Microsoft (MS) has announced an innovative technology that could change the landscape of the quantum computer market. The unveiling of the world's first quantum chip utilizing "topological qubits," known as "Majorana 1," is expected to be a significant turning point in the competition for quantum computers led by Google and IBM.
The Majorana 1, unveiled on the 19th (local time), is the size of a palm. While MS revealed a prototype equipped with 8 topological qubits this time, they stated that the ultimate goal is to pack 1 million qubits into a chip small enough to hold in one hand.
MS noted, "We consider the integration of over 1 million qubits as the starting point for the 'commercialization of quantum computers,'" adding, "With the development of this chip, the era of quantum computers may become a reality within years, not decades."
◇Solving qubit instability with topological mathematics principles
Conventional computers represent the absence or presence of electrons as 0s and 1s, i.e., in 1-bit units. In contrast, the unit of a quantum computer is a qubit, which exists in a superposition of the 0 and 1 states. This is because, in quantum mechanics, which describes the microscopic world, matter can exist in multiple overlapping states rather than just one.
Quantum computers are known as "dream computers." While a classical computer with 2 bits can represent one of four states—00, 01, 10, or 11—2 qubits can represent all four simultaneously. If there are 300 qubits, it allows for 2 raised to the power of 300 states, which is more than the number of atoms in the universe, significantly enhancing computational capability. Google has already demonstrated this, solving a problem that would take a supercomputer 10,000 years in just 3 minutes using a 53-qubit quantum computer in 2019.
Although quantum computers theoretically offer enormous potential, practical application has been challenging due to the instability of qubits. Qubits are prone to errors due to minor changes in external environments. Overcoming this requires complex error correction techniques and a considerable amount of additional computational resources.
To overcome this limitation, MS has been developing topological qubits since the early 2000s. Topological mathematics studies properties that remain unchanged even when the shape of an object is altered. For instance, a doughnut and a coffee cup are considered topologically identical due to the common feature of having one hole in their surfaces. Likewise, topological qubits are protected by their topological properties, offering strong resistance to external influences. They can be viewed as qubits with a kind of protective shield.
The key technology that made this possible is a new type of material called "topoconductor." A topoconductor is a material that combines the properties of semiconductors and superconductors, allowing it to control electric current like a semiconductor while maintaining the movement of electrons without distortion like a superconductor, thus making qubits more stable. It functions like a library equipped with soundproof walls, securely storing information without interference from external noise.
The MS research team created the topoconductor by combining indium arsenide semiconductor and aluminum superconductor. They successfully manipulated conditions of extreme cold and magnetic fields to generate particles known as "Majorana zero modes." In conventional superconductors, electrons move in pairs, but in Majorana modes, the paired particles are spatially separated. As a result, if one particle encounters an issue, the other remains unaffected, demonstrating resilience to external disturbances and stable retention of quantum information.
Song Young-ik, a professor of electrical and electronic engineering at Korea Advanced Institute of Science and Technology (KAIST), said, "It is remarkable that a method once thought to be only theoretically possible has been experimentally validated," and added, "The Majorana qubit was a difficult technology that wasn’t even included among the five quantum computer platforms projected for commercialization success in the McKinsey report, but this demonstrates the potential for its realization."
MS has also introduced a simple error correction method utilizing digital signals. They read and adjust the quantum state of nanowires connected to Majorana particles using microwaves. In existing quantum computers, qubits had to be finely tuned using analog methods. This approach simplifies the process of detecting and correcting errors compared to traditional methods, achieving an accuracy rate of less than 1% with just one measurement.
MS explained, "Traditional quantum computers require complex algorithms and a large number of additional qubits for error correction, but Majorana 1 inherently suppresses errors due to its physical properties, which makes it highly stable and scalable," and added, "We expect to realize a quantum computer capable of solving meaningful problems within years, not decades."
◇Taking a different path from Google and IBM
Google and IBM, which have led quantum computer development, are using superconducting qubits. Superconductivity refers to the phenomenon where certain materials exhibit zero electrical resistance at extremely low temperatures, allowing electric currents to flow without loss. They store information in quantum computers using electrons in a superconducting state.
Superconducting qubits are the most widely studied and used method to date; however, they are sensitive to small changes in external environments, such as light or vibrations, necessitating algorithms for error correction. As a result, corporations, including Google, are focusing on increasing the number of qubits while minimizing quantum errors.
Last year, the 105-qubit superconducting quantum chip "Willow" unveiled by Google exhibited that as the number of qubits increases, errors decrease exponentially. The supercomputer Frontier, ranked second in computational speed worldwide, solved calculations that would take 10 zettayears (1 zettayear equals 10 to the power of 24 years) in under 5 minutes. IBM unveiled the "Quantum Heron", a 399-qubit quantum chip with three 133-qubit processors connected, last November. They plan to develop a 2000-qubit quantum computer by 2033.
The quantum chip Majorana 1 announced by MS this time has only 8 qubits, making it incomparable to Google and IBM's chips in terms of the number of qubits. However, MS's announcement could serve as a turning point that changes the competition structure of quantum computers, which has predominantly been based on the number of qubits. Song Young-ik, a KAIST professor, explained, "Once a topological qubit is successfully created, it becomes much easier to increase the number of qubits thereafter," adding, "If even one qubit is made correctly, expanding to 1 million qubits is just a matter of time."
George Booth, a professor at King's College London, stated, "Corporations generally measure development progress by the number of qubits, and by that measure, MS seemed to lag behind other corporations," but added, "MS has developed a system more resilient to environmental interference, focusing on the long game."
Korea has developed a 20-qubit quantum computer through the Korea Research Institute of Standards and Science (KRISS) using a superconducting method and is currently working to increase the number of qubits. Sung Eun-jung, head of the Quantum National Technology Strategy Center at KRISS, stated, "There have already been successful cases using superconducting methods from companies like Google and IBM, and we have a foundation for superconducting technology domestically," adding, "We're researching ion traps, solid-state couplings, and photon methods at the lab level, but finding researchers related to topological methods is quite difficult in the country."
However, there is also a skeptical view regarding the practical implementation of topological qubits. Previously, a research team from Delft University in the Netherlands, which collaborated with MS, announced that they implemented an experiment generating Majorana states but later withdrew their paper. The current research team also stated through this Nature paper, "We are not yet at a stage to draw clear conclusions," and remarked, "If topological qubits can expand based on this research, we will validate whether actual quantum operations can be performed."
References
Nature (2025), DOI: https://doi.org/10.1038/s41586-024-08445-2