AI for Disability Inclusion in Africa
This report summarises the opportunities, challenges and enablers for disability-inclusive AI in Africa and key research questions that were identified in three online roundtables held in February, March and April 2024. The roundtables were convened and hosted by UCL’s Global Disability Innovation Hub (GDI Hub) at the request of the Foreign, Commonwealth and Development Office (FCDO).
The roundtables brought together AI, disability, and development experts to share their knowledge and experiences.
Roundtable #1 was held on 8 February 2024. Theme: AI and Disability inclusion in Africa. The discussion focused on examples of African AI-powered assisted technologies, the potential and opportunities for AI to enhance the lives of disabled people in Africa, and the risks it may pose.
Roundtable #2 was held on 7th March 2024. Theme: How do we ensure AI increases access to justice and inclusion for disabled people? A legal and policy perspective. This focused on the challenges relating to digital access, inequalities, social inclusion, and the role of governance and policy in guiding the development of ethical and inclusive AI that helps to address these issues rather than compounding intersectional disadvantages.
Roundtable #3 was held on 11th April 2024. Theme: Innovation perspective on AI for development and Disability inclusion in Africa. This focused on the key strategic issues and challenges that need to be addressed to enable the AI sector in Africa to grow in a way that will help to realise its potential to enhance the lives of disabled people in Africa and address the risks.
The discussions underscored the need for AI to be developed ethically in ways that respect and include the rights of disabled people in Africa and the challenges to achieving this, including ongoing development challenges that are not specific to AI but remain pertinent if AI is to fulfil its potential as a tool for good, and not compound existing disadvantages and inequalities.
Opportunities identified include the potential to increase access to life-changing AT through investment in local development of AI-driven AT, alongside improved digital infrastructure, affordable devices and data-plans; and mainstreaming the inclusion of disabled people by ensuring that all AI is developed in a way that includes the rights of disabled people and doesn't diminish or exclude them.
Key challenges include unreliable Internet and electricity supplies that hinder the development and use of all digital technologies in many African countries; lack of access to the education and training to develop digital and AI skills; biased data-sets that under- and mis-represent all marginalised people being used for training AI; lack of effective data and technology governance strategies; and lack of financial investment in African AI and AT entrepreneurs leading to a ‘brain drain’ of talent from Africa.
Important enablers of inclusive and ethical AI in Africa include: technology infrastructure; education and training; public investment in generating representative datasets; appropriate governance of AI developed with the involvement of all legitimate actors, including citizens; policy alignment across all domains to support the delivery of ethical AI; and mechanisms to increase financial investment in African digital ventures and entrepreneurs who are committed to disability inclusion.