albert Sabater,
  • Beatriz López,
  • Roger Campdepadrós

摘要

This chapter examines the development and application of AI in social services through stakeholder engagement within the AI FORA project, drawing on meetings held in May 2022 and 2023 with diverse participants, including marginalized communities, social workers, policymakers, technologists, and academics. Focusing on Spain as a case study, we identify key challenges and opportunities in AI-driven social service provision, emphasizing the importance of compositional (demographic and socioeconomic factors), contextual (local systems and biases), and collective (community trust and participation) dimensions. Our findings reveal that AI risks exacerbating inequities when these factors are overlooked. However, stakeholders pointed out that context-aware AI applications, designed with adaptability, transparency, and participatory oversight can mitigate these risks. The chapter highlights three key policy lessons: (1) AI systems must evolve through continuous auditing and community input, (2) predictive tools should account for local realities to avoid bias, and (3) participatory governance is essential to ensure equity.