Sweden’s month-long quantum-key test shows why distance is not enough

Researchers at Sweden’s KTH Royal Institute of Technology report a month of quantum-key generation across an 80.85-kilometer fiber link. The useful finding is not simply that the connection worked. It is that the real cable and the equipment attached to it must be assessed together before a quantum-network promise becomes an operating service. Quantum key distribution, or QKD, uses quantum signals to share encryption keys; it is not a replacement for the ordinary data traffic those keys protect.^[1]

The study, submitted November 9 as a preprint, used a commercial ThinkQuantum Quky Edu-Pro system on a dedicated, rented fiber strand between Stockholm and Norrtälje. The authors report positive secret-key generation throughout the test, averaging 2.4 millionths of a key bit per transmitted pulse. That is not an internet download speed, and the paper is not evidence that Sweden has completed a national secure network. The paper does not specify the calendar dates of the month-long test.^[1]

From a route length to a useful connection

The link’s measured loss was 19.32 decibels. Converting that logarithmic number into an ordinary fraction gives about 1.17 percent transmission: roughly one photon in 86 would survive the fiber loss, before additional receiver losses. This calculation describes optical transmission, not the fraction of pulses that become usable secret-key bits.^[1]

That distinction matters when reading performance claims. A distance describes where the endpoints can be. A loss figure describes how much signal remains. A secret-key rate describes the output after a protocol has done its work. None can substitute for the others.

The reported key-basis error rate was 4.2 percent. The authors associate a later rate reduction with increased errors and suggest automatic realignment as a possible cause, not a proven diagnosis. Their simplified feasibility model omits several practical penalties, including finite data samples and detector imperfections.^[1]

For a buyer, the practical question is therefore more demanding than “Does this box reach 80 kilometers?” It is whether a particular box, on the actual available fiber, produces enough usable keys under the conditions the application requires.

A testbed, not a new network launch

KTH announced the inauguration of its pilot quantum-communications facility in February 2025. That facility was designed to let companies, public authorities and other organizations test systems, including equipment they bring themselves. The new preprint should be read as operational evidence from that research infrastructure, not as a newly announced construction project.^[2]

The distinction also prevents a misleading comparison with ordinary network rollouts. A successful experiment is valuable evidence for the next deployment decision. It is not, by itself, a guarantee of service availability or an independent certification of an entire security system.

Quantum hardware is not the whole security transition

Europe is pursuing another response to the quantum threat: post-quantum cryptography, which uses mathematical algorithms intended to resist both classical and quantum attacks. The European Commission’s current guidance calls for member states to begin that transition by the end of 2026, with high-risk uses addressed as soon as possible and no later than the end of 2030.^[3]

That is a separate workstream from installing quantum-key hardware. The Swedish result does not remove the need to migrate vulnerable cryptography. Its contribution is more specific: evidence that a real fiber connection can support key generation, plus a clearer way to ask what the next connection will require.