|
Researchers around the world are trying to perfect quantum
computing, which relies on quantum mechanics to reach computing
speeds far faster than classical silicon-based computers. The
challenge has been to create quantum computers that are reliable
enough in the real world to consistently beat conventional
computers.
Microsoft, Alphabet's Google and China's Baidu , along with
startups and nation states, are all racing to develop quantum
machines.
As quantum researchers have made the machines big enough to
outpace classical computers, they have struggled with data
errors. On Monday, IBM showed what it says is a new way of
connecting chips together inside machines and then connecting
machines together which, when combined with a new
error-correction code, could produce compelling quantum machines
by 2033.
The first machine to use them is called Quantum System Two,
which uses three "Heron" chips. Dario Gil, IBM's senior vice
president and director of research, said that progress will
appear fairly steady until 2029, when the full effect of the
error-correction technologies come into play.
After that, the machines should see a sharp uptick in
capabilities, similar to how AI systems that developed slowly
for the past 15 years became vastly more sophisticated over the
past year.
"You're going to have to tie them together," Gil said of IBM's
newest chips. "You're going to have to do many of these things
together to be practical about it. Because if not, it's just a
paper exercise."
IBM is not the only quantum player targeting machines within the
next few years. Startup PsiQuantum, which is working with
GlobalFoundries to make its chips, told Reuters earlier this
year it plans to have a commercial machine within six years.
(Reporting by Stephen Nellis in San Francisco; editing by Diane
Craft)
[© 2023 Thomson Reuters. All rights
reserved.]
This material may not be published,
broadcast, rewritten or redistributed.
Thompson Reuters is solely responsible for this content.
|
|
