IBM has ambitions to become a leader in quantum computing, and the new government partnership is set to help. According to the Japanese news publication Nikkei, information about a joint effort between IBM and Japan's National Institute of Advanced Industrial Science and Technology (AIST) to create a quantum computer containing 10,000 qubits by 2029 has been leaked into the public domain. That's far superior to today's class-leading 133-cubic-inch machines. That said, according to IBM's quantum computing roadmap, the company's commercial products won't even reach 2,000 qubits until 2033. The deal will be the company's largest quantum computing deal with the public sector.
Quantum computing has been one of IBM's main areas of activity for several years. The 10,000-qubit computer goes beyond IBM's current quantum computing roadmap. According to her, the company's commercial products won't even reach 2,000 qubits until 2033 and beyond. Earlier, IBM planned to release the 1000-qubit Condor computer in 2025, but the prototype was delayed.
The goal of the 10,000-qubit computer is to perform quantum computing without the need for backup on a traditional supercomputer. The thing is that modern 133-qubit machines often make too many mistakes, which is why their work has to be checked on auxiliary computers.
According to Nikkei sources, IBM and AIST will sign a cooperation agreement in the coming days. This partnership has ambitious goals. IBM and AIST plan to develop semiconductors and circuits capable of operating at temperatures close to absolute zero. The closer to zero Kelvin (-273.15 °C), the more efficiently and accurately quantum computers work. In today's largest machines, qubits and microcircuits are housed in separate chambers or rooms to maintain extremely low temperatures. Thus, the creation of components that function at extremely low temperatures is a necessary step to advance research in the field of quantum computing.
AIST will use its patents, knowledge base in the field of artificial intelligence and connections with Japanese component manufacturers during the production of the future supercomputer. The institute will provide access to future quantum computers for Japanese companies and industries. This will be achieved by training company specialists and lobbying for the introduction of quantum technologies in Japanese business. Access to key players in the Japanese industry was reportedly the main reason IBM struck the deal. This is the company's largest agreement with the public sector in the field of quantum computing.
However, a large number of qubits does not guarantee quality. IBM is betting on 133-cubic-inch machines that outperform 1,000-cubic-inch prototypes in terms of efficiency. Just as traditional processors use multithreading and caching to increase performance, quantum computing has other methods that increase its power beyond simply increasing the number of qubits. After all, the more qubits, the less stable quantum computers become. The future of quantum computing lies in smart engineering that will ensure the stability and cost-effectiveness of 10,000-qubit and more powerful machines.
The partnership between IBM and AIST could have a major impact on the development and implementation of quantum computing. However, today's quantum computers are still in their infancy and have a long way to go before they become useful for ordinary users or professionals. The recent case, when a team of researchers using a classic Commodore 64 computer outperformed IBM's 2021 quantum processor in solving a certain problem, serves as a clear example that IBM and the industry are still far from the widespread use of quantum computers.