New infrastructure loading 3- Data centers and the world order

In the previous installment, I discussed how computing capacity is becoming one of the strategic factors of production in the age of artificial intelligence (AI). However, to understand the economic, political, and strategic importance of computing capacity, we need to take another step. We must look at where computing capacity is produced, how it changes our communication order, and the strategic relationship between data centers.

For example, when we type a few sentences into an AI application on our mobile phones in Istanbul, it is not just software working behind the scenes. The request reaches data centers via communication networks; it is processed there in large computing clusters, and the result returns to the user through the same networks. For the user, the process lasts only a few seconds. But behind those few seconds are thousands of processors, a high amount of electricity, cooling systems, fiber networks, and a physical facility. This situation is not limited to AI chatbots and applications. Watching a YouTube video, opening a movie on Netflix, uploading a video on Instagram, or search results on a search engine all depend on the collaborative work of data centers and communication networks in different parts of the world. In other words, no matter how wireless and cloud-based digital media may seem, it actually runs on a massive physical infrastructure.

The rise of artificial intelligence adds a new function to this infrastructure: while data centers carry communication, they also begin to produce the communication itself. In other words, while the internet puts information into circulation and platforms link this circulation to data, AI pulls content production into the same computing infrastructure. Summarizing a news article, translating a video, generating an image, or answering a search query means producing communication through computation. For this reason, it is not enough to see data centers as facilities in the background of technology. Because the geography of data centers is increasingly turning into the economic geography of communication and computing power.

Northern Virginia, one of the world's largest data center clusters, is a good example of this. The region's proximity to fiber networks, availability of large land tracts, and energy infrastructure have turned it into one of the key nodes of the digital economy over the years. Similarly, Ireland has become one of Europe's major data center hubs. The recent shift of Gulf countries toward large-scale data center and AI investments points to another development. Energy wealth is now also being sought to produce computing capacity. This transformation is not just about how much AI technology companies in Turkey can develop. Turkey's position in the global digital economy must be rethought in terms of data centers, energy capacity, fiber networks, submarine cables, and regional data traffic. While its location between Europe, the Caucasus, the Middle East, and Central Asia provides Turkey with a significant communication and data transit advantage, transforming this advantage into real computing capacity requires strong data center infrastructure, reliable energy, and high-capacity network investments. In other words, while Turkey's position in the digital economy's infrastructure depends on all of this, will Turkey be a corridor through which data passes in this new geography, or will it be one of the centers where data is processed, computing is produced, and regional digital power is concentrated? When a data center is established, servers, electricity, fiber networks, cooling systems, transformer stations, and new energy investments come along with it. Thus, the data center creates a new digital infrastructure economy that surrounds it. The correlation between the map of data centers and the new map of the world economy is increasing.

At this point, a profound transformation is also taking place in terms of media and communication. In the past, communication power was largely related to the capacity to produce and distribute content. Newspapers gained power by controlling printing presses, television channels by controlling broadcasting networks, and news agencies by controlling global networks of correspondents. The internet added platforms and data collection capacity to this structure. AI brings a third layer: computing capacity. Now, the power of a platform may not be measured solely by how many users it has or how much data it collects. How fast it can process that data, which AI models it can run, and the infrastructure on which it can perform this also gain importance. For this reason, Google's data centers, Microsoft's cloud infrastructure, or Amazon's massive server networks are increasingly becoming the physical nodes of the global communication system. In other words, while the visible face of the media economy remains with platforms, a large portion of the economic and strategic weight is shifting toward the infrastructure behind the platforms and the infrastructure monopolies.

The geopolitical consequence of this transformation is even more interesting. Just as the routes of submarine fiber optic cables have become strategic in terms of countries' communication security, where data centers and computing capacity are concentrated may gain similar importance in the future. Because economic and strategic focus will now be associated with where information is processed, where data centers are gathered, by which algorithms they are transformed, and who has control over this process. The large AI infrastructure investments of Gulf countries, the conversion of energy capital into computing capital, also point to a search for a new position in global power balances. Similarly, the control of the US and China over advanced chips and AI infrastructure is increasingly making technology policy a part of national security policy.

In short, the machines running inside data center buildings constitute the computing capacity of the global digital economy; the networks connected to it carry global communication; the energy system feeding it determines production capacity; and the country where it is located tries to gain a new economic and geopolitical advantage through this infrastructure. Therefore, when thinking about freedom of communication or the power of the media in the AI age, looking only at platforms and the platform economy is no longer sufficient. A significant part of the future communication order is being built in data center facilities behind the screens. And for these facilities to operate, besides chips or capital, more "electrical energy" is required. As AI factories grow, the competition over the energy systems that will feed them will also grow. The question for the next installment arises right here: If electricity is required to store, process, and circulate data for artificial intelligence, which countries and which energy systems will be able to carry the computing power of the future? On the other hand, the importance of computing infrastructure and data centers is not limited to the economy. Because the same computing capacity is increasingly becoming a part of military capacity, from communication systems to intelligence, and from autonomous systems to arms industry technologies, it is turning into a strategic and military infrastructure. If the war of the future is to be fought with smarter weapons, it means that the computing that makes the smart arms industry possible will be determined by the capacity to control energy and communication infrastructure. In this context, what is the geopolitical and security meaning of this for NATO? Where does NATO place the relationship between AI, computing capacity, critical infrastructure, and next-generation weapon systems? We will pursue these questions next week.