Multilateralism

This piece is part of Reframing Impact, a collaboration between AI Now Institute, Aapti Institute, and The Maybe. In this series we bring together a wide network of advocates, builders, and thinkers from around the world to draw attention to the limitations of the current discourse around AI, and to forge the conversations we want to have.

In the run-up to the 2026 India AI Impact Summit, each piece addresses a field-defining topic in AI and governance. Composed of interview excerpts, the pieces are organized around a frame (analysis and critique of dominant narratives) and a reframe (provocations toward alternative, people-centered futures).

Chinasa T. Okolo is the founder of Technecultura and a policy specialist at the United Nations Office for Digital and Emerging Technologies (ODET). Her research examines how African governments can effectuate robust AI and data governance, investigates the geopolitical impacts of AI, and analyzes datafication and algorithmic marginalization in Africa. 

In this conversation, Okolo unpacks the current state of AI global governance, and Big Tech’s dominance, from a Majority World perspective. At a moment when Big Tech is leading with “AI for good” and “AI for development” hype to access the consumers and data of the Majority World, Okolo pushes for a nuanced conversation to understand the real potentials of AI in light of entrenched structural issues. She cautions against corporate partnerships and makes a strong case for richer peer exchanges between Majority World countries as the way forward for feasible solutions for local contexts and communities.

Following is a lightly edited transcript of the conversation.

Big Tech’s influence in Majority World AI policy is concerning.

Whenever I participate in policy spaces, there is always someone from Anthropic, Google, Meta, OpenAI, Hugging Face. In these spaces, I’m very wary of these companies. Meta, for example, focuses their policy agenda on their open-weight—not open-source—LLMs, while avoiding many of the big questions around algorithmic harms on their social media and communication platforms, such as WhatsApp. They have also been involved in countries, particularly within the African continent, like Nigeria, in shaping and participating in drafting national AI strategies. They are influencing how governments regulate AI, for example by pushing for data privacy rules to be weakened against Big Tech. This increases my concern about the influence that Big Tech will have in India and within or across other Global Majority countries.

Majority World AI strategies face a choice between grand hype driven by fickle outside actors and more grounded, feasible solutions.

I am aware of the hype surrounding AI for development, particularly in the Indian context. I did my dissertation research on AI in rural India for community healthcare workers. As someone who is now working at the UN, I understand the diplomacy and the multilateral side of it as well. I hope the Summit shifts the narrative towards feasible solutions, away from grand ideas that would not be implemented and that may shift if priorities from prominent governments and frontier tech companies change.

REFRAME:  We need detailed research and multilateral dialogue—not Big Tech hype and short-term corporate partnerships—to build robust, contextually relevant systems that can deliver lasting developmental change.

The first question is: What are the long-term costs, not just the financial ones? A big company may approach a Global Majority government with the promise of reducing costs for the first couple of years. But as the model and usage scales, it may become unsustainable to afford in the long run. They will have wasted all this money on an AI system that they can’t even use anymore, money that could have been directed towards pressing social problems. I think governments really have to employ analysts or researchers to figure out how such issues may look in the long run.

We also need to understand what the agreements are. For example, there was an issue around Kenya having a health data partnership with the United States which would provide the US with access to sensitive Kenyan health data for decades. That’s unreasonable. Many other governments across Africa, across the Majority World, may experience similar clauses in their partnerships with companies. These companies are finally seeing the majority as both a new market to increase revenues and to provide new avenues for data to train and refine their models to become more culturally and contextually relevant. They may introduce clauses that provide them with access to sensitive data or general usage data about consumers, clauses that governments may not be aware of because they don’t have the technical capabilities or expertise to understand. This is something I really want governments pursuing these partnerships to be aware of.

South-South dialogue and peer exchanges are essential to understanding how AI solutions can be used, developed, and adopted to address local problems.

Governments, particularly those in Global Majority countries, need to understand how AI solutions are already being used, developed, and adopted elsewhere and figure out what issues they raise. For example, AI Incident databases don’t provide a lot of context on issues outside of the US, the UK, the EU, and other Western countries. So [it is important] to collaborate with civil society and academia to do landscape mapping to really understand: Is AI helping solve these problems or is it just creating more of them?

Another way to get that information is to talk to other peer countries at big summits, UN forums, or events to understand where AI could potentially be a solution, what approaches peer countries have taken, and what issues they are experiencing. For example, if they are focusing on small language models versus trying to develop a large national LLM sponsored by OpenAI, what has been the value in that approach? What benefit did it bring to the government? Has it enabled you to streamline social services distribution? Has it closed a communication gap with minority communities in your country? 

Having these very intricate conversations and trying to understand the landscape and problems is something that politicians don’t do a lot. They’re usually more focused on trying to maximize partnerships and to understand how they can get funding. Obviously, you need money to pursue these projects, but you have to be realistic. I’m adamant that African governments in particular, as well as others across the Global Majority, have to be very intentional about not just pursuing lofty goals but ensuring that if you want to adopt AI, it’s reasonable and feasible for your context and for your communities.

Watch the full conversation between Chinasa T. Okolo and Alix Dunn here.