By: Nur El-Mahrakawy

The “Responsible AI in the Global South” webinar series co-hosted by the MENA Observatory on Responsible AI and the African Observatory on Responsible AI represents a critical milestone in the ongoing global conversation about artificial intelligence. This webinar series aims to create a dedicated space for knowledge-sharing and the exploration of responsible AI, not just in the Middle East and Africa, but across the entire Global South. The webinar series highlights Middle Eastern and African perspectives, focusing on nuanced and context-specific approaches to AI governance. By amplifying the voices of local knowledge holders, addressing crucial topics such as ethical considerations, regulatory frameworks, and societal impacts of AI technologies in these regions, we can ensure that communities who have been historically marginalized are no longer left out of the conversation. Our first webinar, Responsible AI in the Global South: Narratives from the Middle East and Africa, marked the inaugural session in our collaborative series. On October 30th, the MENA Observatory on Responsible AI and the African Observatory on responsible AI co-hosted the second webinar in the series, “Multi-Stakeholder Collaboration for Responsible AI Development in Africa.” The event focused on the African Union AI Strategy and explored multi-stakeholder collaborations for the responsible development and implementation of AI across the continent. 

Dr. Nagla Rizk, founding director at the Access to Knowledge for Development Center (A2K4D), moderated the webinar. The session delved into strategies for developing AI frameworks that are both ethical and regionally relevant, with an emphasis of collaboration. The webinar featured a diverse panel of experts, each offering valuable insights into the intersection of AI and development in Africa and the MENA region included: Mr. Amr Safwat, Manager of the African Organizations Affairs in the International Relations Division - Ministry of Communications and Information, Ms. Isobel Acquah, Executive Director at Certa Foundation,  Ms. Leonida Mutuku, Director of AI at LDRI, and Mr. Silas Adekunle Co-Founder and CEO at Reach Industries.

Re-molding AI to Fit the Need of Africa

Mr. Adekunle provided an in-depth overview of the challenges facing AI adoption in MENA and Africa. He described AI adoption as a “multidimensional problem,” emphasizing that current AI systems were not designed with the realities of Africa in mind. For example, African languages, often rich in symbolism and tonality, present significant barriers for AI models built on tokenization. Furthermore, much of Africa’s history and culture is communicated visually or through storytelling - elements that current AI systems are poorly equipped to capture. Finally, the uneven rate of technology adoption across the continent has led to inconsistent and fragmented data, which contributes to biases in AI systems, ultimately diminishing their effectiveness in African contexts and missing out on opportunities to capitalize on critical native knowledge.

Building on this argument, Mr. Safwat underscored the necessity for cross-border collaboration. Given AI’s inherently transnational nature, collaboration is not just beneficial, but essential for its ethical development. He stressed that African voices must be at the forefront of the dialogue on ethical AI and governance, ensuring AI solutions are tailored to meet the continent’s unique socio-economic challenges in healthcare, education, public service, and peace and security. His call to action was clear: academics, governments, the private sector, and grassroots organizations must work together to build the necessary frameworks for responsible AI.

Sustainability and Inclusivity

Ms. Acquah took the discussion in a critical direction by highlighting the environmental impact of AI and the need to work towards “green transition.” She drew attention to the extensive amount of energy required to train algorithms and the growing concern over e-waste. Acquah also highlighted the importance of engaging youth in AI design, pointing out that the continent’s young, creative population has the potential to create a legacy of responsible AI. However, she also noted that AI research in Africa remains severely underfunded and underdeveloped, a gap that needs to be addressed. She called for academics to prioritize applied research to address real-world challenges.

Finally, Mutuku wrapped up the webinar by emphasizing the importance of sustainable business models for scaling responsible AI. She pointed out that much of the funding for AI development on the continent currently usually supports pilots and proof-of-concepts, with limited support for full-scale implementation. LDRI, she shared, is working with policymakers and innovative communities across Africa “to explore how communities of practice can drive scalable and sustainable AI solutions.” She argued that “communities of practice can strengthen the open-source movement by focusing on specific community needs, creating opportunities for a unified voice in GPT-5 advocacy.” She concluded the webinar with a call to develop favorable regulatory frameworks to ensure AI benefits everyone.

The Path Forward

The conversation from the webinar makes one thing clear: the future of AI in the Global South must be shaped through collaboration. Policymakers, tech developers, academics, and civil society must work together to build AI systems that fit the diverse needs of the Global South. As AI becomes an integral part of societies, it is more than ever that the voices of all local stakeholders are included.