E. Micheni

Governing Artificial Intelligence in a Fragmented World: Toward a Multi-Level Policy Framework for Global AI Governance

Artificial intelligence has moved from a specialised technical concern to a central object of international economic and security policy, yet global governance arrangements remain fragmented across competing regulatory models. This article addresses how policymakers can reconcile innovation, competitiveness, human rights, security and sustainability within a coherent governance architecture for artificial intelligence operating across national, regional and multilateral levels. The goals include the review of the governance theory and current practices applicable to AI regulations, analysis of stakeholders' interests and influence, estimation of the possible economic, social, legal, technological and environmental effects of such an arrangement, as well as the design of a feasible multi-level governance system with monitoring and evaluation mechanisms. The article utilises a qualitative comparative policy analysis based on primary legal documents, which include Regulation (EU) 2024/1689, OECD Recommendation on Artificial Intelligence (as amended in 2024), UN Resolution A/RES/79/325 of 2025, and the Global Digital Compact of 2024, in addition to the academic literature on the subject from peer-reviewed sources and books. This paper proposes a framework that takes into account the multi-level and adaptive governance approaches with a particular emphasis on digital sovereignty, thereby creating a three-tier architecture that will include global normative coordination, regional and plurilateral regulatory clusters, and national or sectoral implementation, illustrated by the case study of Kenya, which has created its National Artificial Intelligence Strategy 2025-2030. The main findings in the paper suggest that regulation fragmentation among the European Union, the United States, and China is growing rather than converging, that multilateral instruments do not possess any kind of binding enforcement mechanism, and that low- and middle-income countries like Kenya experience capacity limitations even when developing a proactive national strategy. This paper argues that a layered subsidiarity approach, with specific financing for capacity-building and technical standards that work across the board, is more likely to deliver effective global AI governance than calls for a binding treaty.

June 2026: Top 10 Cited Articles in Software Engineering & Applications

Purchasing Practices and Food Security among African Indigenous Vegetable Farmers in Nyanza Region, Kenya

Food insecurity remains a major concern among smallholder farming households in Kenya despite continued efforts to increase agricultural productivity and improve rural livelihoods. Previous studies have mainly linked household food security to factors such as farm production, income levels, access to extension services, land ownership, and the adoption of improved technologies. However, little attention has been given to the role of farmers’ purchasing decisions and input sourcing practices. In addition, theories widely used in purchasing and supply chain management, including Value Chain Theory, Transaction Cost Theory, the Relational View, Social Embeddedness Theory, and Institutional Theory, have rarely been applied to explain food security outcomes among smallholder farmers. This study addresses this gap by examining how purchasing practices influence food security among African Indigenous Vegetable (AIV) farmers in the Nyanza region of Kenya. A cross-sectional correlational research design was employed, and primary data were collected from 430 AIV farmers selected through stratified random sampling across six counties. Data were analysed using descriptive statistics, Pearson correlation, and multiple regression techniques. The findings showed that purchasing practices were positively and significantly associated with household food security (R² = 0.326, F (8,421) = 25.43, p < 0.001). Collaboration with input suppliers and other value chain actors had the strongest positive association with food security (β = 0.237, p < 0.001), followed by aligning purchases with consumer preferences (β = 0.205, p < 0.001) and input supplier selection (β = 0.192, p < 0.001). Household income, farming experience, and monthly sales also had significant positive relationships with food security. This study contributes to the literature by integrating insights from food security and supply chain management and by recognising smallholder farmers as active purchasing decision-makers whose sourcing choices influence production, market participation, and household well-being. The findings highlight the importance of strong relationships within indigenous vegetable value chains in improving food security outcomes. Strengthening partnerships between farmers and input suppliers, promoting collaborative marketing, improving access to market information, and investing in aggregation centres and cold storage facilities could enhance the efficiency and resilience of indigenous vegetable markets. Farmers can further improve household food security by building long-term supplier relationships, responding to consumer preferences, and adopting more strategic purchasing practices. Key words: African Indigenous Vegetable Farmers, Food Security. Purchasing Practices, Kenya