The New Infrastructure Battle: Who Controls Connectivity – and Knowledge?

A single private company can now determine whether an entire country has access to the internet. At the same time, a handful of technology firms are becoming the main interface between humanity and knowledge.

Two of the most critical infrastructures of the 21st century – satellite networks and artificial intelligence – are increasingly controlled not by governments, but by private actors. The obvious question is: who should own the infrastructure that shapes how the world connects and thinks?

The Return of an Old Political Question

This question isn’t entirely new. It echoes the classical liberal idea of the night-watchman state: governments should protect borders, enforce laws, and safeguard property rights while private enterprise builds and runs the economy.

For much of the industrial age, that seemed reasonable. Railways, factories, shipping firms, and newspapers were built and run by private businesses. Governments mostly stayed out of it.

By the late 19th century, industrialists controlling railroads, oil, and steel had accumulated extraordinary influence over society. They didn’t just operate markets; they controlled the infrastructure those markets depended on - they became known as The Robber Barons.

The 20th century was, in part, a response to that concentration of power. Governments stepped in – not to replace markets, but to balance them. They built highways, power grids, telecom networks, and satellite systems – infrastructure that wasn’t just economic, but strategic.

Today the pattern is reversing. Two of the most important infrastructures of the digital age – global connectivity and access to knowledge – are increasingly owned and operated by private companies. The result is a new political question: who ultimately controls the systems through which societies communicate, learn, and make decisions?

Now, in the 21st century, we have a new kind of infrastructure – digital, global, and deeply political.

The Infrastructure of the AI Age

Two technologies define this new era: satellite constellations that determine who can access the internet, and large language models (LLMs) that determine how people access knowledge

Satellite constellations now form a privately owned communication layer around the Earth. Connectivity in remote regions – and even in war zones – can depend on the decisions of a single company.

At the same time, AI is becoming the primary interface to the internet. Instead of navigating websites, more users simply ask a system that summarizes research, interprets data, and synthesizes knowledge on their behalf.

Together, these two layers – the physical infrastructure of connectivity and the cognitive infrastructure of intelligence – are reshaping how humanity communicates and learns. And increasingly, both are privately controlled.

When Infrastructure Becomes Power

From a traditional free-market perspective, this seems fine. Governments set the rules, companies innovate, everyone benefits.

But this logic assumes infrastructure is purely economic. What happens when it becomes strategic?

A company that controls a satellite network during a geopolitical conflict wields enormous power. A handful of AI firms that shape how billions understand the world have extraordinary influence over culture, politics, and economics.

The Concentration Problem

Both satellite networks and frontier AI models require massive capital. Training state-of-the-art AI systems costs hundreds of millions of dollars, and building and launching satellites at global scale costs even more.

The result is extreme concentration of capability and capital. Only a handful of companies can play in this league. When so few actors control critical infrastructure, the implications are inevitably political.

What Governments Should Do

If AI models and satellite systems are becoming strategic infrastructure, governments must rethink their role. 

1. Ensure technological sovereignty. Avoid total dependence on foreign or private systems and build redundancy – public, private, or hybrid.

2. Maintain competitive ecosystems. Concentration breeds vulnerability. Regulators must encourage competition and interoperability.

3. Treat AI as public infrastructure. Even privately built systems that shape access to information may require governance frameworks similar to utilities.

What Companies Must Do

For businesses, this shift redefines visibility and reputation. Success in the AI age isn’t just about being online – it’s about being understood by machines.

1. Become legible to AI systems.
Make sure corporate information is structured, credible, and easy for AI to interpret.

2. Build authoritative knowledge footprints. AI systems prioritize consistent, authoritative signals.

3. Treat AI visibility as infrastructure.
Just as SEO shaped the web era, AI interpretability will shape the next decade.

At KontentPlus, we focus on helping organizations thrive in this new landscape – ensuring their knowledge is not only distributed across digital infrastructure, but also correctly interpreted by the AI systems that increasingly mediate access to it.

Companies can influence how they appear in this new machine-mediated world. The larger political questions, however, will ultimately be decided at the ballot box.