When Physics Becomes the Firewall
How a New York City experiment brought the vision of a quantum internet a step closer
For as long as people have been connecting computers, others have been trying to break into them. The internet was built to move information quickly, not to keep it safe, and every layer of security added since has been a patch on top of a system that was never designed to be tamper proof. Scientists have spent decades imagining something different, a network where the laws of physics themselves protect information. In that vision, any attempt to intercept a message would leave unmistakable fingerprints. Researchers call this idea a quantum internet, and a team from New York University, the Brooklyn based startup Qunnect, and Cisco has just taken a meaningful step toward making it real.
The experiment, described in the paper High rate Scalable Entanglement Swapping Between Remote Entanglement Sources on Deployed New York City Fibers shows that quantum signals can be linked across multiple points using the same fiber that already runs under New York City. It is the first demonstration of polarization entanglement swapping across a metropolitan network using deployed telecom infrastructure. That achievement matters because it begins to turn a set of laboratory techniques into something that resembles a working network.
To understand why this is important, it helps to start with what a quantum internet actually is. Instead of sending electrical pulses or radio waves, a quantum network uses individual particles of light to carry information. These particles, called photons, can be prepared in delicate quantum states that behave in ways that have no counterpart in everyday experience. One of those behaviors is that measuring a quantum particle changes it. If someone tries to intercept a quantum signal, the act of looking at it disturbs the particle and reveals the intrusion. This property makes quantum communication fundamentally different from the classical internet, where eavesdropping can be invisible.
Photons are the preferred carriers for quantum information because they travel easily through fiber, they can be created on demand, and they interact very weakly with their surroundings. That weak interaction is a double-edged sword. It keeps the information intact, but it also means photons are easily lost. A quantum internet cannot simply amplify a weak signal the way the classical internet does, because amplifying a quantum state destroys the very information it is meant to preserve. This limitation is one of the reasons the field has been slow to move from theory to practice.
To overcome this, researchers rely on a phenomenon called entanglement. When two photons are entangled, they behave like a single system even when separated by large distances. Changing one affects the other in predictable ways. Entanglement allows information to be shared without sending a fragile quantum state directly from one place to another. Instead, the correlations between entangled particles do the work.
The challenge is that entanglement does not naturally stretch across long distances. Fiber absorbs photons, and after a few tens of kilometers the signal becomes too weak to use. The solution is a technique called entanglement swapping. It allows two photons that have never interacted to become entangled by using a pair of intermediate photons that meet at a central point. The process stitches together shorter entangled links into a longer one, much like joining two pieces of rope by tying their ends together. Entanglement swapping is essential for building a quantum internet, but it has been difficult to perform reliably outside the lab.
The New York team focused on a specific form of this process called polarization entanglement swapping. Here, the information is encoded in the orientation of the light wave, similar to how polarized sunglasses block light oriented in certain directions. The researchers used two independent sources of entangled photon pairs, each producing one photon at 795 nanometers and another at 1324 nanometers. The shorter wavelength photons stayed at the outer nodes, where they were measured with relatively inexpensive detectors. The longer wavelength photons traveled through the city’s fiber to a central hub, where a specialized measurement linked them together. When that measurement succeeded, the two 795 nanometer photons at the outer nodes became entangled even though they had never met.
This setup followed a hub and spoke design. The outer nodes were simple and operated at room temperature. The central hub housed cryogenic detectors that are needed to catch the faint telecom wavelength photons. This architecture is important because it allows a network to grow without requiring every location to host complex equipment. Only the hub needs the specialized hardware, which makes the system more practical for real world deployment.
The experiment connected two nodes at Qunnect’s facility in the Brooklyn Navy Yard with a third at QTD Systems, a commercial data center at 60 Hudson Street in Manhattan. The fibers between them stretched about 17.6 kilometers. The team used Qunnect’s Carina system to generate entangled photons and to stabilize the polarization of the light as it traveled through the fiber, which can drift with temperature and vibration. Cisco provided orchestration software that synchronized the three sites and coordinated the timing of the measurements. NYU contributed both experimental expertise and one of the network nodes, and its researchers validated the quality of the entanglement produced across the city.
The results showed that the system could perform entanglement swapping at about 1.5 events per second across the deployed network while maintaining the correlations needed for a functioning quantum link. The team also demonstrated that the entanglement remained stable over many hours, even as the fiber experienced environmental fluctuations. This stability is essential for any future quantum network that must operate continuously rather than in short bursts.
Although the experiment does not create a full quantum internet, it addresses one of the field’s central obstacles: how to connect independent quantum devices over real-world infrastructure. Photons traveling through fiber are vulnerable to loss and noise, and the quantum states they carry are easily disturbed. The New York experiment shows that these effects can be managed well enough to sustain entanglement across a metropolitan network. It also shows that the system can scale. Additional nodes could be added without replicating the cryogenic hardware, which is a practical requirement for any citywide or data center scale network.
One of the near term applications often discussed in this context is quantum key distribution. QKD uses quantum states to share encryption keys in a way that reveals any attempt at interception. It is an interesting scientific technique, but it has a significant limitation. QKD requires a trusted intermediary to create and distribute the keys, which introduces a built-in vulnerability. The system is only as secure as the party that manages the keys. This design makes QKD unsuitable for many real-world applications and is one reason it has been embraced by authoritarian governments that prefer centralized control.
The broader impact of the New York experiment lies not in QKD but in the foundation it lays for more advanced uses. A quantum network could eventually link quantum computers, allowing them to work together on problems too large for any single machine. It could enable new forms of sensing that rely on entanglement to measure tiny changes in the environment. It could support distributed clocks that keep time with extraordinary precision. All of these applications require reliable entanglement across distance, and the New York experiment shows that this is becoming feasible on existing infrastructure.
The researchers recommend continued development of scalable architectures that minimize the need for specialized equipment at each node. They also emphasize the importance of integrating quantum hardware with orchestration software that can manage a network in real time. Their work demonstrates that quantum networking is no longer confined to laboratory benches. It can operate in the same ducts and conduits that carry the classical internet.
The next steps involve increasing the rate of entanglement swapping, extending the distances involved, and connecting more nodes. As networks grow, they will need automated systems that monitor and correct for environmental changes without human intervention. They will also need standardized interfaces so that devices from different organizations can interoperate. These challenges are significant, but the path forward is clearer than it was even a few years ago.
New York City may be one of the first places where a quantum internet begins to take shape. The density of institutions, the availability of fiber, and the concentration of potential users create a natural testbed. NYU already operates a local quantum network across Manhattan and is expanding it to additional campuses. The launch of the NYU Quantum Institute provides a hub for turning research into deployable technology. As one of the researchers noted, Manhattan is compact, and within a few miles there are hundreds of financial institutions that depend on secure communication. The groundwork being laid today could support technologies that become essential in the next decade.
The experiment in New York shows that the dream of a quantum internet is moving from theory toward practice. It demonstrates that entanglement can be created, preserved, and linked across a city using real infrastructure. It shows that physics can serve as a firewall, not as a metaphor but as a literal property of nature. And it suggests that the networks of the future may be built not only with glass and light, but with the strange and powerful rules that govern the quantum world.