The Global South and the AI Battleground: Who’s Really Calling the Shots?

The Global South and the AI Battleground: Who’s Really Calling the Shots?

Whenever we usually start to look into the debate of who’s really shaping the future of AI, we keep hearing the same story about it being a two-way struggle between the US and China. But the more dug into it, I realized that there’s this whole other narrative bubbling up where some experts also believe that the Global South, those emerging economies in Africa, Latin America, and Southeast Asia, are actually the ones who’ll decide how this all plays out. It’s a compelling argument, but like most things in geopolitics, the reality is way more complicated than the headlines suggest.

Let me start with what got me interested in this perspective. The numbers are pretty striking in this context as the Global South represents about 85% of the world’s population. That’s a massive chunk of humanity, and they’re not just sitting around waiting for Silicon Valley or Shenzhen to tell them what to do with AI. The world is witnessing key contributions from the Global South. For instance, Brazil helped lead UNESCO’s AI ethics guidelines, pushing hard for data sovereignty. Indonesia built its own AI roadmap focused on transforming work and building smart cities. Kenya positioned itself as a leader in AI-driven financial inclusion. These aren’t countries following someone else’s playbook, in some or other way they’re writing their own rules.

Let me start with what got me interested in this perspective. The numbers are pretty striking in this context as the Global South represents about 85% of the world’s population. That’s a massive chunk of humanity, and they’re not just sitting around waiting for Silicon Valley or Shenzhen to tell them what to do with AI. The world is witnessing key contributions from the Global South. For instance, Brazil helped lead UNESCO’s AI ethics guidelines, pushing hard for data sovereignty. Indonesia built its own AI roadmap focused on transforming work and building smart cities. Kenya positioned itself as a leader in AI-driven financial inclusion. These aren’t countries following someone else’s playbook, in some or other way they’re writing their own rules.

What really caught my attention was the data story which is astonishingly surprising. Interestingly, these regions are providing the training grounds that make AI actually work better everywhere. For instance, Google’s tuberculosis detection system got way more accurate when it learned from African medical data. Similarly, Climate prediction models got sharper using Latin American agricultural information and Language processing improved dramatically when it started learning Southeast Asian languages. Thus, it demonstrates that for the countries in the Global South it’s not just about consuming AI but also about contributing to making it smarter.

With respect to India, in particular, it has positioned itself as a bridge between the developed world and emerging economies. When Prime Minister Modi co-chaired the Paris AI Action Summit in February 2025 with President Macron, it wasn’t just ceremonial, it showed how India is being recognized as a legitimate voice in global AI governance. Looking at the bigger picture, I think India exemplifies the complexity of the Global South’s role in AI geopolitics. Therefore, when India advocates for “tech neutrality” and open-source approaches, it’s offering a genuine third way between American corporate dominance and Chinese state control.

Though all these arguments suggest a rosy picture about the role of Global South in AI, here’s where I started seeing cracks in this “Global South as Kingmaker” narrative. When I looked at the actual numbers behind AI governance,something very striking jumped out: out of nearly 500 AI policies and guidelines developed from 2011 to 2023, about two-thirds came from the US, Europe, or China. Only 7% originated in Latin America and Africa. That’s a pretty big gap between the story of leadership and the reality of who’s actually setting the rules.

I also started noticing and understanding what experts call the “AI divide”. While everyone’s talking about how the Global South will shape AI’s future, only half of AI experts think the technology will actually improve productivity in low-income countries. Most of the experts expect AI to widen the gap between rich and poor nations by exacerbating inequality. That’s not exactly the empowerment story we’ve been hearing. In the similar context, India also shows the limits of the influence it has in AI governance. Despite impressive initiatives and growing diplomatic clout, the country still operates within structural constraints it didn’t set. Therefore, it is important to acknowledge that the global AI supply chain, the fundamental research infrastructure, and the capital markets that fund AI development, all remain concentrated in traditional power centers and are not going to get decentralized very soon.

In this regard, the China angle is particularly interesting because it cuts both ways. China has been smart about offering what we can call“AI in a box” which is complete, ready-to-use systems through the Digital Silk Road that countries can adopt without needing armies of AI experts. It looks like a great deal: smart cities, surveillance systems, government automation tools that just work. But when you dig deeper, it’s creating what researchers call “technological dependencies.” Countries get access to the technology but lose control over their digital infrastructure.

Then there’s the uncomfortable truth about data contribution. Sure, Global South countries are providing valuable data that makes AI systems better, but the way this happens often looks a lot like old-school exploitation. Underpaid workers in these regions train algorithms, while intellectual property barriers keep local companies from benefiting. The data gets extracted, processed elsewhere, and turned into products that get sold back to the same regions that provided the raw material. This already sounds very familiar!

On top of that, what I find most critical and eye-opening is looking at the choices these countries actually have. The research shows they’re often stuck choosing between Western systems that are complicated and expensive to deploy, and Chinese alternatives that are cost-effective but come with serious long term implications for data sovereignty and geopolitical alignment. That’s not really a strategic choice that the countries of the Global South have to make, but that’s only a picking between two problematic options.

Furthermore, the multilateral governance story is similarly nuanced and perplexing. Yes, Global South countries have numerical majorities in international forums, but structural power imbalances persist regardless of the numbers. It has been observed continuously that they’re often invited to discussions only, rather than setting agendas. To give the final calls and make the real decisions about AI development, the ones involving massive capital investment, advanced infrastructure, and cutting-edge research are still happening in countries with the resources and expertise to make them.

Now, looking at the future scenarios, I think what’s happening is more complex than either the “Global South as kingmaker” or the “continued Northern dominance”narratives suggest. These countries definitely have more agency than in previous technological transitions. They’re setting some rules, making some choices, and influencing some outcomes. But they’re doing it within constraints that limit their ability to fundamentally reshape AI geopolitics. The reality seems to be that while the Global South matters more than ever before, the structural advantages of wealthy nations in AI development ranging from the capital, infrastructure, expertise, and control over key technologies are proving pretty resistant to demographic rebalancing. These countries are players in the game, but they’re not necessarily the ones writing the rules.

As I see it, the future probably won’t be dominated by two superpowers or determined by emerging economies alone. Instead, we’re heading toward something messier and more unpredictable like decentralized blockchain system—a world where influence is more distributed but still unequal, where demographic weight matters more than before but economic power still counts for a lot, and where the Global South has real agency but operates within limits they didn’t set.

Thus, to sum up , the question isn’t whether the Global South will control AI’s future, but how much influence they’ll have in shaping it. Additionally, whether that influence will be enough to make the technology work for everyone, not just the countries that built it first. India’s approach suggests it’s possible, but the jury’s still out on whether possibility becomes reality.

Sudhanshu Kumar is a Subject Matter Expert on AI, Cybersecurity and Cyberwarfare at CENJOWS (Centre for Joint Warfare Studies), Ministry of Defence, New Dehli.