FILE PHOTO: IBM logo is seen in this illustration taken February 16, 2025. REUTERS/Dado Ruvic/Illustration/File Photo
SAN FRANCISCO (Reuters) -IBM announced on Wednesday it has built a new experimental quantum computing chip called Loon that demonstrates it hit a key milestone toward making useful quantum computers before the end of the decade.
Quantum computers could someday solve problems that would take classical computers thousands of years. But due to the uncertain nature of quantum mechanics, the chips are prone to errors.
Correcting those errorsis the key focus of tech giants such as Alphabet's Google and Amazon that are chasing quantum computers alongside IBM. In 2021, IBM proposed a new way of doing error correction: adapt an algorithm for improving cellphone signals to quantum computing and run it on a combination of quantum chips and classical computing chips.
The downside of IBM's idea is that the quantum chips become harder to build because they must contain not only basic building blocks of quantum chips called "qubits" but also new quantum connections between the qubits, Mark Horvath, a vice president and analyst at research firm Gartner, told Reuters in an interview.
"It's very, very clever," Horvath said. "Now, they're actually putting it in chips, so that's super exciting."
Jay Gambetta, director of IBM Research and an IBM fellow, said the key was tapping the Albany NanoTech Complex in New York, which houses the same chipmaking tools as the most advanced factories in the world.
Loon remains in its early stages, and IBM did not disclose when outsiders can test it. But the company also announced on Wednesday a chip named "Nighthawk" that will be available at the end of this year.
IBM believes Nighthawk could beat classical computers on some tasks by the end of next year and is working with a group of startups and researchers to share its code openly so that others can test those claims.
"We're confident there'll be many examples of quantum advantage," Gambetta told Reuters. "But let's take it out of headlines and papers and actually make a community where you submit your code, and the community tests things, and they select out which ones are the right ones."
(Reporting by Stephen Nellis in San Francisco; Editing by Muralikumar Anantharaman)
