¿Es posible encadenar computadoras cuánticas? Los investigadores tienen una respuesta

Escalando la computación cuántica

Quantum computers are expected to one day become the core technology for advanced computation within biotechnology, material sciences, finances, cryptography, and other complex systems that silicon-based computers struggle to deal with.

Sin embargo, dos cosas han estado frenando a las computadoras cuánticas para que se vuelvan más comunes.

Lo primero es que, cuántos más qubits (el equivalente cuántico a un bit binario) haya en un sistema, mayor será el error, y cualquier inestabilidad provoca un efecto dominó.

Esto podría pronto ser un problema del pasado, gracias a los recientes avances logrados por Google (GOOGL ) y su chip Willow, que logra una tasa de error decreciente a medida que se añaden más qubits.

El otro problema es que probablemente exista un límite de tamaño para cuán grande puede llegar a ser una computadora cuántica individual, de la misma manera que un chip de silicio solo puede volverse tan potente antes de enfrentar problemas de suministro de energía, sobrecalentamiento, etc.

Esto también podría estar resolviéndose, gracias al trabajo de investigadores de la Universidad de Oxford. Lograron crear una red de computación cuántica distribuida, similar a cómo funcionan los servidores con chips de silicio, potencialmente allanando el camino para supercomputadoras cuánticas.

They announced their results in the prestigious publication Nature, under the title “Distributed quantum computing across an optical network link”¹.

Conectando computadoras cuánticas

The way quantum computers currently work is as isolated units, containing a few hundred or maybe a thousand qubits for the most advanced ones.

El problema es que ampliarlas es increíblemente complejo, ya que cuanto más grande es el sistema, más inestabilidad y perturbaciones del entorno pueden interrumpir los efectos cuánticos de los que depende la computadora.

Al mismo tiempo, los cálculos útiles probablemente requerirán millones de qubits, lo que llevaría a la industria a un posible callejón sin salida. Al menos, mientras las computadoras cuánticas no puedan conectarse entre sí.

Y esto es más fácil decirlo que hacerlo porque las computadoras cuánticas son muy diferentes a los chips de silicio. En las computadoras tradicionales, el 0 & 1 es una señal eléctrica que puede transmitirse a otra máquina relativamente fácil.

For quantum computers, it is the very quantum state of the qubits that need to be communicated. These quantum states are difficult to keep stable out of very specific conditions, usually involving ultra-low temperatures barely a few degrees above absolute zero.

Aprovechando la teletransportación cuántica

Experimento previo de teletransportación cuántica

En enero de 2025, informamos la primera indicación de que la teletransportación cuántica en fibra óptica era una posibilidad. Apareció que incluso podría llevarse a cabo en redes de fibra óptica ordinarias junto a flujos de datos “normales”.

Si bien suena como un concepto fantasioso de una película de ciencia ficción, la teletransportación cuántica es en realidad un fenómeno real estudiado durante décadas.

Esto ocurre cuando 2 partículas diferentes están “emparejadas/vinculadas” entre sí, algo llamado entrelazamiento cuántico.

En este caso, cuando dos partículas están enlazadas, sin importar la distancia entre ellas, intercambian información a grandes distancias — sin transportarla físicamente. En algunos casos, incluso podría ser posible que el intercambio de información ocurra más rápido que la velocidad de la luz, algo teóricamente imposible.

Este fue un paso muy importante, ya que indicó que no necesitaríamos nueva tecnología para que los cables realicen teletransportación cuántica entre computadoras cuánticas.

But it was not yet a direct connection between 2 quantum computers.

Conectando qubits

This is the step achieved by the Oxford researchers. They used optical fibers to connect together qubits and get them entangled, using photons (light particles).

Fuente: Nature

In the lab experiment, the modules were separated by about 2 meters, each containing dedicated networks and circuit qubits. But longer distances could also be achieved.

By carefully tailoring these interactions, we can perform logical quantum gates – the fundamental operations of quantum computing – between qubits housed in separate quantum computers.

This breakthrough enables us to effectively “wire together” distinct quantum processors into a single, fully connected quantum computer.

Dougal Main – Graduate Student at Ion trap quantum computing research group at Oxford

Fuente: Nature

The type of quantum computer used was trapped-ion modules. Trapped-ion quantum computers generally contain a lot fewer qubits, but with much higher reliability and efficiency, with a similar amount of qubits achieving 100-1,000 fold more reliable results than other quantum computing methods.

Fuente: Quantinuum

Alta eficiencia

The entanglement between the networked qubits achieved an 86% fidelity.

When performing Grover’s search algorithm, a useful benchmark of computing where quantum computers radically outperform normal computers, the success rate was 71%.

Grover’s search algorithm searches for a particular item in a large, unstructured dataset much faster than a regular computer can, using the quantum phenomena of superposition and entanglement to explore many possibilities in parallel.

This is remarkably high, especially when considering it was a first-ever achievement before any optimization or improvement to the method was done.

Impacto de este descubrimiento

¿Supercomputadoras cuánticas?

This opens the way to an entirely different way of building quantum computers, imitating how supercomputers are built from separate subsystems that communicate their calculations to one another.

Such a system would also be remarkably modular, instead of needing an increasingly complex massive centralized quantum computer.

By interconnecting the modules using photonic links, the system gains valuable flexibility, allowing modules to be upgraded or swapped out without disrupting the entire architecture.

Dougal Main – Graduate Student at Ion trap quantum computing research group at Oxford

Fuente: Nature

Resurgimiento de la tecnología de iones atrapados

Until now, trapped-ion Technology was the more reliable one, and the one closest to practical use, but limited by the difficulty of adding more qubits to a given system. For this reason, it was expected that other systems using superconducting qubits would be the final form of quantum computer.

If trapped-ion technology can overcome this limit through improved networking, this might not be true anymore. This would make the companies that have specialized in this field much more interesting to investors.

Empresa de computación cuántica de iones atrapados

IonQ

IonQ is a quantum computing company using trapped-ion technology, founded by pioneering scientists in the field from the University of Maryland and Duke University. It was publicly listed on the NYSE in 2021.

IonQ quantum computing platforms are able to produce a 99.9% fidelity result. It currently uses a 64-barium ion chain, producing a 36-algorithmic qubit (AQ). The chain organization allows for much quicker computing than other trapped-ion designs without losing fidelity.

Fuente: IonQ

IonQ acquired Qubitekk in enero de 2025, adding to its operations the company’s team and 118 patents to IonQ. Qubitekk’s specialty is in quantum networks, using photonic interconnects, enabling quantum clusters, and advancing quantum internet capabilities.

Quantum networks should facilitate highly secured communications and ultimately allow for distributed quantum computing. Considering how quickly the field is moving, expertise and IPs on this topic might prove crucial for IonQ’s future.

IonQ is also developing a partnership with NKT Photonics (NKT.CO) to help develop future data center-ready quantum computers.

It is also collaborating with Imec on photonic integrated circuits and chip-scale ion trap technology to scale up the company’s qubit count and system size and costs.

Instead of developing its own SDK (Software Development Kit), the company is supporting all the major ones at once, and partnering with many leading companies for developing new quantum computing applications.

Fuente: IonQ

IonQ is the closest to a pure quantum computing stock for investors who are uninterested in the main activities of other leaders like Google, Intel (INTC ), IBM, or Honeywell.

So together with its competitor Quantinuum, part of Honeywell (HON ), IonQ is closer to developing commercial quantum computers, with its focus on high fidelity, lower qubit count trapped-ion systems.

Its early success has helped it build a strong network of partnerships with other quantum computing innovators to keep pushing this technology forward, with a recent re-focus on networked quantum computers.

Referencia del estudio:

^{1. Main, D., Drmota, P., Nadlinger, D.P., et al.(2025) Distributed quantum computing across an optical network link. Nature 638, 383–388. https://doi.org/10.1038/s41586-024-08404-x}