China’s Quantum Networking and QKD — World’s Most Ambitious Quantum Communication Program

Introduction

On September 29, 2017, a video call connected two men separated by 7,600 kilometres. On one end, in Beijing, Bai Chunli, president of the Chinese Academy of Sciences. On the other, in Vienna, Anton Zeilinger, the physicist who would later win the 2022 Nobel Prize. The 75-minute call consumed roughly 560 kilobits of encryption key in total, refreshing AES-128 session keys every second. What made it historic was not the call itself but the source of those keys: photons beamed from a satellite named after an ancient Chinese philosopher, bounced between ground stations on two continents, carrying quantum states that no eavesdropper could intercept without detection.

That call was a demonstration. What China has built since is something else entirely — an operational, carrier-grade quantum communication network spanning over 12,000 kilometres of fiber, 145 backbone nodes across 80 cities in 17 provinces, connected to two quantum satellites (as of Jan 25, 2026 operates with Jinan-1 only as Micius decayed), and serving hundreds of government agencies, banks, and state-owned enterprises. No other nation has anything comparable. Whether it represents a visionary investment in information-theoretic security or an expensive hedge against a threat that post-quantum cryptography already addresses is a question I will examine carefully. But the engineering achievement itself is beyond dispute.

This article is the comprehensive reference I have been wanting to write for years — tracing every major deployment, satellite mission, protocol breakthrough, and record from the earliest metropolitan testbeds through early 2026. It is part of my ongoing series on China’s quantum technology initiatives.

The Infrastructure

The Beijing-Shanghai Backbone: Where It All Started

The story begins in July 2013, when construction started on the Beijing-Shanghai Quantum Communication Backbone Network — the Jinghu trunk line. Led by Pan Jianwei at the University of Science and Technology of China (USTC), it took 42 months to build: a 2,032 km fiber-optic QKD link connecting Beijing, Jinan, Hefei, and Shanghai through 32 trusted relay nodes and 135 QKD links.

The technical specifications matter because they define what “deployed QKD” actually looks like in practice. The backbone uses polarization-coded decoy-state BB84 protocol. Inter-node distances range from 34 to 89 km with fiber losses between 7.26 and 22.27 dB per link. The system multiplexes four quantum channels onto a single fiber and delivers 100 Gbps of classical bandwidth alongside the quantum channel. Hardware came from QuantumCTek — the USTC spinoff that would later become China’s first quantum technology IPO. The entire base project cost less than RMB 600 million (roughly $86 million) — remarkably modest for infrastructure of this scale.

When the backbone officially launched on September 29, 2017, it was the same day Pan’s team conducted the first intercontinental quantum-encrypted video call via the Micius satellite. The timing was deliberate — a statement of integrated capability.

Extensions followed quickly. The Shanghai-Hangzhou commercial trunk (260 km, ~$25 million) opened in October 2017 as China’s first commercial quantum communication line. A Wuhan-Hefei extension (609 km, 11 stations, ~200 million yuan) completed in November 2018. A Beijing-Guangzhou trunk was approved by the National Development and Reform Commission, with construction underway by 2019.

Metropolitan Networks: The City-Level Foundation

Beneath the backbone, metropolitan QKD networks were built in parallel across China’s major cities — each functioning as an access layer for the national infrastructure.

The Hefei metropolitan network is the best documented scientifically, evolving from a modest 3-node prototype in 2009 to a 46-node system that operated continuously for 31 months, with results published in Chen et al., npj Quantum Information 7, 134 (2021). The Jinan network reached up to 56 nodes by 2013, making it one of the earliest large-scale deployments. The Wuhan quantum network, launched in November 2017 with 71 nodes and a maximum key rate of ~141 kbps, was built by a subsidiary of China Aerospace Science and Industry Corporation — a signal that the military-industrial complex was fully engaged. Beijing deployed QKD links to secure communications for the 18th CPC Party Congress as early as 2012.

Each of these networks served dual purposes: operational security for government communications and testbed environments for advancing the technology. The pattern of building real infrastructure and then publishing the scientific results — rather than the reverse — distinguishes China’s approach from Western quantum networking programs, which have largely remained in the testbed phase.

The 4,600 km Integrated Network: Space Meets Ground

The definitive integration milestone came in January 2021, when Chen, Y.-A. et al. published in Nature 589, 214–219 a description of the world’s first integrated space-to-ground quantum communication network spanning 4,600 kilometres. This combined the 2,032 km fiber backbone with over 700 QKD links and two satellite-to-ground free-space links via Micius — one to the Xinglong ground station near Beijing, the other to Nanshan near Ürümqi — adding roughly 2,600 km of satellite coverage.

The network served more than 150 industrial users including banks, municipal power grids, and e-government platforms across the four metropolitan networks. For the first time, an end-to-end quantum-secured communication path existed from Shanghai to Ürümqi — a 4,600 km span requiring both fiber and satellite links, with keys relayed through a combination of fiber-optic trusted nodes and the satellite acting as an additional trusted relay.

12,000+ Kilometres: The Carrier-Grade Network

By 2025, the network had undergone another dramatic expansion. The China Quantum Communication Network (CN-QCN), described in Chen, H.-Z. et al., npj Quantum Information 11, 137 (2025), represents a genuine step-change: 145 fiber backbone nodes, 144 fiber links, and over 10,103 km of total fiber, with metropolitan access networks in 20 cities supporting more than 800 user nodes. Combined with earlier segments, total fiber mileage exceeds 12,000 km across 17 provinces and 80 cities.

The architecture follows a “two horizontal, two vertical” ring topology — Beijing-Jinan-Hefei-Wuhan east-west, and Shanghai-Hangzhou-Hefei-Nanjing north-south — providing redundancy that commercial telecommunications requires. Six satellite ground stations in Beijing, Shanghai, Guangzhou, Chongqing, Hainan, and Xinjiang link to the Jinan-1 microsatellite. The average inter-node distance is approximately 70 km with average attenuation of 18.61 dB per link.

The word “carrier-grade” in the 2025 paper’s title is significant. It means the network operates with the reliability, uptime, and service-level agreements expected of commercial telecommunications infrastructure — not as a laboratory demonstration. This is the network that China Telecom Quantum now operates commercially, serving 6.8 million quantum communication users across 40-plus cities.

Micius and the Satellite Program

The World’s First Quantum Communication Satellite

The Micius satellite (墨子号, named after the ancient Chinese philosopher Mozi) launched on August 16, 2016 from the Jiuquan Satellite Launch Center aboard a Long March 2D rocket. At roughly 635 kg in a sun-synchronous orbit at approximately 500 km altitude, it became the world’s first — and for six years, the only — quantum communication satellite. The mission cost roughly $100 million with a designed two-year lifespan, though it far exceeded its two-year design life, remaining operational until its orbital decay and reentry in late January 2026 — nearly a decade of service.

The payload is what matters: a quantum key communicator, an entangled photon source generating roughly 5.9 million entangled pairs per second at ~810 nm via a PPKTP crystal in a Sagnac interferometer, and a 300 mm Cassegrain telescope with ~10 µrad beam divergence. This hardware enabled three landmark papers in 2017 that fundamentally proved space-based quantum communication was possible.

Satellite-to-ground QKD (August 2017). Liao, S.-K. et al. in Nature 549, 43–47 demonstrated the first satellite-to-ground decoy-state QKD, generating approximately 300 kilobits of secure key during a single 273-second pass over the Xinglong ground station. The channel efficiency was roughly 20 orders of magnitude better than equivalent fiber at the same 1,200 km distance — a comparison that fundamentally justified the satellite approach.

Entanglement distribution over 1,200 km (June 2017). Yin, J. et al. in Science 356, 1140–1144 demonstrated satellite-based entanglement distribution to ground stations in Delingha (Qinghai) and Lijiang (Yunnan), violating Bell’s inequality under strict Einstein locality conditions. This was ten times the previous distance record and won the AAAS Newcomb Cleveland Prize — the oldest peer-voted prize for scientific papers in the United States.

Ground-to-satellite quantum teleportation (July 2017). Ren, J.-G. et al. in Nature 549, 70–73 achieved the first ground-to-satellite quantum teleportation over up to 1,400 km from the Ngari ground station in Tibet, teleporting six quantum states with an average fidelity of 0.80 ± 0.01.

Intercontinental QKD: Beijing to Vienna

The intercontinental demonstration followed in January 2018. Liao et al. in Physical Review Letters 120, 030501 reported satellite-relayed QKD between China and Austria across approximately 7,600 km. Micius acted as a trusted relay, performing separate QKD sessions with Chinese and Austrian ground stations and relaying XOR-combined keys. The system generated ~215 kbit keys daily with a ~1.5% error rate. This secured the 75-minute video conference between Bai Chunli and Zeilinger that I opened with.

A critical upgrade came in 2020. Yin et al. in Nature 582, 501–505 demonstrated entanglement-based QKD over 1,120 km between Delingha and Nanshan — achieving a secret-key rate of 0.12 bits per second without requiring a trusted satellite relay. This was the first time satellite-based QKD eliminated the trusted-relay assumption entirely, though the key rate was so low as to be primarily of proof-of-concept value.

Micius was also used for a China-Russia satellite QKD link spanning 3,800 km between Moscow and Ürümqi during 2021–2022, yielding 310 kbit of final key.

Jinan-1: The Microsatellite Revolution

The Jinan-1 microsatellite, launched July 27, 2022, represents a paradigm shift from flagship demonstrations to scalable infrastructure. With a payload of just ~23 kg (versus Micius’s ~200 kg) and portable ground stations under 100 kg (versus Micius’s ~13,000 kg ground equipment), the microsatellite reduced cost by roughly 45× while increasing key generation rates by two to three orders of magnitude.

Using a single laser diode at 850 nm with 625 MHz repetition rate, Jinan-1 transmits ~250 million quantum photons per second. The Nature paper reports up to 1.07 million bits of secure key generated in a single satellite pass. The microsatellite established real-time QKD with ground stations in Jinan, Hefei, Nanshan, Wuhan, Beijing, and Shanghai — and critically, Stellenbosch, South Africa, achieving the first quantum satellite communication link in the Southern Hemisphere and setting a 12,900 km intercontinental QKD record, published in Li, Y. et al., Nature 640, 47–54 (2025).

The significance is not the distance but the economics. At 45× lower cost per satellite and 100× lower ground station cost, microsatellite QKD becomes feasible as a constellation — and constellation-scale deployment is exactly what Pan Jianwei has announced: 2–3 additional LEO quantum satellites launching in 2025, a medium-Earth orbit satellite at ~10,000 km altitude by 2027, and a high-orbit satellite called “Dawn” in geostationary orbit above 35,000 km for daytime QKD with continuous communication windows. The stated goal is a global quantum communication service by 2027 and a constellation-based global quantum network by 2030.

Pushing the Protocol Limits

Twin-Field QKD: Breaking the 1,000 km Barrier

The fundamental rate-distance limit of point-to-point QKD is set by the repeaterless bound — key rates scale linearly with channel transmittance, making long-distance fiber QKD impractical. Twin-field QKD (TF-QKD), first proposed by Lucamarini et al. in Nature 557, 400–403 (2018), offered a way around this by scaling with the square root of transmittance (√η).

Chinese teams seized on TF-QKD and systematically drove it to record distances. The sending-or-not-sending (SNS) protocol, proposed by Wang Xiang-Bin at Tsinghua University in 2018, became the workhorse due to its tolerance of large misalignment errors. The progression tells the story:

In 2020, Chen, J.-P. et al. in Physical Review Letters 124, 070501 demonstrated SNS-TF-QKD over 509 km, exceeding the repeaterless bound by a factor of seven. In 2021, field deployments linked metropolitan areas at 428 km and 511 km over commercial fiber. The Guo Guang-Can group at USTC then achieved 833.8 km using an optimized four-phase TF-QKD protocol (Wang, S. et al., Nature Photonics 16, 154–161, January 2022), with channel loss exceeding 140 dB.

The crowning achievement came in May 2023, when Liu, Y. et al. in Physical Review Letters 130, 210801 reported TF-QKD over 1,002 km of fiber — the first time any QKD system crossed the thousand-kilometer barrier. Using pure silica core fiber (maximum 0.16 dB/km) and ultra-low-noise superconducting nanowire single-photon detectors (SNSPDs, ~0.02 Hz dark count rate) developed by the Shanghai Institute of Microsystem and Information Technology, the system achieved a key rate of 0.0034 bits per second.

That number, 0.0034 bps, deserves emphasis. It is far too slow for any practical application. But it demonstrated the physical feasibility of repeater-less fiber QKD at intercity scales, validating the protocol’s theoretical advantage over the repeaterless bound.

Speed Records: 115.8 Mb/s at Short Range

Distance records tell only half the story. For QKD to be practical, key rates must approach useful thresholds. In March 2023, Li, W. et al. in Nature Photonics 17, 416–421 demonstrated 115.8 Mb/s over 10 km of standard fiber — roughly 10× the previous world record. The system used a multipixel SNSPD with ultrahigh counting rate, an integrated polarization transmitter, and fast real-time post-processing. Keys were distributed up to 328 km on ultra-low-loss fiber, and the same group demonstrated over 47 kbit/s at 200 km using SNS-TF-QKD.

While 115.8 Mb/s is still orders of magnitude below the throughput of classical AES encryption (multiple gigabits per second), it begins to approach rates that could support real-time key refresh for high-bandwidth encrypted communications — a meaningful operational threshold.

Device-Independent QKD: The Gold Standard Reaches 100 km

Device-independent QKD (DI-QKD) represents the highest level of quantum cryptographic security — requiring no trust in the measurement devices themselves. Security is certified entirely through observed Bell inequality violations, making it immune to all device-level side-channel attacks. Previous DI-QKD demonstrations had been limited to roughly 700 meters.

In January 2026, Lu, B.-W. et al. published in Science a demonstration of DI-QKD over 100 km of spooled fiber using two ⁸⁷Rb single atoms trapped in optical tweezers, with quantum frequency conversion to telecom wavelength. The team prepared 1.2 million heralded Bell pairs over 624 hours, extracting finite-key secure rates at 11 km and demonstrating positive asymptotic rates at 100 km. This represents a roughly 3,000× improvement in attainable distance — a leap comparable in scale to the improvement from early fiber QKD demonstrations to the Beijing-Shanghai backbone.

As I analyzed in my coverage of this breakthrough, the 100 km DI-QKD result has implications beyond QKD itself: it demonstrates the kind of long-distance atom-photon entanglement and Bell state measurement capability that quantum repeaters will eventually require.

Measurement-Device-Independent and Continuous-Variable QKD

Two other QKD variants deserve mention for completeness. Measurement-device-independent QKD (MDI-QKD) — which removes all detector side-channel attacks by placing the measurement station between two users — saw a Chinese record of 404 km in 2016 (Yin, H.-L. et al., PRL 117, 190501), then the world record for all fiber QKD types. The first free-space MDI-QKD demonstration (19.2 km between buildings in Hefei) followed in 2021.

Continuous-variable QKD (CV-QKD), which uses coherent states rather than single photons and can operate with standard telecom detectors, reached 202.81 km (Zhang, Y.-C. et al., PRL 125, 010502, 2020). More recent work has pushed CV-QKD bit rates to 190 Mbps at 5 km and demonstrated 126.56 km with local local oscillator systems.

Chip-Based QKD: The Scalability Play

In February 2026, Wang Jianwei’s group at Peking University published in Nature a 20-user TF-QKD network based on integrated photonic quantum chips with an aggregate communication capability exceeding 3,700 kilometres — the first demonstration of a QKD network built entirely on photonic chips. This matters because chip-based systems offer a path to mass manufacturing and cost reduction that discrete-component QKD systems cannot match. If QKD is ever to move beyond government and banking applications to broader commercial deployment, it will likely need to follow this photonic chip route.

Quantum Repeaters and Memory — The Missing Piece

Why Repeaters Change Everything

The fundamental limitation of all current QKD networks — including China’s entire 12,000+ km infrastructure — is the trusted relay node problem. Every long-distance QKD link requires intermediate stations where quantum states are measured, keys exist momentarily as classical data, and then are re-encoded for the next hop. At each relay, the key is exposed to conventional cybersecurity threats: insider attacks, hardware tampering, software vulnerabilities. The quantum security guarantee applies only to individual point-to-point links, not to end-to-end communication.

Quantum repeaters would relay entanglement without ever exposing key material — eliminating the trusted node vulnerability entirely. They remain the missing piece for truly secure long-distance quantum communication, and China is investing heavily in building them.

The January 2026 Breakthrough: A Scalable Building Block

In January 2026, USTC researchers published what may be the most consequential result in this domain: the first scalable building block for a quantum repeater (Liu, W.-Z. et al., Nature, 2026). Using trapped-ion quantum memories with a high-efficiency ion-photon interface, the team achieved remote memory-memory entanglement over 10 km of spooled fiber with an entanglement lifetime of approximately 550 milliseconds — critically exceeding the average entanglement establishment time of ~450 ms.

This crossing of the “entanglement lifetime exceeds generation time” threshold is the fundamental requirement for scalable operation. It means entanglement can be stored in one link while additional links are being established, enabling eventual chaining across longer distances. The gap is narrow — 550 ms versus 450 ms — and the distance is only 10 km, but the principle has been demonstrated.

The Road from 10 km to 1,000 km

Earlier work built the foundation for this result. In 2020, Pan’s group entangled two cold-atom quantum memories over 50 km of fiber — extending the previous record of 1.3 km by roughly 40× (Nature 578, 240–245). A multiplexed quantum repeater using absorptive rare-earth-ion-doped crystal memories with ~80.4% entanglement fidelity was demonstrated in 2023 (Li et al., Nature). A metropolitan three-node quantum network with quantum memories separated by up to 12.5 km was demonstrated in 2024 (Liu et al., Nature 629, 579–585).

Quantum memory performance has improved dramatically alongside these networking results. Zhou Zongquan’s group at USTC achieved one-hour coherent optical storage in a europium-doped yttrium silicate crystal, maintaining fidelities above 96% at 5, 30, and 60 minutes (Nature Communications 12, 2381, 2021). Nuclear spin coherence times in the same material system reached 10 hours. Integrated solid-state quantum memories achieved 99.3% storage fidelity, and laser-written waveguide memories reached 1.021 ms storage (Science Advances, 2025).

These are impressive individual results. But chaining multiple repeater nodes into a functional long-distance network — with real-time classical feedback, error correction, and integration with existing fiber infrastructure — remains years away. The gap between a 10 km lab demonstration and a 1,000 km operational repeater chain is vast, and anyone who tells you otherwise is selling something.

The People and Institutions

Pan Jianwei: China’s “Father of Quantum”

Virtually every major Chinese quantum communication achievement traces back to USTC and specifically to Pan Jianwei (潘建伟). Pan completed his PhD under Anton Zeilinger at the University of Vienna, held postdoctoral positions at Innsbruck and Heidelberg, and returned to China in 2008. By 2011, at age 41, he became the youngest-ever member of the Chinese Academy of Sciences. Nature named him one of “10 People Who Mattered” in 2017.

Pan leads a core team that includes Chen Yu’ao (chief designer of the Beijing-Shanghai backbone), Peng Chengzhi (co-leader of satellite programs), Zhang Qiang (QKD protocols), and Bao Xiaohui (quantum memory and DI-QKD). His relationship with Xi Jinping is direct — in July 2013, just three months after the Snowden leaks, Xi met Pan to discuss quantum communications and visited QuantumCTek. The Snowden revelations convinced Chinese leadership that communications security required a quantum solution.

The Supporting Cast

A second major USTC group led by Guo Guang-Can — who founded CAS’s Key Laboratory of Quantum Information in 2001 and was instrumental in establishing early government funding — holds the 833.8 km TF-QKD record through his collaborator Han Zheng-Fu.

Beyond USTC: Tsinghua University contributes through Wang Xiang-Bin (inventor of the SNS protocol underpinning most TF-QKD records) and Gui-Lu Long (quantum secure direct communication pioneer). The Beijing Academy of Quantum Information Sciences (BAQIS), co-founded in 2017, supports quantum communication research. Peking University hosts Hong Guo’s group (CV-QKD records) and Wang Jianwei’s group (chip-based QKD). The Shanghai Institute of Microsystem and Information Technology (SIMIT) develops the ultra-low-noise SNSPDs that enable extreme-distance TF-QKD — hardware that is as important to China’s QKD records as the protocols themselves.

Government Strategy and Funding

China’s quantum communications program benefits from sustained, escalating government support spanning two decades. The 2006 National Medium- and Long-Term Science and Technology Development Plan elevated “quantum control” to one of four major scientific research areas. The 13th Five-Year Plan (2016–2020) designated quantum technologies as a national “megaproject.” The 14th Five-Year Plan (2021–2025) tripled attention to quantum, mentioning it six times. The 15th Five-Year Plan (2026–2030), approved in March 2026, elevates quantum technology to first among seven “future industries.”

In October 2020, Xi Jinping presided over a dedicated Politburo study session on quantum technology — an extraordinary signal of political priority at the highest level.

Funding estimates are substantial but contested. The National Laboratory for Quantum Information Sciences in Hefei occupies a 37-hectare campus with reported investment of up to $10 billion. U.S. Congressional reports estimate China’s total quantum R&D funding at approximately $15.3 billion, though USTC physicist Chao-Yang Lu has suggested actual spending may be roughly one-third of publicized figures. Even at one-third, China’s quantum spending exceeds U.S. government investment.

A massive new funding vehicle emerged in 2025: the National Venture Guidance Fund, financed through ultra-long-term government bonds. Three regional sub-funds specifically targeting quantum-related technologies were established in December 2025, totaling approximately RMB 121.8 billion (~$17.5 billion) across Beijing-Tianjin-Hebei, the Yangtze River Delta, and Guangdong-Hong Kong-Macao — though these cover multiple technology sectors, not quantum exclusively.

The Commercial Ecosystem

QuantumCTek: From USTC Spinoff to State-Telecom Subsidiary

The commercial ecosystem centers on QuantumCTek (科大国盾量子, stock code 688027), founded in 2009 as a spinout from Pan’s USTC group. The company went public on the Shanghai STAR Market on July 9, 2020 — the first quantum technology IPO in China — with shares surging 924% on the first day, the largest first-day gain in Chinese stock market history at the time. QuantumCTek manufactures QKD devices, quantum random number generators, quantum network switching equipment, and key management systems, and has led the drafting of over 100 quantum technology standards.

In early 2025, China Telecom’s Quantum Information Technology Group — established in Hefei in May 2023 with a 3 billion yuan (~$434 million) investment — became QuantumCTek’s controlling shareholder. This merger of state telecom muscle with quantum hardware capability has accelerated commercialization dramatically.

China Telecom Quantum: 6.8 Million Users

By May 2025, China Telecom launched the world’s first commercial hybrid QKD+PQC distributed cryptography system, completing a 1,000 km quantum-encrypted phone call between Beijing and Hefei and rolling out across 16 cities. The Hefei metropolitan quantum network alone spans 1,147 km of QKD fiber, 8 core nodes, and 159 access points, serving 500 government departments and 380 state-owned enterprises.

Products include “Quantum Secret” (secure messaging), “Quantum Cloud Seal” (document workflow), 5G quantum encryption with approximately 500,000 users, and quantum SIM cards reaching roughly 6 million users. China Telecom Quantum reported 65.4% quantum revenue growth in 2025, serving 6.8 million quantum communication users across 40-plus cities with 5,000-plus industry customers.

The Broader Ecosystem

Qasky (安徽问天量子), founded in 2016 by Guo Guang-Can’s group in Wuhu, provides provincial and enterprise-level QKD equipment. XT Quantech, founded by Shanghai Jiao Tong University professor Zeng Guihua, has explored quantum communication integration with high-speed rail systems. China Mobile and China Unicom have both launched quantum-secure leased lines and positioned quantum communications as part of next-generation infrastructure.

The Complete Record Book

The progression of Chinese world records in quantum networking and QKD tells a story of systematic, accelerating capability expansion:

2016: MDI-QKD over 404 km — world record for all fiber QKD types (Yin et al., PRL). Micius satellite launch.

2017: Satellite-to-ground QKD, ~300 kbit per pass over 1,200 km (Liao et al., Nature). Entanglement distribution over 1,200 km (Yin et al., Science). Ground-to-satellite teleportation over 1,400 km (Ren et al., Nature). Beijing-Shanghai backbone operational.

2018: Intercontinental QKD, China-Austria, ~7,600 km via Micius (Liao et al., PRL).

2020: SNS-TF-QKD over 509 km, breaking repeaterless bound (Chen et al., PRL). Entanglement-based satellite QKD over 1,120 km without trusted relay (Yin et al., Nature). CV-QKD record of 202.81 km (Zhang et al., PRL). Cold-atom memory entanglement over 50 km fiber (Nature).

2021: Integrated space-ground network, 4,600 km, 150+ users (Chen et al., Nature). One-hour coherent optical storage (Nature Communications).

2022: TF-QKD over 833.8 km (Wang et al., Nature Photonics). Jinan-1 microsatellite launch.

2023: QKD bit rate record of 115.8 Mb/s at 10 km (Li et al., Nature Photonics). TF-QKD over 1,002 km (Liu et al., PRL). Multiplexed quantum repeater with absorptive memories (Li et al., Nature).

2024: Metropolitan quantum memory network, 3 nodes over 12.5 km (Liu et al., Nature).

2025: Microsatellite intercontinental QKD over 12,900 km (Li et al., Nature). CN-QCN carrier-grade network, 10,000+ km, 145 nodes (Chen et al., npj Quantum Information).

2026 (February): Scalable quantum repeater building block, 10 km with 550 ms entanglement lifetime (Liu et al., Nature). DI-QKD over 100 km (Lu et al., Science). 20-user chip-based TF-QKD network spanning 3,700 km (Wang et al., Nature).

How China Compares to Everyone Else

China’s position in quantum communications is unambiguous in scale. Its 12,000+ km operational network dwarfs all other deployments combined.

The EU’s EuroQCI program, signed by all 27 member states, aims for a pan-European quantum communication infrastructure by 2027 but remains in the testbed and cross-border interconnection phase. ESA’s Eagle-1 — Europe’s first QKD satellite — targets launch in late 2026 or early 2027, a decade after Micius. The United States has no comparable national QKD infrastructure program, instead pursuing fundamental R&D through DOE testbeds like the Chicago Quantum Loop (80-mile, three-node) and Brookhaven-Stony Brook network, with total quantum networking funding of roughly $61 million. South Korea has the most substantial non-Chinese deployment: an 800 km, 48-node government network completed in 2022, leveraging SK Telecom’s acquisition of ID Quantique. Japan’s NICT-led Tokyo QKD network has operated since 2010 at smaller scale. Singapore’s National Quantum-Safe Network focuses on enterprise-grade hybrid QKD+PQC.

In commercial QKD hardware, Toshiba (Japan/UK) leads in key rates and telecom integration, and ID Quantique (Switzerland/South Korea) remains the global market leader in commercial systems. BT and Toshiba demonstrated the world’s first commercial quantum-secured metro network in London in 2022. These are strong capabilities — but they are products, not infrastructure. The gap between building good QKD hardware and operating a 12,000 km carrier-grade network is enormous.

The most significant area where China trails is post-quantum cryptography standardization. NIST has finalized three PQC standards (ML-KEM, ML-DSA, SLH-DSA) with HQC as backup, while China’s national PQC algorithm standards have not yet been publicly released. The U.S. has a clear PQC migration timeline with government-wide completion targeted by 2035. China is actively working to close this gap, and China Telecom’s hybrid QKD+PQC system reflects a dual-track strategy — using QKD as an “enhanced defense line” while PQC serves as the “basic defense line.”

My Analysis — The QKD Debate and What This Means

The Trusted Node Problem Is Real

I want to be direct about something: China’s quantum communication network is genuinely impressive engineering, but its security still fundamentally depends on classical security at relay nodes. Every trusted relay in the 12,000 km network is a potential attack point where keys exist as classical data. A single compromised relay breaks the quantum security advantage for any communication path that traverses it.

China knows this. The January 2026 quantum repeater demonstration targets exactly this limitation. But chaining multiple repeater nodes into functional long-distance infrastructure is years away. Until then, the operational security of the network rests on physical security at relay sites — which, in China’s case, reportedly includes military bases and government communications laboratories. That is good classical security, but it is not the information-theoretic security that QKD promises.

Why Countries Disagree on QKD

As I analyzed in detail in my article on why countries differ on QKD’s future, the global divergence in QKD policy is one of the most fascinating strategic disagreements in cybersecurity.

The NSA has stated unequivocally that it “does not support the usage of QKD” for protecting national security communications. The UK’s NCSC similarly does not endorse QKD for government or military applications. ANSSI (France), BSI (Germany), and agencies in the Netherlands, Sweden, and the Czech Republic have expressed similar preferences for PQC. Their argument is straightforward: QKD requires expensive specialized hardware, cannot provide authentication or digital signatures, must be combined with PQC anyway, and the bit rates remain impractical for high-bandwidth applications. PQC, by contrast, can be deployed as a software update to existing infrastructure.

The counterargument rests on information-theoretic security — QKD’s security is guaranteed by the laws of physics rather than computational hardness assumptions. The cracking of SIKE (a NIST PQC finalist candidate) on a single classical computer demonstrated that mathematical security assumptions can fail catastrophically. If lattice-based PQC were to fail, countries with operational QKD infrastructure would have an advantage that could not be quickly replicated.

Most programs outside the U.S. and UK — including the EU, China, South Korea, Japan, and Singapore — pursue hybrid QKD+PQC as defense in depth. China’s own approach confirms this: China Telecom’s commercial system combines both. That is a pragmatic position that I have considerable sympathy for, even as I maintain that PQC migration should be every organization’s first priority.

Implementation Vulnerabilities Deserve More Scrutiny

Beyond the trusted node problem, I am concerned about the gap between QKD’s theoretical security proofs and real hardware. Multiple demonstrated attacks exploit this gap: electromagnetic emanation attacks recovering nearly 100% of keys from detector emissions, injection-locking attacks on laser sources, Trojan horse attacks on modulators, and detector blinding attacks. A 2025 analysis of Micius’s China-Russia QKD data found 98.7% distinguishability between signal and decoy states due to ~300 ps timing mismatches — a side-channel vulnerability in deployed hardware.

No globally recognized certification standard for QKD devices yet exists. This is a problem. If QKD is to be taken seriously as a security technology, it needs the kind of rigorous third-party evaluation that cryptographic algorithms receive through processes like NIST standardization. The current situation — where QKD hardware is manufactured and deployed by nationally-controlled entities without independent security certification — should make security professionals uncomfortable, regardless of how elegant the physics is.

The Strategic Dimension: It Is Not Just About Cryptography

Where I part company with QKD skeptics is on the strategic question. As I discussed in my broader analysis of China’s quantum technology initiatives, dismissing China’s QKD program as a misallocation of resources misses the larger picture.

The engineering capabilities built through quantum communication — single-photon detection, entanglement generation and distribution, quantum memory, real-time feedback control, cryogenic engineering, space-qualified quantum hardware — transfer directly to quantum computing and quantum sensing. The SNSPD technology that enables 1,002 km TF-QKD is the same technology needed for optical quantum computing readout. The ion-trap quantum memories being developed for repeaters are essentially the same hardware platform as trapped-ion quantum computers. The satellite quantum links could enable distributed quantum computing or quantum sensor networks.

China’s QKD program has created a quantum engineering ecosystem — manufacturing capability, trained personnel, institutional knowledge, supply chains — that has strategic value regardless of whether QKD itself proves to be the long-term communications security solution.

Related Content

How ECC Became the Easiest Quantum Target

A New Algorithm Shrinks the Quantum Attack Surface for ECC

Shor’s Algorithm: A Quantum Threat to Modern Cryptography

Post-Quantum Cryptography (PQC) Standardization – 2025 Update