Translation tutorial presented by: Corinne Cath-Speth, Mark Latonero, Vidushi Marda, Roya Pakzad
This tutorial discusses how human rights frameworks can inform, guide and govern AI policy and practice in a manner that is complementary to Fairness, Accountability, Transparency, and Ethics (FATE) frameworks.
Using the case study of researchers developing a facial recognition API at a tech company and its use by a law enforcement client, we will engage the audience to think through the benefits and challenges of applying human rights frameworks to AI system design and deployment. We will do so by providing a brief overview of the international human rights law, and various non-binding human rights frameworks in relation to our current discussions around FATE and then apply them to contemporary debates and case studies.
The development and deployment of facial recognition demonstrates how human rights frameworks can give AI designers, researchers, and policymakers, a holistic view of potential social impacts of such systems. It helps us identify how, as was the case for example in recent Hong Kong protests, the deployment of facial recognition impedes on the rights to privacy (Article 12 of the UDHR); the right to non-discrimination (Article 2), the right to freedom of expression and opinion (Article 19), the rights to assembly and association (Article 20), and so on. It also allows us to formulate the following avenues for recourse and redress of these violations, while also showing the limits of this framework.
By looking at international human rights standards, national frameworks, and localized applications, we hope to leave participants with a deeper understanding of how to integrate human rights frameworks in their technical development, policymaking and/or advocacy efforts. This tutorial also aims to answer calls within academia and civil society to move the debate beyond voluntary ethics frameworks and develop approaches based in law, that provide concrete obligations for industry.
Who will be delivering this tutorial?
We will spend the first thirty-minute of the tutorial to cover the following topics:
- A brief introduction to human rights framework and its history in relation to digital technologies; Human rights and ethical frameworks in AI: commonalities and differences.
- The international human rights law: AI regulations and states’ responsibilities
- Delivered by Vidushi Marda, Legal researcher and Program Officer at ARTICLE 19‘s Team Digital
- The UN Guiding Principles on Business and Human Rights: Private sector responsibilities
- Delivered by Roya Pakzad, Human rights and technology researcher; director and founder of Taraaz
The remaining 60 minutes will consist of an activity moderated by all four presenters.
Who is the target audience?
The tutorial is designed for the general ACM FAT* community:
Legal scholars and civil society advocates can benefit from this tutorial by applying human rights frameworks to governance of AI – not just as another voluntary ethical framework, but as a legal framework grounded in international human rights law. They can also learn how human rights frameworks can help them with stronger advocacy campaigns and regulatory proposals.
Machine learning researchers/practitioners can benefit from this tutorial by learning how to use human rights frameworks to proactively assess the risks of designing, developing, and deploying certain tools against society’s most vulnerable groups. They will learn how practices such as human rights impact assessments can complement and strengthen current practices in algorithmic impact assessments, value-sensitive and human-centric design.
Slides for this tutorial are available here.