“Anyone who is not shocked by quantum theory has not understood it.” That’s why the epithet “weird” is so often used. Good news for random number generators everywhere…
But researchers at the National Institute of Standards and Technology announced Wednesday they’ve harnessed quantum mechanics to not only generate truly random numbers, but to prove that they’re random.
“Our quantum source and protocol is like a fail-safe. We’re sure that no one can predict our numbers,” NIST mathematician Peter Bierhorst said.
It’s a new illustration of how the wackiest, most abstract science research can turn out to have commercial relevance years or decades later.
Quantum mechanics, pioneered in the early 20th century, offers explanations of the bizarre realm of the ultra-small, but it’s famously hard to comprehend. Yet one of its weirdest aspects — “entanglement,” which link the states of two separated particles — is key to the NIST work.
NIST plans to make its truly, certifiably random ones and zeros available to anyone who needs them. The agency already offers a service called the NIST Randomness Beacon using commercially available random numbers. Now it’s planning on adding its own technology to the mix when it matures out of its prototype finickiness, Bierhorst told CNET.
“It is also currently not quite fast enough to generate bits at the rate requested by the beacon — the beacon publishes 512 bits a minute and our reported experiment took 10 minutes to generate 1,024 bits — but we are working to improve the rate,” Bierhorst said.29In 500 years, robots haven’t become any less creepy
NIST’s method starts with a laser that generates pairs of photons — particles of light — that are entangled. Measuring the state of each photon in a pair then comparing them lets the researchers prove they were observing quantum mechanically generated ones and zeros.
That’s because the states are measured far enough apart that even the speed of light wouldn’t be fast enough to coordinate the photons in each pair. And that means the data couldn’t have been generated through some ordinary physics fakery.
There is one bother, though. The photons, as originally generated, aren’t equally likely to be ones or zeros. So a second phase of the process extracts a random subset. That subset’s one-or-zero probability is certified to be balanced to within 1 trillionth of 1 percent.
For more, check out the NIST researchers segment in the April 12 issue of the journal Nature.