As Artificial Intelligence (AI) rapidly reshapes knowledge systems, governance structures, and global development pathways, a fundamental question arises: Who owns the algorithms shaping our collective future? To date, AI development has been largely concentrated in the hands of a few powerful corporations and countries—predominantly in the Global North. Yet a growing chorus of voices, particularly from the Global South, is calling for a multilateral, rights-based approach that centres public interest, equity, and transparency.
The urgency for such a shift is gaining momentum at the highest levels of global governance. In a landmark move, the United Nations General Assembly (UNGA) adopted the “Pact for the Future,” which includes the Global Digital Compact (GDC)—a framework aimed at fostering inclusive, ethical, and accountable digital cooperation. Further, UNGA Resolution 79/1 calls for the establishment of a multidisciplinary Independent International Scientific Panel on AI and the launch of a Global Dialogue on AI Governance—both major steps toward reimagining AI as a global public good rather than a proprietary asset.
At present, the development of Large Language Models (LLMs) and other advanced AI systems remains overwhelmingly concentrated in the Global North. These models are trained, governed, and deployed within a limited set of cultural, political, and economic contexts, often reinforcing longstanding inequalities. Meanwhile, countries in the Global South are left digitally dependent, with minimal influence over how AI technologies are designed or applied. This growing asymmetry threatens to transform AI into a new form of technological colonialism—where marginalized communities become passive recipients of tools that neither reflect their realities nor respond to their needs.
Ironically, much of the data used to train these AI systems originates from the Global South—whether through social media interactions, environmental monitoring, or public infrastructure. And yet, the regions generating this data often see little of the resulting value. If AI were truly localized—designed with the linguistic, cultural, and developmental contexts of the Global South in mind—it could become a catalyst for equity, sustainability, and self-determination.
That’s the vision behind Poseidon-AI, founded not in Silicon Valley, but in Singapore—at the heart of Asia’s aquaculture and coastal food systems. Poseidon-AI integrates AI, machine learning, and Internet of Things (IoT) technologies to support climate-resilient aquaculture and sustainable agriculture—especially in Southeast Asia and China, where nearly 90% of the world’s seafood is produced. In these regions, rising temperatures threaten the survival of poikilothermic (cold-blooded) aquatic species, while overuse of antibiotics, contamination from microplastics, and declining water quality are escalating concerns. Poseidon-AI works directly with local producers, communities, and governments to deploy AI tools tailored to real-world challenges—from early disease detection to water monitoring and biodiversity conservation—ensuring that technology works for the people and ecosystems who need it most.
However, the true impact of Poseidon-AI’s technology emerged during the COVID-19 pandemic, when global supply chains were disrupted, quarantine measures were severe, and marginalized communities were often left behind. In response, Poseidon-AI deployed its Integrated Aquaculture Systems (IAS) to support Indigenous and vulnerable communities in Costa Rica and assisted in sustainable rice-fish farming practices in Ghana. These systems utilized rainwater harvesting, clay tiles, solar energy, and AI-powered monitoring algorithms to help communities grow their own food sustainably—and even generate income from surplus production. Despite challenges such as linguistic barriers, traditional dietary customs, limited access to clean water and energy, and patriarchal social structures, Poseidon-AI’s tailored, community-centred approach helped overcome these obstacles using innovative, culturally sensitive technology.
Support from various institutions made this possible, including the United Nations Development Programme (UNDP), the Food and Agriculture Organization of the United Nations (FAO), the Global Environment Facility (GEF), and national partners such as Banco Nacional. Perhaps the most symbolic moment came when U.N. Deputy Secretary-General Amina Mohammed visited the Indigenous territory of Talamanca, Costa Rica. With the help of a translator, she spoke with local women and girls—some of whom, for the first time in their lives, were growing tomatoes using only rainwater, clay tile greenhouses, and AI-assisted techniques. It was a defining moment for Poseidon-AI, and a powerful example of how inclusive innovation can empower even the most remote and underserved communities.
Years later, the momentum is growing. U.N. agencies are already charting a path forward: the United Nations Development Programme (UNDP) supports Digital Public Infrastructure (DPI) that places open-source, public-interest digital systems at the centre of governance and service delivery. The United Nations Educational, Scientific and Cultural Organization (UNESCO) has advanced this vision with its Recommendation on the Ethics of Artificial Intelligence, adopted by over 190 member states, establishing the first global framework that emphasizes human rights, inclusivity, and cross-cultural dialogue. The United Nations Institute for Training and Research (UNITAR), for its part, is expanding AI training and governance literacy among policymakers and civil society actors in Small Island Developing States (SIDS), Least Developed Countries (LDCs), and conflict-affected zones—precisely where AI inclusion can have transformative impacts.
The International Telecommunication Union (ITU) is also playing a pivotal role by convening global stakeholders to develop AI standards, foster interoperability, and ensure that telecommunications infrastructure—a prerequisite for AI deployment—is accessible, secure, and inclusive. Through initiatives like AI for Good, ITU connects innovators, policymakers, and communities—particularly in developing countries—to ensure that AI technologies are harnessed for sustainable development and real-world impact. Its work complements the efforts of UNDP, UNESCO, and UNITAR by providing the digital backbone necessary to support AI equity across geographies.
To unlock AI’s true potential for humanity, we must move toward an AI Commons: a globally shared ecosystem of open, ethical, and inclusive AI resources governed in the public interest. Just as we safeguard clean air, public health, and scientific research, AI—particularly in its applications to education, agriculture, conservation, and public services—should be treated as a common good. This means prioritizing open-source models, investing in culturally and linguistically diverse datasets, and developing participatory governance structures that empower communities—not just corporations—to shape the AI systems that influence their futures.
This is not a call to slow down innovation, but a demand to democratize it. If guided by justice and inclusion, AI can help bridge—not widen—the divides of our world. And as the story of Poseidon-AI demonstrates, when innovation is grounded in equity, sustainability, and community leadership, even the most advanced technologies can uplift humanity’s most overlooked voices.
Amaj Rahimi-Midani is a Costa Rican-Iranian environmental scientist & founder of Poseidon-AI. He specializes in sustainable soil & water management, aquaculture, & ocean conservation, collaborating with organizations like the ITU, FAO, UNDP, & UNITAR on climate resilience & food security. An advocate for equity in AI, he bridges technology & policy to promote global sustainability & innovative solutions.