Dr. Reham Hosny
University of Cambridge
Minia University
It was a great honor to be invited by the UN Tech Envoy, Amandeep Gill, earlier this year to contribute to the global discussion on AI. As a member of the consultative network advising the UN Secretary-General’s High-level Advisory Body on AI, I had the opportunity to engage in this critical conversation.
The discussions were far-reaching, with extensive consultations on the AI interim report. Over 500 experts participated in 18 “deep dives” on specific issues, accompanied by 250+ written submissions from more than 150 organizations and 100 individuals. An AI risk pulse check further engaged around 350 experts worldwide.
Last month, the UN AI Advisory Body released its final report, “Governing AI for Humanity”, which includes seven key recommendations aimed at bridging governance gaps and fostering global collaboration [link].
One of the most striking challenges highlighted is the representation gap. Out of 193 Member States, only seven are engaged in prominent AI governance initiatives, while 118—primarily from the Global South—remain entirely absent.
Despite the importance of global efforts such as the UN and the EU AI ACT, they often miss a crucial element: shifting power to the powerless. The underlying power dynamics of AI technologies need more attention. Even when biases and challenges are identified, there’s no guarantee that those in control will address them (as noted by scholars such as Agnew et al., 2023).
Effective AI governance must prioritize marginalized and vulnerable communities, who are most affected by AI development and misuse. The imbalance between the Global North—home to tech corporations and powerful governments—and the Global South, where AI misuse causes the most harm, must be urgently corrected.
In the MENA region, particularly the Arab World, where AI is increasingly embedded in state and military operations, the need for ethical AI governance has never been more pressing. AI is often used by governments, tech corporations, and occupying forces to stifle free speech, enforce censorship, conduct surveillance, and perpetuate conflict.
To engage meaningfully in global AI governance, the Arab World should first develop a framework that reflects its unique cultural, social, and political values while prioritizing those most impacted by AI misuse. This framework will serve as a foundation for meaningful participation in the global conversation on AI regulation.
Existing regional efforts, including Egypt’s Charter for Responsible AI by Ministry of Communications and Information Technology (MCIT), UAE’s Principles of AI by Digital Dubai, KSA’s AI Ethics Principles by Saudi Data & AI Authority (SDAIA), Qatar’s Guidelines for
Secure Adoption and Usage of Artificial Intelligence by National Cyber Security Agency (NCSA), Oman’s draft National Artificial Intelligence Policy by Oman’s Ministry of Transport‚ Communications‚ and Information Technology, should be unified and re-evaluated to ensure a cohesive and comprehensive approach to AI governance across the region.
Globally, AI development and governance have often excluded groups such as creatives, activists, women, refugees, and dissidents. Correcting this imbalance requires greater inclusivity and participation from these underrepresented sectors.
A bottom-up regulatory approach is essential—one that empowers marginalized communities through knowledge-building, resource investment, and active participation in decision-making. Governments, stakeholders, and society must focus on shifting power to the powerless to ensure justice and equity.
By actively shaping AI’s role in shifting power dynamics, we can transform it from a tool of control into one of justice and freedom. The future of AI in the Arab World—and globally—is at a crossroads. Whether it serves as a tool for liberation or oppression will depend on the frameworks we build today.