Servers can be purchased within months, but building the knowledge and companies that turn it into revenue takes years.

This is where the economic value of Saudi-French cooperation in artificial intelligence becomes apparent. The Kingdom is expanding its digital economy and computing capacity, while France has research institutions and companies working on language models, robotics, and intelligent systems.

As Crown Prince Mohammed bin Salman makes a state visit to France, a question emerges that goes beyond any agreements that may result from it: Will the partnership lead only to the use of technology, or to participation in developing and selling it?

Data Centers Alone Do Not Create Returns

Saudi Arabia’s digital economy grew by 75% over eight years to reach $139 billion. The Kingdom is working to build 6.9 gigawatts of data-center capacity by 2034, beginning with 3 gigawatts by 2030.

These figures reflect a significant expansion of digital infrastructure, but they do not, on their own, reveal the scale of returns that artificial intelligence can generate.

Data centers are like ports: building a port increases its capacity to handle trade, but actual revenue depends on the goods passing through it and the entities that produce them and earn profits from them.

Similarly, servers provide the capacity to run artificial intelligence models and applications, but they do not create value on their own. If they run products developed entirely outside the Kingdom, local returns may be concentrated in construction, energy, operations, and improving the efficiency of various sectors.

But involving local companies and researchers in product development can generate revenue from software and services, while building intellectual property and companies capable of growth.

This does not mean that every technology must be developed locally; purchasing some solutions may be faster and less expensive. The economic difference, however, lies in how much knowledge and revenue remains within the Kingdom, rather than limiting the benefit to operating the product.

What Does France Add to the Equation?

Meetings held by Minister of Communications and Information Technology Abdullah Al-Swaha in Paris in July 2025 reveal that potential cooperation spans different links in the artificial intelligence industry.

The discussions included opportunities to establish joint laboratories and exchange researchers with "INRIA", a French national institute specializing in digital science and technology research.

The Kingdom also discussed cooperation on open-source large language models with "Mistral", a French company specializing in the development of artificial intelligence models.

The discussions also extended to intelligent systems and space technologies with "Thales", a French group working in advanced technologies, and to interactive robots powered by artificial intelligence with "Enchanted Tools", a French company specializing in the development of interactive robots.

They also covered technology investment and support for the expansion of Saudi startups into promising markets with "Eurazeo", a French investment group, as well as data governance and artificial intelligence regulation with "CNIL", the French authority responsible for data protection and privacy.

These are not separate areas; they are links in a chain that begins with research, moves through product development and financing, and ends with reaching the market under clear rules.

But they remain opportunities that have been discussed, not laboratories that have begun operating or joint models whose development has been completed. The transition from discussions to implementation therefore remains the step that will determine the real returns.

From Using Technology to Exporting It

The partnership’s potential impact can be understood across three levels:

Using technology: This is the fastest path, as institutions acquire ready-made solutions that help improve services or reduce costs. But its impact on building a local technology sector remains limited if product development and ownership remain with the foreign company.

Developing technology jointly: This is where laboratories and researcher exchanges become important. Local talent’s participation in development gives them knowledge that purchasing a ready-made product cannot provide, and this expertise may flow into new companies and projects.

Exporting technology: This is the highest-value and most difficult stage to achieve. At this stage, Saudi companies use computing and research infrastructure to develop products that can be sold outside the domestic market, benefiting from financing and access to new markets.

These levels are not mutually exclusive; cooperation may begin with the use of ready-made solutions, then move to joint development, and eventually result in technology that can be exported. The further the relationship advances along this path, the greater the local economy’s share of knowledge and revenue.

The Real Test Begins with Implementation

The partnership’s success will not be measured by the number of meetings or participating entities, but by what reaches the market as a result: functioning laboratories, researchers involved in development, models reaching commercial use, and companies securing investment and entering new markets.

Governance also plays an economic role at this stage, because clear rules for using and protecting data help companies assess the costs and risks of their projects instead of postponing investment due to uncertainty.

The most significant risk is that computing capacity expands faster than products and commercial demand. In that case, servers may be available while the applications running on them remain imported or limited domestically.

The current visit gives Saudi-French cooperation a political boost, but the returns will emerge during the implementation phase.

Ultimately, every new expansion of data centers raises an economic question: Will these servers run a product we import, or one we help develop and can sell?

The answer to this question will determine the Kingdom’s share of the artificial intelligence economy—not the number of servers alone.