Who Actually Runs the Internet? ICANN, IANA & the RIRs

Nobody runs the internet the way a company runs an office, yet it holds together beautifully. This guide explains who actually runs the internet, and how ICANN, IANA and the Regional Internet Registries keep its shared parts coordinated.

Who actually runs the internet?

The honest answer to who runs the internet is: no one, and everyone. The internet is not a single machine owned by a single organisation. It is a network of networks, tens of thousands of independent systems run by internet providers, universities, companies, and governments, all agreeing to connect to one another. Nobody is in charge of the whole thing.

That raises an obvious question. If no one owns it, why does it work so smoothly? The answer is that a small number of shared resources must stay consistent for the whole system to function. Two devices cannot use the same public IP address, and a domain name must point to the same place for everyone. Coordinating those shared identifiers is the quiet, essential job handled by a handful of organisations.

Think of it a little like the rules of the road. No single company owns the world's roads, yet driving works because everyone agrees on a shared set of conventions and a few authorities keep the signs and numbering consistent. The internet is much the same. The coordinating bodies do not drive the cars or build every road; they just make sure the map stays coherent so that everyone else can travel freely.

ICANN: coordinating names and numbers

ICANN, the Internet Corporation for Assigned Names and Numbers, is the body most often named when people ask who runs the internet. It is a non-profit organisation with a multi-stakeholder structure, meaning its decisions involve technical experts, businesses, governments, and ordinary internet users rather than a single authority.

ICANN's role is coordination, not control. It does not decide what you can publish or which sites you can visit. Instead it oversees the systems that keep unique identifiers unique: the domain name system that turns names like example.com into addresses, and the top-level allocation of IP address space. Its aim is a single, stable, interoperable internet.

The multi-stakeholder model is worth dwelling on, because it is unusual. Rather than answering to a single government or company, ICANN makes decisions through open processes in which many different groups have a voice. Technical volunteers, businesses, civil-society advocates, and government advisory representatives all take part. The idea is that the internet, which belongs to no one, should be coordinated by a body that likewise belongs to no single interest. It is deliberately messy, and that messiness is a feature rather than a flaw.

IANA: the master list-keeper

Underneath ICANN sits the IANA function, the Internet Assigned Numbers Authority. If ICANN is the coordinating organisation, IANA is the careful record-keeper that carries out the actual allocations. IANA maintains the top of several critical hierarchies.

IANA manages the root of the domain name system, keeps the registries of protocol numbers that engineers rely on, and holds the master pool of IP addresses. It does not hand addresses to individuals or companies directly. Instead it distributes very large blocks to the regional bodies below it, which is where the Regional Internet Registries come in.

The Regional Internet Registries

The world is divided into five regions, each served by a Regional Internet Registry, or RIR. These non-profit organisations receive large blocks of IP addresses from IANA and then allocate smaller portions to the internet providers and organisations in their region. When your internet provider needs addresses to give to its customers, it obtains them from its RIR.

This tiered structure is why an IP address can be traced to a rough region and provider. The RIRs keep public records of which organisation holds which block, information that also underpins how tools guess location, as explained in how IP geolocation works.

The RIRs do more than hand out numbers. They maintain the public databases that let network operators see who is responsible for a given block of addresses, which is essential for resolving problems and tracing abuse. They also play a central role in managing the transition to newer addressing, having spent years carefully rationing the last of the older IPv4 space as it ran low. Our guide to IPv4 address exhaustion tells that side of the story in more detail.

How the pieces fit together

Picture the address system as a chain. IANA holds the master pool and hands enormous blocks to the five RIRs. Each RIR breaks its allocation into smaller pieces and gives them to internet providers. Each provider then assigns individual addresses to homes and businesses, often dynamically, as our guide to static vs dynamic IP addresses describes. At every step the goal is the same: make sure no two networks claim the same address.

Names work in a parallel way. ICANN coordinates the domain name system, accredited registrars sell domain names to the public, and the whole hierarchy stays consistent from the root downward. The result is a system where a name or number means the same thing everywhere in the world.

There is one more group worth mentioning, because it shapes the internet just as profoundly. The technical rules that all these networks follow, the protocols themselves, are developed in open engineering communities rather than dictated by any authority. Standards for how data is addressed, routed, and secured emerge from proposals that anyone can read and comment on, and they only take hold when networks choose to adopt them. So the internet is coordinated on two fronts at once: the shared identifiers on one side, and the shared rulebook of protocols on the other.

What these bodies do not do

It is worth being clear about the limits of this coordination. ICANN, IANA, and the RIRs manage identifiers. They do not run the physical cables, they do not decide the content of websites, and they do not police what people do online. The actual routing of your data is handled by the independent networks themselves, cooperating through protocols like BGP. Governments regulate networks within their borders, but no single authority governs the global whole.

This division of labour is what keeps the internet both free and functional. If a single organisation truly ran everything, it could impose order but would also become a single point of failure and a tempting lever of control. If nobody coordinated the shared identifiers at all, the system would collapse into chaos as networks collided over names and numbers. The arrangement we have threads the needle: just enough central coordination to keep the shared parts consistent, and no more. It is not perfect, and it is argued over constantly, but it has held a sprawling global network together for a remarkably long time.

The surprisingly human answer to who runs the internet, then, is that it is run by cooperation. A modest set of trusted organisations keeps the shared numbering and naming tidy, and everyone else agrees to play along. To see where this all began, read our brief history of the internet, or head to IP Animals to see the public address your own corner of this vast cooperative was assigned.