While classical computing encodes information in bits represented by ones and zeroes, quantum computing works with quantum bits, or qubits, which can be ones and zeroes at the same time. That quality, researchers believe, could make quantum computing perform certain tasks much faster than current computers can, like drug discovery and materials science research. But today's quantum chips often require very cold temperatures and only work for short bursts of time.
Intel has previously produced quantum chips for testing, but the new quarter-sized 17-qubit superconducting test chip is the first one the company is discussing publicly since announcing its $50 million partnership with QuTech — a research institute that involves the Netherlands' Delft University of Technology and Dutch Organisation for Applied Research — two years ago. One reason is the new system includes special packaging that can make the chips for longer stretches of time.
"The design improves reliability and thermal performance, as well as reducing radio frequency interference," Jim Held, the director of emerging technologies research at Intel Labs, told CNBC in an interview.
It makes sense that Intel is opening up now about the state of its quantum program, given that in recent months Google, IBM and Microsoft have all shown off their quantum chips. (IBM's latest chip also has 17 qubits. Having that number means QuTech will be able to run certain error correction protocols on the chip, Held said.) Meanwhile, privately held companies including Canada's D-Wave Systems and the Bay Area's Rigetti Quantum Computing are also active in the field.
As one of the world's biggest chipmakers, Intel is surely keen to ensure that its future is bright. Founded in 1968, Intel has long kept buyers coming back for more powerful chips in accordance with Moore's Law, the concept from cofounder Gordon Moore that says the number of transistors on a chip will double roughly every two years. While Intel executives say Moore's Law is doing fine, last year the MIT Technology Review declared that it was dead, and earlier this year Nvidia CEO Jensen Huang said the same thing.
What is clear is Intel believes the world should use different terms to talk about advancements in silicon. And that stance comes as the company has been building up a portfolio of chips designed for artificial intelligence, a domain where Nvidia has seen gains. For instance, Intel Labs recently developed "a first-of-its-kind self-learning neuromorphic chip."
But that doesn't mean today's chips will become obsolete anytime soon.
Quantum computing "is not going to replace today's computer," Held said. "It's going to augment them. It doesn't do all the things today's computers do well and will continue to do well."
