THE MACHINE THRESHOLD
Quantum / EXPLAINER / 2 MIN READ + OPTIONAL DEEP DIVE

Why making a quantum state of light is a milestone of its own

A chip-based source creates a useful kind of quantum building block. Turning those blocks into a fault-tolerant computer takes more work.

AI-assisted synthesis · Published 2026-09-17 · Updated & sources checked 2026-09-17
How we research and correct our work

The first challenge is producing the right state of light. That success has its own evidence—and its own limits.

Start with the right kind of light

In a June 2025 Nature paper, a Xanadu team reported an integrated photonic source of approximate Gottesman–Kitaev–Preskill, or GKP, states. These are structured quantum states that can encode a qubit. The experiment addresses the creation of those states, a building block for a proposed computing architecture. [1]

A conditional source of quantum light
  1. Prepare and mix light
  2. Count three outputs
  3. Identify the remaining state
Mechanism guide · lower loss is still needed

A detector pattern announces a candidate

The experiment combines four prepared light modes on a chip. Photon-counting detectors measure three outputs; selected count patterns signal a GKP state in the remaining output. This conditional announcement is called heralding. The team reconstructs the state from many measurements to examine its structure. [1]

The part and the finished machine

Hypothetical example: Imagine a workshop demonstrating that it can make a difficult component with recognizable features. That is useful evidence about the component. It does not, by itself, demonstrate the speed or reliability of a finished machine assembled from many such parts. This manufacturing analogy illustrates the distinction; a quantum state is not a literal mechanical part.

Loss remains a major boundary

The authors report features needed for fault tolerance, but their path to sufficiently high-quality states relies on simulations with much lower optical loss. They do not demonstrate a complete fault-tolerant computer. This is developer-authored research; we reviewed the paper, not an independent replication. [1]

Go a little deeper

Optional reading · about 1 more minute

Read the promise at the right level

Our interpretation: We would keep three questions separate: can the source make the intended state, can that state reach the required quality, and can a larger system use it reliably? Evidence for one question should not silently become the answer to all three. This is why a component result can matter without being a finished computer.

What would change the assessment

Our interpretation: The next evidence we would look for is measured improvement under lower loss and demonstrated integration with later stages of the architecture. A funding announcement, a manufacturing plan and a state-generation experiment answer different questions. This article is an explanation of the 2025 experiment, not a prediction of availability.

Original sources

Attributed synthesis, not original reporting. Examples labeled hypothetical or illustrative are explanatory. Reviewing a source does not independently validate its findings.

  1. Larsen and colleagues: Integrated photonic source of Gottesman–Kitaev–Preskill qubits ↗

    Published June 4, 2025. Accessible article read in Chrome September 17: abstract, main, experiment, discussion and author affiliations. Supplementary derivations not independently checked; no figures reused.

Suggest a correction

Know someone who would find this interesting?

Share this story on Facebook ↗ ·

Follow on Facebook ↗ for story highlights and questions to explore next.

Where this question leads next

Follow new explainers and updates →