While India builds data centres, China is building data markets

While Washington debates how artificial intelligence may influence the next phase of warfare, and New Delhi focuses on attracting global technology investment, Beijing is building that may sound almost dull: data exchanges.

In cities such as Shanghai and Shenzhen, marketplaces are growing where companies, universities and Government agencies can buy and sell datasets. Industrial production records, logistics flows, medical images and transport patterns are all becoming tradable assets. Though officials describe these platforms as part of China’s digital economy strategy. Analysts estimate that the country’s data economy could exceed $1 trillion before the end of the decade.

On the surface, it looks bureaucratic — the sort of institutional machinery China often creates when it wants to organise a sector at scale. But look closer and the logic begins to reveal itself. Artificial intelligence runs on information much the way the industrial age ran on coal and oil. China appears to be building the infrastructure to treat data as a national resource for machine intelligence. Over the past two years, more than 20 Government-backed data exchanges have opened across Chinese cities. The Shanghai Data Exchange alone lists hundreds of tradable datasets, ranging from logistics flows and industrial production records to medical imaging and urban transport patterns.

Chinese officials estimate the country’s data market could reach 60 trillion yuan (about $8 trillion) by 2030, according to projections tied to the national digital-economy strategy. Independent analysts

are more cautious, but many still place the value of China’s data economy at over $1 trillion before the decade ends. And this is where much of the global conversation about artificial intelligence begins in the wrong place.

Behind every impressive AI system sit three basic ingredients: advanced chips, specialised talent and vast quantities of structured data. Remove any one of these and the system weakens. Right now, the United States dominates the first two. American companies design the processors that power modern AI systems. Silicon Valley still attracts much of the world’s top AI talent and investment. Washington has also begun to defend that position aggressively. In recent years the US Government has imposed export controls on advanced semiconductor technologies, restricting China’s access to some of the most powerful AI chips.

China, however, seems to be pursuing a different advantage. The country produces roughly 30 per cent of global manufacturing output, more than the United States, Japan and Germany combined. Its digital platforms process extraordinary volumes of commercial activity. During major online shopping festivals, Chinese e-commerce companies can handle billions of transactions in a single day.

Every transaction produces information — about consumer behaviour, supply chains, pricing and logistics. AI thrives on exactly this kind of data. Machine-learning systems do not “think” in the way humans do. They learn by detecting patterns across enormous datasets. The richer the data, the smarter the system becomes. Yet in most countries, this information remains scattered. It gets limited inside databases only.

China’s emerging data exchanges are an attempt to change that. Instead of letting information remain fragmented, Beijing is experimenting with marketplaces where datasets can be standardised and traded. Companies can monetise data products. Researchers can access information for machine-learning projects. Government agencies can verify and catalogue data before it enters the exchange.

Seen this way, the idea becomes clearer. China is trying to turn the everyday information generated by its economy into training material for artificial intelligence. At a time when most countries ask how to build the smartest AI model.

China appears to be asking a different question: what happens if an entire economy becomes a training environment for AI? History suggests that this kind of systems thinking often matters more than individual inventions.

Britain pioneered many early developments in computing, yet it was the United States that ultimately built the modern computer industry. Germany produced remarkable rocket engineers during the Second World War, but it was the United States and the Soviet Union that turned rockets into space programmes.

The decisive advantage may not lie with the country that produces the smartest algorithm. It may lie with the country that builds the largest ecosystem capable of absorbing and learning from

intelligent machines. From that perspective, China’s data exchanges start to look less like administrative tools and more like infrastructure for intelligence.The United States continues to dominate frontier research and semiconductor design. China is exploring ways to convert its industrial scale and digital activity into a continuous learning resource for AI systems.

Both approaches reflect the same underlying recognition: artificial intelligence is no longer just a technology sector. It is becoming part of the architecture of national power.

For India, it raises an obvious question For India, the more interesting question is not whether the country has enough data for artificial intelligence. It almost certainly does. India has become one of the largest generators of digital activity in the world. The Unified Payments Interface now processes more than 12 billion transactions every month, according to the National Payments Corporation of India, moving values that frequently exceed INR 18 trillion. Aadhaar covers over 1.3 billion people, making it the largest biometric identity system ever built. Internet access has crossed 850 million users, creating one of the largest connected populations on the planet. Taken together, these systems generate something unusual: a constantly updating digital record of how a billion-person economy actually behaves.

Many advanced digital economies are built around private data ecosystems. In the United States, vast amounts of behavioural data sit inside large technology firms such as Google, Amazon and Meta. In Europe, strong privacy frameworks like the GDPR protect citizens but also make large-scale data sharing across institutions slow and complicated. China, by contrast, has moved toward a more centralised model, experimenting with Government-backed data exchanges that allow datasets to be standardised and traded across industries.

India sits somewhere between these models. Rather than relying purely on corporations or centralised state systems, India built much of its digital foundation as public infrastructure. Aadhaar provides identity, UPI enables real-time payments, and the broader India Stack allows banks, startups and Government services to operate on shared digital rails.

In effect, India has created something resembling a national digital nervous system. Every digital payment, subsidy transfer, logistics transaction or online service leaves behind structured information about economic activity. Yet much of this information still remains locked inside institutional silos. Banks hold financial transaction histories. Hospitals store medical records. Telecom firms manage communication data. Government databases track agriculture, welfare and infrastructure systems. Very little of this data circulates in ways that allow artificial intelligence systems to learn from it.

This is India’s real AI challenge. Not producing data, but organising it intelligently. Unlike China’s centralised system, India must design ways for data to move across sectors while preserving privacy, consent and democratic accountability. If it succeeds, the result could be something rare: a trusted data ecosystem where public infrastructure allows innovation without concentrating power entirely in either the state or large corporations. Artificial intelligence ultimately learns from how societies organise their information. The countries that solve that institutional puzzle most effectively will build the technological landscape of the twenty-first century.

Author is a theoretical physicist at the University of North Carolina at Chapel Hill, US, and the author of the forthcoming book The Last Equation Before Silence; views are personal