The historical adage of “data is the new oil’’ has become a geopolitical reality. However, while oil is a physical resource, data is an infinite by-product of human activities that is generated through every digital interaction. While the Industrial Revolution was built on the extraction of minerals and labour from colonised countries, the Artificial Intelligence (AI) revolution has been built on the extraction of “behavioural patterns” of global citizens. Scholars and policymakers refer to this process as “data colonialism.” The actual problem is that the data is generated globally, particularly in the high-growth markets of the Global South, but the profits derived from it are concentrated in the hands of a few multinational corporations in the Global North. However, international trade law, particularly the World Trade Organisation (WTO) framework, is not strengthened enough to prevent this digital resource drain, as it was developed in the late 20th century when the digital world was still evolving.
The Concept of Data Colonialism
In the past, colonial powers took control of land and natural resources to benefit themselves. Today, a similar pattern can be seen in the digital world. Instead of land, it is the user data collected through searches, social media, and GPS tracking, that is utilised by tech giants. This data is exported to foreign servers, processed into AI algorithms, and imported back to the same users as a subscription service. This creates an asymmetry in which the provider of the raw material (the user), receives a convenience, while the extractor (the corporation) gains capital and market dominance. This relationship mirrors the colonial patterns of the 18th century, where raw materials were taken for pennies and sold back as finished goods for pounds.
AI and the Global Data Economy
The growth of Artificial Intelligence has driven a sharp increase in the demand for digital data. AI systems, especially Large Language Models (LLMs), depend on large datasets to function and to keep improving over time. These datasets often extracted from the public internet without any direct compensation to the users, most of whom are in the Global South and provide the diverse linguistic, cultural, and behavioural data that is required to make AI global. This creates a clear imbalance: despite giving up their data, they receive nothing in return and do not own the infrastructure used to process that data. Consequently, developing nations are facing a “dependency loop’’ where they have to pay the subscription, which effectively trained on their own national resources. This is not restricted to only a technical issue, but it is a trade justice issue that widens the global digital divide.
Legal Gaps in International Trade Law
The current international legal framework is no longer adequate to address these developments. The World Trade Organisation (WTO) and the General Agreement on Trade in Services (GATS) drafted at the time when AI and data technologies had not become central to the global economy. Under GATS, there is no clear classification for “data” or “AI models,” whether they fall within goods, services or something an entirely new category remains unclear. This ambiguity allows multinational firms to operate in a gray area. Further, the push for “Free Flow of Data” in trade agreements ignores the concept of “Data Fairness.” Current international trade rules prioritise the removal of trade barriers for corporations but do not offer any protection for the country who is providing the data. This creates a system where trade law promotes liberalisation at the expense of equity.
Recently, Cameroon hosted the WTO’s 14th Ministerial Conference, where the issue of the moratorium on electronic transmissions was a key point of discussion. First introduced in 1998, the moratorium renewed every two years and prevents countries from imposing customs duties on digital or electronic transmissions. However, member countries failed to reach a consensus on extending the moratorium. Consequently, for the first time since its introduction, the moratorium lapsed on March 31, 2026, marking the end of 28 years of duty-free digital trade. While developed countries including the United States and the European Union supported its extension, developing countries such as South Africa and India opposed it, citing concerns over significant revenue losses.
Data Colonialism as Trade Inequality
The framing of data colonialism reveals that digital trade is not equal for all. Powerful nations demand “Data Free Flow with Trust,” and advocate for a system where their companies can access foreign data without restrictions. For developing nations, this can be seen as a threat to “digital industrialisation.’’ If a country cannot control its data, it cannot develop its own AI industry, as it will always be outcompeted by foreign companies that already possess the aggregated data of its citizens. Therefore, data colonialism is a form of trade inequality that traps developing nations as permanent consumers of foreign technology rather than producers of their own.
India’s Approach to Digital Sovereignty: A Case Study
India has emerged as a prominent voice against the status quo of data colonialism. With one of the world’s largest populations of internet users, it accounts for a considerable portion of global raw data generation. Estimates from 2024 suggest that developing countries could lose around $10 billion annually in potential tariff revenues, while India’s own losses estimated to exceed $500 million per year if WTO e-commerce moratorium had been extended. In response to these foreign technology firms, India has advanced “Data Localisation’’ policies, requiring certain types of data be stored on servers within its borders.
India is in a strong position to influence the evolution of global trade rules. As its economy expected to become the third largest in the world by 2027-28, it has also emerged as a leading voice for the Global South, often mediating between developed and developing countries on key issues like climate change, trade and security. This positioning will be most effective if India actively contributes to rule-making rather than opposing existing frameworks.
Way Forward
To make the digital order more balanced, there is a need to reform the current international trade frameworks. It is important to acknowledge that data is not only a by-product of digital activity, but also has a real economic value.
One possible reform can be benefit-sharing. This principle was adopted in the Nagoya Protocol where it was made sure that countries are compensated for the use of their genetic resources. This same principle can be applied when a country or a company exploits the datasets of another country for training commercial AI systems. This way, some value can be given to the source country through which they may develop or strengthen their own AI systems.
Data sovereignty is also one of the key points of discussion. As of now, data is freely extracted from developing countries and is controlled and stored by multinational companies or a few developed countries. It is later stored on foreign servers and can be processed elsewhere. This means that even when data originates from a country, it has very limited control over how it is ultimately used or monetised. The data generating countries should have the right to decide whether certain kinds of data can leave their countries or not. They should also have the right to require companies to store their data in local servers and set conditions on how the data can be processed or shared.
This will lead to the economic development of the data originating country. It would encourage start-ups, local industries, and AI systems. This will also empower the government to protect sensitive data and enforce security and privacy laws. Developing nations should also be supported through technology transfer, capacity-building initiatives, and access to digital infrastructure. Without such measures, the gap between data-rich and technology-rich nations will continue to widen.
Conclusion
The practice of free data extraction without paying any tax to the data source country should be restricted, otherwise there will be a significant digital divide between developed and developing countries. A shift in power dynamics was witnessed in the recent events of March 2026 at the WTO, where countries like Brazil and South Africa have emphasised the need to end the electronic transmission moratorium and asserted their position at the global level.
The objective is to make international digital trade more inclusive by addressing the needs of developing nations. To ensure this, it is necessary to follow the principles of justice, equity, and sovereignty so that the benefits of the AI revolution are evenly distributed. In the end, the question of global justice can only be answered by shaping the evolving international digital order.
