At CES 2018, Intel CEO Brian Krzanich announced Intel successfully built a 49-qubit chip, which aims to allow researchers to improve error correction techniques and simulate computational problems.
Intel 49-qubit quantum chip builds upon Intel’s two months old 17-qubit chip, which introduced a new architecture improving reliability, thermal performance, and reducing interference between qubits. It also relied on new processes, materials, and designs to enable better scaling.
Despite the fast progress that Intel and other tech giants, including Microsoft, Google, and IBM, are reporting, Mike Mayberry, corporate vice president and managing director of Intel Labs, tried to provide a more realistic perspective:
We expect it will be five to seven years before the industry gets to tackling engineering-scale problems, and it will likely require 1 million or more qubits to achieve commercial relevance.
Researchers still need to figure out how solve a number of issues, including error correction to work around individual qubits fragile quantum state, mapping software algorithms to the quantum hardware, and build local control electronics to control the quantum system and get its results.
Intel’s 49-qubit chip is based on superconductors, which require very low operating temperatures of about 20 millikelvin (–273 degrees Celsius). Nevertheless, Intel is also researching so-called “spin qubits” that are based on more traditional semiconductors. Spin qubits are usually smaller than superconductive qubits and promise to scale up to thousands or millions of qubits. According to Intel, spin qubits resemble a single electron transistor similar to conventional transistors and could be manufactured with comparable processes. Furthermore, Intel claims it has already developed a process to fabricate spin qubits on 300 mm silicon wafers.