The AI Governance Gap and What It Means for the Global South

The confusion in Washington this week over AI incident reporting — in which an Anthropic safety disclosure became entangled in a dispute over the Trump administration’s approach to AI oversight — is being read in Silicon Valley as a regulatory story. From Islamabad, or Nairobi, or Jakarta, it reads differently: as a reminder of how little say most of the world has over the infrastructure that is increasingly shaping its economic and political life.

The governance gap

The United States and the European Union are locked in a familiar competition over who sets the global standard for AI regulation. The EU’s AI Act, now in phased implementation, establishes a risk-tiered framework with genuine extraterritorial reach. Washington has oscillated between a light-touch approach and more interventionist instincts depending on which administration holds office. China has its own regulatory architecture, oriented toward state control rather than liability frameworks.

What these three powers share is agency — the ability to shape rules that others must follow. Pakistan, in common with most of the developing world, has none. Pakistani firms that use AI tools from American providers operate under terms set in California. Pakistani citizens whose data flows through these systems are protected — or not — by American regulatory decisions and EU adequacy determinations, not by anything Islamabad has enacted.

Why this matters more than it appears

The practical consequences are not abstract. AI-driven credit scoring, hiring algorithms, and content moderation systems are already operating in Pakistan, largely through tools built and governed elsewhere. When those systems encode biases or make consequential errors, the redress mechanisms are in foreign jurisdictions. Pakistani consumers and workers bear the risk without having shaped the rules.

At the macro level, the countries that lead in AI will capture a disproportionate share of the productivity gains the technology delivers. The compute divide — which we have written about previously — is one dimension of this. The governance gap is another: states that cannot participate in setting AI rules also tend to struggle to attract AI investment, since the regulatory environment is partly set elsewhere and may change without notice.

Pakistan’s position

Pakistan has the beginnings of a digital economy strategy and a nascent tech sector that punches above its weight in certain niches. What it lacks is a coherent AI governance framework of its own, a seat at any of the multilateral tables where these standards are being negotiated, and — most fundamentally — the leverage that comes from being home to significant AI infrastructure or capability.

This is not unique to Pakistan. It describes most of the world. But Pakistan’s particular combination of a large young population, a digitalising economy, and a governance environment that has struggled to keep pace with technological change makes the gap especially consequential. The rules being written in Washington and Brussels will shape Pakistani lives. Pakistani voices are not in the room.