Quantum Internet Is Closer Than You Think

The Light That Refuses to Die

Every time you send a message across the internet, your data gets copied. Routers read it, amplify it, and pass it along. That copying is what makes the internet work. It is also what makes it insecure. Anyone with access to a router can read your message, and if they are careful, you will never know.

Now imagine a network where copying is physically impossible. Where a single photon carries your message, and the moment anyone tries to read it, the message vanishes. That is the promise of the quantum internet. The problem is that photons, left to their own devices, travel about 100 kilometers through optical fiber before they die. Beyond that, the signal is gone.

For twenty years, that distance limit has been the hard wall between a laboratory curiosity and a global network. But a comprehensive review published in Reviews of Modern Physics by Koji Azuma of NTT Corporation, Sophia Economou of Virginia Tech, David Elkouss of the Technical University of Madrid, and Paul Hilaire of the University of Innsbruck argues that we have finally found a way through that wall (Azuma et al., 2023). The solution is something called a quantum repeater, and it does not copy your data. It teleports it.

Why Photons Are So Fragile

The basic unit of quantum internet is the qubit, and the most practical qubit for sending information over distance is a single photon. You encode information in its polarization: horizontal means 0, vertical means 1, or some superposition of both. This is elegant and secure, but it has a brutal limitation.

Optical fiber is not perfectly transparent. Photons scatter. They get absorbed by impurities in the glass. After about 100 kilometers, the probability that your photon arrives intact drops below 50 percent. For longer distances, you need to boost the signal. But here is the problem.

Classical repeaters work by measuring the incoming signal, making a perfect copy, and sending that copy onward. You cannot do that with a quantum signal. The no cloning theorem, a fundamental law of quantum mechanics, states that you cannot make an exact copy of an unknown quantum state. If you try to measure the photon to figure out what it is, you destroy the information you were trying to send.

For years, this seemed like an absolute barrier. If you cannot amplify the signal, you cannot send it further than 100 kilometers. The quantum internet would be limited to metropolitan areas, never crossing continents or oceans.

The Repeater That Does Not Copy

The quantum repeater solves this problem by doing something that sounds like magic. It does not measure the photon. It does not copy it. Instead, it uses a phenomenon called entanglement swapping to extend the range of the signal.

Here is how it works. Imagine you want to send a qubit from New York to Los Angeles. You cannot send the photon directly because the distance is too far. So you break the path into segments, say 80 kilometers each. Between each segment, you place a quantum repeater.

The repeater does not touch your photon. Instead, it creates two pairs of entangled photons. One pair connects the repeater to the previous node. The other pair connects it to the next node. Then the repeater performs a measurement on the two photons it holds, one from each pair. That measurement entangles the two photons that are far apart, one at the previous node and one at the next node. The information jumps across the repeater without ever passing through it.

This is entanglement swapping, and it is the core mechanism behind every quantum repeater design (Azuma et al., 2023). The authors describe it as a way to "extend entanglement over long distances by dividing the total distance into shorter segments." Each segment establishes entanglement independently, then the repeaters connect those segments together. The information never gets copied. It gets teleported.

The Three Architectures That Could Work

The review identifies three main approaches to building these repeaters, each with different tradeoffs between speed, reliability, and technological difficulty.

First Generation: The Slow and Steady Approach

The first generation quantum repeaters rely on a technique called heralded entanglement generation. Two nodes send photons toward a central repeater. When the photons arrive, the repeater performs a measurement that tells you whether entanglement was successfully created. If it fails, you try again.

The problem is that this process is probabilistic. Each attempt might succeed only 10 percent of the time. For a single segment, you can retry quickly. For a chain of 10 segments, the probability of all of them succeeding simultaneously becomes vanishingly small. You need quantum memory, a device that can store an entangled state for a few milliseconds while you wait for the other segments to succeed.

The authors note that "first generation repeaters require quantum memories that can store qubits for a time comparable to the classical communication time between repeaters" (Azuma et al., 2023). That means storing the state for about 1 millisecond per 100 kilometers. Current quantum memories can do this, but only with limited fidelity. The qubits degrade while they wait.

Second Generation: Faster with Photon Loss Tolerance

Second generation repeaters use a different trick. Instead of trying to generate entanglement probabilistically and then storing it, they use a technique called loss tolerant encoding. You send multiple photons, and even if most of them get lost, the remaining ones still carry the information.

This is analogous to how classical error correction works. You send three copies of your data, and if one gets corrupted, the other two vote to correct it. But quantum error correction is more complex because you cannot simply copy qubits. Instead, you encode a single logical qubit into multiple physical qubits using special codes that allow you to detect and correct errors without measuring the quantum state.

The review explains that "second generation repeaters use quantum error correction to overcome photon loss, reducing the need for long lived quantum memories" (Azuma et al., 2023). This makes them faster than first generation designs, but they require more qubits per repeater and more sophisticated control electronics.

Third Generation: The Holy Grail

Third generation repeaters aim to combine the best of both approaches. They use quantum error correction not just to handle photon loss, but also to correct operational errors inside the repeaters themselves. This would allow them to operate at speeds approaching the classical internet, with error rates low enough for practical applications.

The authors describe third generation repeaters as "the ultimate goal, where all sources of noise and loss are corrected by fault tolerant quantum error correction" (Azuma et al., 2023). These devices would need thousands of physical qubits per repeater, far beyond what current technology can build. But the theoretical framework is solid. The question is engineering.

What the Research Does Not Prove

The review is comprehensive, but it does not claim that a quantum internet is imminent. The authors are careful to distinguish between theoretical feasibility and practical implementation.

They do not prove that any single approach will work at global scale. Each architecture has unresolved technical challenges. First generation repeaters need better quantum memories. Second generation repeaters need lower error rates in their quantum gates. Third generation repeaters require fault tolerant quantum computing, which does not exist yet.

They also do not address the cost question. Building a quantum repeater every 80 kilometers across an ocean is not just a technical problem. It is an economic one. The authors acknowledge that "the cost of deploying quantum repeaters over long distances remains an open question" (Azuma et al., 2023). A transatlantic quantum link would require dozens of repeaters, each containing sophisticated cryogenic equipment and lasers. The price tag could rival the Apollo program.

The biggest open question is whether we actually need a global quantum internet at all. The most compelling application, quantum key distribution, works over shorter distances using satellites. A satellite can send entangled photons from orbit to two ground stations thousands of kilometers apart, bypassing the need for repeaters entirely. The authors note that "satellite based quantum communication may offer a complementary approach to fiber based quantum repeaters" (Azuma et al., 2023). It is possible that the quantum internet will be a hybrid network, using satellites for long haul links and fiber for metropolitan connections.

The Timeline Nobody Wants to Admit

The authors do not give a timeline. That is unusual for a review paper, and it is intentional. The history of quantum technology is littered with overpromises. In the 1990s, researchers predicted quantum computers within a decade. In the 2000s, they said the same thing. Now, in the 2020s, we have demonstration devices that can solve problems no classical computer can, but they are still too noisy for most practical applications.

Quantum repeaters face the same trajectory. The first laboratory demonstrations of entanglement swapping over a few meters happened in the 1990s. The first demonstrations over tens of kilometers happened in the 2010s. In 2020, researchers in China demonstrated entanglement swapping over 50 kilometers of fiber using a simple first generation repeater.

The review catalogs these milestones but does not extrapolate. The authors seem to be saying: we have the physics. We understand the principles. What remains is engineering, and engineering timelines are notoriously hard to predict.

If you press researchers privately, many will say a metropolitan scale quantum network with a handful of repeaters is likely within a decade. A transcontinental network is probably two decades away. A global quantum internet connecting multiple continents is a mid century project, assuming we decide to fund it.

What This Actually Means

• ▸Quantum repeaters are not amplifiers. They do not copy or boost your quantum signal. They use entanglement swapping to teleport the information across each segment. This means the fundamental architecture of the quantum internet will look nothing like the classical internet. It will be a network of entanglement, not a network of data packets.

• ▸The first useful quantum networks will be small and specialized. Before we get a global quantum internet, we will get city scale networks connecting banks, government agencies, and research labs. These networks will use first generation repeaters with modest quantum memories. They will be expensive and slow, but they will be secure against any future quantum computer.

• ▸Quantum memory is the bottleneck. Every repeater design requires storing a qubit for at least a few milliseconds. Current quantum memories can do this, but with error rates that accumulate over time. Improving memory coherence time by even a factor of ten would dramatically simplify the repeater architecture.

• ▸Satellites may beat fiber for global links. Fiber based quantum repeaters require infrastructure every 80 kilometers. Satellites can cover thousands of kilometers in a single hop. The authors explicitly note that satellite quantum communication is a complementary approach. For long haul links, especially across oceans, satellites may be the practical choice.

• ▸The quantum internet will not replace the classical internet. It will add a new layer on top of it. Most of your web browsing, streaming, and social media will continue to use classical networks. The quantum internet will handle only the tiny fraction of traffic that requires absolute security or quantum specific capabilities, such as connecting quantum computers to each other.

The quantum internet is closer than you think, but closer does not mean here. It means we can see the path. We know what the repeaters need to do. We know which architectures might work. We even have laboratory demonstrations of the core principles. What remains is the slow, difficult work of turning physics into engineering.

That work has begun. The first quantum repeater networks are being planned in Europe, China, and the United States. They will be small, fragile, and expensive. But they will work. And once they do, the question will shift from "can we build a quantum internet?" to "what do we do with it?"