The quantum internet is a real engineering pursuit, not science fiction, but it is still far from anything resembling the classical internet’s maturity. As of mid-2025, researchers have demonstrated small entanglement-based networks spanning a handful of nodes across metropolitan distances, run quantum key distribution over tens of kilometers of fiber, and tested core building blocks like quantum memories and entanglement swapping at telecom wavelengths. The vision is a global network that distributes entanglement between any two points on Earth, operating alongside the conventional internet to enable applications that classical communication provably cannot achieve.1Science. Quantum internet: A vision for the road ahead But almost every component still needs orders-of-magnitude improvement before that vision becomes practical, and the path from lab demonstrations to deployable infrastructure is lined with unsolved problems.
What the Quantum Internet Actually Is
The term “quantum internet” refers to a network that transmits quantum information, typically in the form of entangled photons, between distant nodes. Unlike classical data, which can be freely copied and amplified, quantum signals cannot be duplicated without destroying the information they carry. That restriction, rooted in what physicists call the no-cloning theorem, is simultaneously the quantum internet’s greatest feature and its biggest engineering headache. It means any attempt to eavesdrop on a quantum channel leaves detectable traces, which is the basis for quantum-secure communication. But it also means you cannot use conventional signal amplifiers to boost a weak quantum signal the way the classical internet boosts optical data across long fiber runs.2Europhysics Letters. A survey on advances of quantum repeater
Instead, the quantum internet needs an entirely different infrastructure stack: single-photon sources, quantum memories that can hold fragile quantum states long enough to be useful, quantum repeaters to extend range, and new protocols that look nothing like TCP/IP. A comprehensive survey of proposed quantum internet protocol stacks concluded that adapting the classical internet’s layered architecture to quantum networks is essentially impossible because of the fundamental differences in how quantum information behaves.3arXiv. Quantum Internet Protocol Stack: a Comprehensive Survey The protocol designs needed are a ground-up rethinking, not a retrofit.
How Far Signals Can Travel Without Help
The single biggest limitation facing quantum networking today is distance. Photons carrying quantum information get absorbed as they travel through optical fiber, and because you cannot copy and re-amplify them, the signal simply dies. Research has shown that without quantum repeaters, quantum communication becomes nearly impossible across fibers longer than about 20 kilometers because the fiber’s transmission efficiency drops below the critical threshold needed to maintain a useful quantum link.4PubMed. Restoring Quantum Communication Efficiency over High Loss Optical Fibers That number applies to raw quantum communication without any clever workaround protocols. Practical quantum key distribution systems have pushed somewhat farther by using high-performance detectors and optimized encoding. One recent system demonstrated secure key generation at kilobits-per-second rates over 50 kilometers of fiber, and managed to send encrypted data with information-theoretic security across 100 kilometers.5PubMed Central. Experimental coherent one-way quantum key distribution with simplicity and practical security
For intercontinental quantum links, fiber simply will not work no matter how good the repeaters get, at least not any time soon. The alternative is satellite-based transmission, where photons travel through the vacuum of space with far less loss. China’s Micius satellite demonstrated satellite-to-ground quantum communication and entanglement distribution across thousands of kilometers, proving the basic concept.6Reviews of Modern Physics. Micius quantum experiments in space Satellite links solve the distance problem but introduce their own complications: you need clear line-of-sight, atmospheric turbulence degrades signals, and satellite passes over any given ground station last only minutes.
On the fiber side, the challenge extends beyond raw loss. Metropolitan fiber networks introduce noise that degrades quantum states in ways that depend on how the information is encoded. Researchers have begun systematically measuring how different encoding methods, whether using the photon’s polarization, timing, or frequency, hold up across real-world metropolitan fiber, quantifying how each one drifts over time.7arXiv. Engineering Quantum Links: Noise and Quantum-State-Degradation Metrics over Metropolitan Fiber Network That kind of characterization work is essential groundwork, because a deployed quantum network will need to handle whichever encoding survives contact with real installed fiber.
The Repeater Problem
Classical internet infrastructure relies heavily on optical amplifiers spaced along fiber routes. The quantum internet needs an analog: quantum repeaters that can extend entanglement over long distances. The basic idea is to break a long link into shorter segments, generate entanglement across each short segment independently, and then stitch the segments together through a process called entanglement swapping. In principle, this lets you create entanglement between two nodes that are far too distant to share a direct quantum link.8Europhysics Letters. A survey on advances of quantum repeater
In practice, getting entanglement swapping to work well has been a slog. One recent experiment demonstrated teleportation of entanglement between time-bin qubits at a standard telecom wavelength, achieving a fidelity of 87 percent.9OSTI.GOV. Entanglement swapping systems toward a quantum internet That is encouraging for a lab demonstration, but a practical repeater chain would need to perform this process many times in succession while maintaining quality, and every imperfect swap compounds errors. No one has yet demonstrated a full working quantum repeater chain that extends entanglement beyond what a single fiber span can manage under real-world conditions.
Quantum Memories and Photon Sources
A quantum repeater only works if each intermediate node can store a quantum state while waiting for operations on other segments to complete. That requires quantum memories, devices that can capture a photon’s quantum state, hold it for some useful duration, and then release it on demand without corrupting the information.
Several different physical platforms are competing to serve as quantum network nodes. Researchers have explored single atoms, cold atomic ensembles, trapped ions, nitrogen-vacancy centers in diamond, and rare-earth-doped crystals, each with distinct advantages and drawbacks.10Laser & Photonics Reviews. Towards Real‐World Quantum Networks: A Review A room-temperature atomic memory achieved 25 percent end-to-end storage-and-retrieval efficiency for single photons, and when used to synchronize independently generated single photons, it produced a more-than-tenfold increase in the rate of useful photon-pair coincidences, reaching over a thousand detected synchronized pairs per second.11PubMed. Single-Photon Synchronization with a Room-Temperature Atomic Quantum Memory An earlier demonstration of a fast ladder memory in rubidium vapor stored 1.7-nanosecond pulses with the same 25 percent efficiency and a memory lifetime of 86 nanoseconds, while adding fewer than one noise photon per ten thousand retrievals.12PubMed Central. Fast, noise-free memory for photon synchronization at room temperature
Those numbers illustrate both the progress and the gap. A 25 percent round-trip efficiency means three-quarters of stored photons are lost, and an 86-nanosecond lifetime means the memory forgets its contents in a fraction of a microsecond. For a quantum repeater chain spanning hundreds of kilometers, you need memories that last milliseconds at minimum, and ideally much longer, while maintaining high efficiency. The technology is improving rapidly, but it is still several generations away from what practical networks demand.
On the photon-source side, the quantum internet needs emitters that produce exactly one photon at a time and do so at telecom wavelengths compatible with existing fiber infrastructure. Semiconductor quantum dots embedded in specially designed structures have demonstrated bright single-photon emission in the telecom O-band, with count rates up to 10 megahertz.13ACS Photonics. High-Performance Single-Photon Sources at Telecom Wavelength Based on Broadband Hybrid Circular Bragg Gratings A broader family of solid-state emitters at telecom wavelengths is under active development, and recent reviews suggest the field is converging on designs suitable for scalable quantum networks.14PubMed Central. Solid-state single-photon sources operating in the telecom wavelength range The key challenge remains pairing high brightness with high photon purity and doing so reproducibly at manufacturing scale.
Cleaning Up Noisy Entanglement
Even when entanglement is successfully generated and distributed, it is rarely pristine. Real-world noise from fiber imperfections, imperfect memories, and lossy interfaces degrades the quality of entangled states. This is where entanglement distillation comes in: protocols that take multiple copies of noisy entangled pairs and distill them into fewer pairs of higher quality.
The difficulty is that the very operations used to purify entanglement also inject noise, creating a frustrating chicken-and-egg problem. Recent work on IBM quantum devices tested distillation under realistic noise conditions and found that the technique broadly succeeds at improving average entanglement quality and removing coherent errors, though performance depends heavily on the noise model.15Physical Review Research. Basic entanglement distillation with realistic noise Theoretical work has also generalized purification protocols to more powerful error-correcting frameworks, showing improved fidelity in noisy-circuit simulations that incorporate realistic device imperfections.16PRX Quantum. Efficient Entanglement Purification Circuit Design for Dual-Species Atom Arrays Distillation is critical because without it, errors accumulate across a repeater chain until the output entanglement is useless. Getting distillation to work well in practice, not just in simulation, remains an open challenge.
The First Real Networks
The most convincing evidence that quantum networking is more than a lab curiosity comes from multi-node demonstrations over real fiber infrastructure. A team reported the first multi-node entanglement-based quantum network operating across a metropolitan area, with three quantum nodes equipped with atomic memories and telecom interfaces. They demonstrated heralded entanglement between two nodes 12.5 kilometers apart, stored entanglement longer than the time it takes light to make a round trip between the nodes, and showed concurrent entanglement generation across three separate links.17Nature. A multinode quantum network over a metropolitan area
That is a landmark result, but it is important to keep the scale in perspective. Three nodes across a city is a long way from a continental or global network. The entanglement rates achieved were low, the quantum memories had short lifetimes, and the whole system required extraordinary stabilization of the fiber links. These are proof-of-concept demonstrations. They prove the physics works at urban distances. They do not yet prove the engineering can scale.
What You Could Actually Do With a Quantum Internet
The most commercially mature application is quantum key distribution, which uses quantum channels to generate encryption keys whose security is guaranteed by physics rather than computational difficulty. QKD has already been deployed in real-world networks, and improved protocols continue to push the boundary of what is practical over metropolitan and intercity distances.18PubMed Central. Experimental coherent one-way quantum key distribution with simplicity and practical security But QKD is only one application, and arguably the least interesting from a long-term perspective.
Blind quantum computing lets a client delegate computations to a quantum server without the server ever learning what it is computing. Researchers demonstrated this experimentally over a decade ago, establishing that a client with minimal quantum capability can use a remote quantum computer while keeping inputs, operations, and outputs completely private.19PubMed. Demonstration of blind quantum computing As quantum cloud computing becomes more common, blind computation could become essential for users who need quantum processing power but cannot trust the provider with their data.
Distributed quantum computing is another major draw. Current quantum processors are limited in the number of qubits they can pack onto a single chip, and connecting multiple processors through quantum links could allow them to work as a single, larger system. Industry roadmaps have included plans for multi-chip quantum processors with communication links supporting inter-chip entanglement.20ScienceDirect (Elsevier). Distributed quantum computing: A survey This is not the same as a full quantum internet, but it is the near-term use case most likely to drive development of quantum networking technology.
Distributed quantum sensing uses entanglement shared across a network to achieve measurement precision that scales better than classical sensor arrays can manage. A recent experimental demonstration of a quantum-enhanced optomechanical sensor network showed a sensitivity advantage that grew with the number of sensors, confirming theoretical predictions.21Laser & Photonics Reviews. Quantum‐Enhanced Optomechanical Sensor Network Applications in navigation, timekeeping, and environmental monitoring could benefit substantially.
Quantum clock synchronization is a less flashy but potentially transformative application. Sharing a precise time reference between distant nodes is fundamental to navigation, telecommunications, and financial networks. Quantum approaches exploit entanglement to synchronize clocks with precision that can exceed what classical methods achieve, while also offering resistance to certain types of spoofing attacks.22arXiv. Quantum Clock Synchronization Networks: A Survey
QKD Versus Post-Quantum Cryptography
The security argument for quantum networking deserves a more nuanced look than it usually gets. QKD provides security based on physics: an eavesdropper disturbs the quantum states in detectable ways, so any interception is caught. But QKD requires dedicated hardware, specialized fiber or free-space links, and becomes impractical over long distances without repeaters that do not yet exist at scale. Post-quantum cryptography takes a completely different approach, using new mathematical algorithms that are believed to be hard for both classical and quantum computers to crack, and runs on existing classical infrastructure with no special hardware.
A critical analysis comparing deployed QKD systems with post-quantum cryptographic solutions concluded that the two approaches are suited to different scenarios. QKD excels in contexts where information-theoretic security is essential and infrastructure investment is justified, while post-quantum cryptography offers easier scalability and integration with existing high-density networks.23PubMed Central. A critical analysis of deployed use cases for quantum key distribution and comparison with post-quantum cryptography Research into hybrid systems that combine both approaches is underway. One experimental demonstration integrated QKD and post-quantum cryptography into a single framework, leveraging QKD’s physics-based security for key generation and post-quantum algorithms for scalable authentication.24Advanced Quantum Technologies. Experimental Integration of Quantum Key Distribution and Post‐Quantum Cryptography in a Hybrid Quantum‐Safe Cryptosystem
The honest assessment is that for most organizations worried about quantum computing breaking their encryption, post-quantum cryptography is the near-term practical answer. QKD matters for the subset of use cases where adversaries might record encrypted traffic today to decrypt it once quantum computers mature (the “harvest now, decrypt later” threat), and where the data is so sensitive that even a theoretical future vulnerability is unacceptable. Government and military communications, certain financial transactions, and critical infrastructure control systems fall into this category. For everyday internet security, the classical internet upgraded with post-quantum algorithms is the realistic path forward for the next decade or more.
Why the Timeline Is So Uncertain
Predictions about when a functional quantum internet will exist vary wildly, from optimistic industry roadmaps suggesting initial utility within a decade to academic assessments that place a full-scale network further out. The uncertainty stems from the fact that the field faces not one bottleneck but several simultaneous ones. Quantum memories need to improve in both storage time and efficiency. Quantum repeaters need to work reliably in chains. Single-photon sources need to be manufactured reproducibly at telecom wavelengths. Entanglement distillation needs to work under real device noise. And entirely new protocol stacks need to be designed, tested, and standardized.
Each of these problems is individually solvable in principle. The challenge is that a working quantum internet needs all of them solved at once and integrated into a single system where every component is compatible with every other. The history of the classical internet shows how long that integration takes even when the individual technologies are comparatively well understood. Quantum networking adds fundamental physical constraints that make the integration harder.
Staged development is the realistic model. Early stages, achievable with near-term technology, include point-to-point QKD links, small trusted-node networks, and intra-campus or intra-data-center quantum links connecting nearby quantum processors. Later stages involve untrusted repeater chains, long-haul entanglement distribution, and eventually a fully connected network where any node can share entanglement with any other. Each stage enables some applications while falling short of others. The field is currently transitioning from single-link demonstrations to the earliest multi-node stage.
Quantum Networking for Connecting Processors
Interestingly, the application most likely to drive quantum networking investment in the near term has little to do with communication security. As quantum processors grow, they hit physical limits on how many qubits can fit on one chip while maintaining quality. Linking multiple smaller processors through quantum interconnects could create effectively larger quantum computers without requiring any single chip to do everything. This distributed quantum computing model is already on industry roadmaps, with plans for multi-chip systems where processors are linked by quantum communication channels.25ScienceDirect (Elsevier). Distributed quantum computing: A survey
The distances involved are short, maybe within a single building or across a campus, which sidesteps the fiber-loss and repeater problems that plague long-distance quantum communication. But the fidelity requirements are extreme. A distributed quantum computer needs entangled links that are good enough to run quantum error correction across chips, which is a higher bar than QKD or sensing applications demand. So while the distance problem is easier, the quality problem may be harder. This corner of quantum networking is where some of the most intense engineering effort is concentrated right now, partly because the payoff is clear and partly because quantum computing companies are the ones funding it.
How Deployed Fiber Actually Behaves
Lab experiments over clean, short fiber spools give an optimistic picture of quantum channel performance. Deployed metropolitan fiber is a different beast. It runs through underground conduits where temperature fluctuates, it bends around corners in ways that affect polarization, and it shares physical pathways with high-power classical signals that create background noise. The gap between lab and field is substantial, and bridging it requires painstaking characterization work that is less glamorous than record-setting entanglement demonstrations but arguably more important for real deployment.
Recent field studies have measured how quantum states degrade across actual metropolitan fiber for different encoding strategies, tracking not just instantaneous error rates but how those rates drift over hours and days.26arXiv. Engineering Quantum Links: Noise and Quantum-State-Degradation Metrics over Metropolitan Fiber Network Time-bin encoding, where information is carried in when a photon arrives rather than its polarization, appears more robust against the kinds of disturbance that deployed fiber introduces. Frequency encoding also shows promise. These engineering-level details will determine which quantum networking approaches survive contact with reality, even if they rarely make headlines.