Scholarlink

<p>Digital technologies are seen as a&#xa0;driver of social innovation. At the same time, however, there is a&#xa0;risk that they will technically reinforce existing social inequalities not only technically, but also socially, economically, and culturally. A&#xa0;key example of this is the gender data gap, the systematic underrepresentation and distortion of gender-related data in the development and application of artificial intelligence. The article analyzes the emergence of digital invisibility throughout the life cycle of AI systems, as well as the structural mechanisms that lead to gender-specific exclusions. These mechanisms include binary acquisition logics, stereotypical modeling, and a&#xa0;lack of representation in system design. The focus is on the question of how AI systems can be developed to enable gender-equitable participation. It is argued that technological solutions alone are not enough. A&#xa0;change of perspective is needed in design, data practice and institutional responsibility. The healthcare sector serves as an exemplary field of application to illustrate the effects of digital invisibility.</p><p>On this basis, the article develops practice-oriented impulses for a&#xa0;gender-sensitive technology design. Four guiding principles—reflexivity, representation, participation and transparency—form an orientation framework for the design of inclusive AI systems. The article is intended as a&#xa0;theoretically grounded impulse for research and practice and invites the further development of gender-equitable technologies, including through inclusive co-design processes and participatory governance models.</p>

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

Der Gender-Datengap in Künstlicher Intelligenz: Digitale Unsichtbarkeit und ihre Folgen für Inklusion und Teilhabe

  • 期刊论文

  • 发表日期：

    2025年10月7日

查看全文
  • Laura Steffny,
  • Daniela Podevin,
  • Nanna Dahlem,
  • Tobias Greff

摘要

Digital technologies are seen as a driver of social innovation. At the same time, however, there is a risk that they will technically reinforce existing social inequalities not only technically, but also socially, economically, and culturally. A key example of this is the gender data gap, the systematic underrepresentation and distortion of gender-related data in the development and application of artificial intelligence. The article analyzes the emergence of digital invisibility throughout the life cycle of AI systems, as well as the structural mechanisms that lead to gender-specific exclusions. These mechanisms include binary acquisition logics, stereotypical modeling, and a lack of representation in system design. The focus is on the question of how AI systems can be developed to enable gender-equitable participation. It is argued that technological solutions alone are not enough. A change of perspective is needed in design, data practice and institutional responsibility. The healthcare sector serves as an exemplary field of application to illustrate the effects of digital invisibility.

On this basis, the article develops practice-oriented impulses for a gender-sensitive technology design. Four guiding principles—reflexivity, representation, participation and transparency—form an orientation framework for the design of inclusive AI systems. The article is intended as a theoretically grounded impulse for research and practice and invites the further development of gender-equitable technologies, including through inclusive co-design processes and participatory governance models.