The National Data Group Is Established: In the AI Era, the Most Valuable Thing May Be Data Others Cannot Obtain

You may not expect it: in the AI era, the most valuable thing in the future may be neither chips nor large models, but data that others cannot obtain, cannot buy, and cannot easily replicate.

On October 20, the National Data Group Co., Ltd. was established in Xiong'an New Area. It is a central state-owned enterprise whose main business is the development and utilisation of data resources.
Many people, upon seeing this news, have the first reaction: the state has begun to value data.
But what truly deserves attention is that China is trying to solve a deeper problem:

Data is everywhere, so why are there not as many enterprises as one might imagine that can truly turn data into productivity?

Think about it: a manufacturing enterprise generates equipment operation data, production data, quality data, and maintenance records every day.
A hospital has a large amount of medical data, a bank holds transaction data, and an industrial software system records the business rules accumulated by an enterprise over many years.
All this data is valuable.
But the problem is that it is scattered across different enterprises, departments, and systems, with different formats, different standards, and different ownership and usage permissions.

The data may be there, but it cannot necessarily flow; even if it can flow, it may not be directly usable.

What the National Data Group aims to promote is precisely the development and utilisation of data resources and the construction of a data industry ecosystem. Official reports also explicitly mention that data circulation still faces problems such as cross-domain barriers, high governance costs, and insufficient supply of high-quality datasets.

This means:

In the second half of the data industry, the competition may no longer simply be about who owns more data, but about who has the ability to turn scattered, chaotic, and hard-to-use data into trustworthy, compliant data products that can create value.

01.This Has an Especially Big Impact on the AI Industry

Today, large models are becoming more and more capable, and AI programming is making software development easier and easier.
But you will find that having AI write a general-purpose tool is not difficult – what is truly difficult is enabling it to accurately understand a factory's processes, the patterns of equipment failures, and the rules behind a complex business process.

Because models can be replicated and general-purpose code can be generated, but the real business experience accumulated by an enterprise over more than a decade cannot be produced out of thin air.

So there is a value shift happening in the AI era:

Competitive advantage may be shifting from simply possessing technology to possessing industry data and business knowledge that technology cannot easily replicate.

02.What Does This Mean for Enterprises?

In the future, enterprise informatisation may not just be about purchasing an MES, ERP, or industrial software, but about thinking:

Can our data be connected?Can it be accumulated into reusable knowledge?Can it continuously improve production decisions?

03.For Software Companies, This Is Also a Reshuffle

If you still only make money by writing code, building pages, and delivering features, AI will continuously drive down the price of such work.
But if you truly understand an industry, know how data should be collected, governed, and combined with business rules, and ultimately how to help customers reduce costs, reduce failures, and improve yield, then what you provide is not just software, but industry capability.

Of course, the establishment of the National Data Group does not mean data will automatically circulate, nor does it mean all data can be centrally used. Data security, compliant authorisation, distribution of commercial interests, and actual application effects are still issues that must be resolved.
But the direction is already worth taking seriously.

In the past, many enterprises rushed to build systems and deploy platforms to store data.
Next, the more important question may be:

Can this data actually generate business value?

As AI makes the cost of producing intelligence lower and lower, what is truly scarce may not be intelligence itself, but the data, scenarios, and industry experience that enable intelligence to play a role in the real world.

This is what is truly worth thinking about regarding the establishment of the National Data Group.