IBM Wants to Be at the Forefront of Quantum Computing: It's Planning to Unveil a 10,000-Qubit Computer by 2029

  • IBM and the Japanese research organization AIST signed an ambitious collaboration agreement.

  • One of its objectives is developing semiconductors that can operate at temperatures very close to absolute zero.

No comments Twitter Flipboard E-mail

IBM has an ambitious quantum computing roadmap. In early December 2023, it unveiled Condor, its first 1,121-qubit quantum processor, Heron, a 133-qubit fixed-frequency quantum chip that is three to five times more powerful than the 127-qubit Eagle processor, and System Two, a quantum computer equipped with three Heron chips. But that's not all.

In 2025, it plans to launch Flamingo, a 156-qubit quantum processor capable of mitigating, although not correcting, its own errors. Engineers will achieve this through improvements in the quality of the superconducting qubits, made possible by innovations in logic gates and manufacturing processes. In fact, IBM will continue to refine and scale the Flamingo quantum processor until 2028.

However, the real surprise will come in 2029, when IBM plans to debut Starling, its first quantum hardware with the ability to correct its own errors. The company’s roadmap clearly indicates the chip’s correction (rather than mitigation) capabilities. It’s a big deal. What's more, this is a significant development because it should allow researchers to address problems that current prototypes are unable to handle.

Finally, in 2033, the company aims to release Blue Jay, which will mark the beginning of a large-scale implementation of error-correcting quantum hardware.

IBM and AIST Are Preparing a True Quantum Computer Revolution

IBM revealed this roadmap in early December. At that time, however, it didn’t make its collaboration with Japan’s National Institute of Advanced Industrial Science and Technology (AIST) public. This is very significant given that the goal of this partnership is to develop a 10,000-qubit quantum computer by 2029.

The machine aims to conduct quantum calculations without the need for support from a classical supercomputer.

This quantum computer aims to conduct quantum calculations without the need for support from a classical supercomputer. Current prototype quantum computers, on the other hand, don’t have enough qubits to implement a robust correction system. Due to their error-prone nature, they rely on a supercomputer for validation. IBM and AIST’s 10,000-qubit quantum computer will presumably be able to deal with such problems autonomously.

In addition, the collaboration between IBM and AIST also aims to develop semiconductors and circuits that can operate at temperatures extremely close to absolute zero. This is crucial because quantum computers with superconducting qubits, such as those created by IBM, function more stably and efficiently at temperatures near absolute zero (523.67ºF).

As such, IBM and AIST intend to manufacture integrated circuits that can operate at these extremely low temperatures. If one thing is clear is that significant advancements in the field of quantum computing will arrive before the end of this decade.

Image | IBM via Flickr

Related | The First Qubit Built With 99.9% Control Accuracy on a Silicon Wafer Is Now Ready. It Changes Everything

Home o Index