Five carriers. One direction.

Why one path is never enough

Open any block explorer and look up your own address. What you will find is not a summary. It is a complete history, written by the network itself, permanent by construction, readable by anyone with a browser and a passing interest. Every payment you have received. Every payment you have made. The amounts, the counterparties, the exact second each one settled. It is all there, and it will still be there long after you have stopped caring about any of it.

Most people do not mind this until they do. Then they mind a great deal. A salary becomes public. A supplier becomes public. A settlement with a counterparty who would rather not be named becomes public, and stays public, and is indexed by services whose entire business is joining those dots together.

The uncomfortable part is not that a record exists. It is that the record is shaped like a line. One end is you. The other end is whoever received the money. Everything in between is bookkeeping. There is nothing to interpret and nothing to dispute, which is precisely what makes it so revealing.

The problem was never the record. It was that the record had only two ends.

A great deal of effort has gone into making that line harder to read. Some of it works. Almost all of it works by making the line more complicated, which is a different thing from removing it. A complicated line is still a line. It takes longer to follow, and the people who follow lines for a living are patient.

The mesh takes a different position. Instead of making the line harder to read, it declines to produce one. What leaves you is not the same thing that arrives, not in any sense that can be written down as a single edge in a graph. There is no line to make complicated.

What the mesh is

The mesh is five carriers. Each one takes what it is given, signs for its own part of the journey, and hands the result to the next. No carrier is told where the value came from. No carrier is told where it is ultimately going. Each one knows only what it needs in order to do its job, and then it is done.

You interact with exactly one of them, once. From your side the process is a single approval and a wait of a few seconds. From the other side it is five separate acts, each performed by something that has no view of the whole.

This is the part that matters, and it is worth stating carefully. It is not that the carriers are forbidden from looking at the whole picture. It is that the picture is never assembled anywhere for them to look at. There is no ledger entry, on any network, that records the complete journey. There is no operator console that displays it. The shape of the thing is the privacy.

Carriers

05Each signs for its own part, and only its own part.

From you

01Signature. There is nothing else to approve.

Public links

00No graph connects what you sent to what arrived.

The carriers

Each carrier has a name, because a system that moves value on your behalf should be able to tell you who is doing what. The names describe what happens to the trail, not how it happens. That distinction is deliberate and it is the only thing about the mesh we will not expand on.

The order is never the same twice

The carriers do not run in a fixed sequence. The order is decided per transfer and it differs every time. This is not a flourish and it is not a random number generator for its own sake. A fixed order is a pattern, and a pattern is a thing that can be learned.

If the mesh always ran in the same sequence, then watching enough transfers would eventually tell you what that sequence was. You would not need to break anything. You would simply observe until the shape repeated, and then you would know what to look for. Rotation removes that foothold. There is no sequence to learn because there is no sequence, only a set of carriers and a rule about how many of them a transfer passes through.

The same reasoning applies to timing. Transfers do not run to a fixed schedule, and the time a carrier takes is not a constant that could be used as a fingerprint. Anyone attempting to correlate two ends of a transfer by matching intervals would need those intervals to be stable. They are not.

What we publish, and what we do not

There is a line between describing a product and handing over its design. We try to stay on the right side of it, and it is fair to be explicit about where the line sits.

We publish the networks. They are listed on the landing page under supported technology, and they are not a secret. Value has to settle somewhere, and the person receiving it has a right to know where it will land. Infrastructure is not the same thing as method.

We do not publish which carrier uses which network, or in what order the carriers run. That mapping is the private path. Handing it over would not be transparency. It would be giving away the only thing standing between your transfer and a graph that anyone could draw.

There is a third category, which is the part we decline to discuss at all. How a carrier does what it does, how one hands to the next, and where the seams are. Publishing that would be publishing the recipe, and a recipe can be followed by anyone, including people whose interest in your privacy is not friendly.

We will tell you what the mesh does for you. We will not tell you how to build one.

What we can claim, and what we cannot

A privacy product that overstates itself is worse than one that does not exist, because people make decisions based on what it says. So, plainly:

We can claim that no public ledger connects what you sent to what arrived. That is what the mesh is for and it is what the mesh delivers. There is no edge to follow, on any network, between the two ends of a transfer.

We do not claim that nobody can see anything. While settlement is operated for you, that operation necessarily has a view of a transfer while it is in flight. That is a real limitation and it is the reason the roadmap includes a phase that removes it. Until that phase ships, the honest description of the mesh is that it removes the public record, not that it removes all knowledge.

If you have read this far and what you want is a guarantee that no human being anywhere could ever form an opinion about your transfer, no such guarantee exists in this design or in any other. What exists here is narrower and it is real: the permanent, public, machine readable link between sender and receiver is gone.

Why this is worth the trouble

There is a version of this argument that treats privacy as a matter of taste, a preference some people have and others do not, like whether you keep your curtains open. That framing has never been right.

A financial system where every transaction is public by default is a system where every participant is permanently exposed to everyone with an interest in looking. You do not get to decide who finds it interesting. You do not get to decide when. A record that is public today is public in ten years, when the counterparty is different, the context is different, and the reasons you had for the transfer are nobody's business but your own.

The mesh is our answer to that, and it is the only thing we are trying to build. Not a feature you switch on, but a property of the system you cannot switch off.