IGF 2023 Lightning Talk #33 Framework for Meaningful Engagement in AI

5.1
5.2
5.5
5.b
5.c
10.2
10.3
10.6
16.7

Targets: SDG 5: gender equality– women and gender non-binary persons are disproportionately impacted by AI systems, from bias and discrimination in algorithms to silencing and harassment online. Women who have intersecting identity characteristics, such as racialized women; those from religious minorities; transwomen, queer and non binary persons; disabled women; girls, and those of lower socio-economic status, are furthermore disproportionately at risk of harm. These risks are especially acute for women from the Global South. Yet they’re also generally excluded from conversations related to the design, development, and use of AI systems. Building these systems in a way that considers the unique social, political and cultural contexts in which AI is created and used – often within patriarchal environments – is urgently needed. 5.1; 5.2; 5.5; 5B; 5C SDG 10: reduced inequalities – AI systems can accelerate and exacerbate existing social and economic inequality, from the use of AI for law enforcement and criminal justice, to automated social welfare systems and algorithmic content moderation. The issue is further heightened when looking at the disproportionate impacts and exclusion of Global South-based stakeholders. AI developers and deployers thus have a responsibility to identify, assess, mitigate, and remedy any adverse impacts that their systems may have on human rights. Human rights impact assessments for AI, with meaningful stakeholder engagement, helps promote marginalized and vulnerable users’ (and broader stakeholders’) enjoyment of human rights, instead of harming them, and reduce inequality between demographic groups and regions. 10.2; 10.3; 10.6; SDG16: 16. Peace, Justice and Strong Institutions . Promoting peaceful and inclusive societies through meaningful participation in HRIAs and ensuring responsive, inclusive, participatory and representative decision-making at all levels (16.7)