PRINCETON, N.J., Oct. 1 (UPI) -- It sounds like something out of the movie "Back to the Future," but one forward-thinking physicist says as computer technology progresses, scientists may be able to use wormholes to send answers to calculations back to their own past to solve problems.
By basing computers off wormholes, or tunnels in time, theoretical physicist Todd Brun of the Institute for Advanced Studies in Princeton said they could be used to solve problems that would take a classical computer longer than the lifetime of the universe to solve.
"They could have amazing -- almost magical -- efficiency," Brun told United Press International.
Still, Brun cautions he "would be amazed if anyone built anything like this in my lifetime. I would not be amazed if no one ever built anything like this."
Computers in development work off the bizarre properties of quantum physics, where subatomic particles can apparently spin two opposite directions simultaneously. Such "quantum computers" can in theory operate much more powerfully than classical computers on certain problems -- breaking codes in seconds that would take classical computers years to crack.
These successes led Brun to ask whether other unlikely areas in physics could help solve computer problems. "As long as one is speculating, one might as well speculate wildly," Brun said.
The physicist focused on space-time loops known as "closed timelike curves." These distortions can develop because of wormholes -- tunnels that theoretically can catapult anything traveling through them faster than light or back in time. Wormholes are allowed by Einstein's theory of relativity, although whether they are practically possible is another matter.
Brun developed a program for such a computer to solve extremely difficult mathematical problems, such as factoring very large numbers. At the start, the computer checks the wormhole. If in the future, the computer solved the equation, it sends the answer encoded in bursts of particles back in time through the wormhole.
The program essentially works because of steps "which are never actually executed," Brun said. Such a computer essentially works similarly to a familiar time travel paradox.
"A brilliant young inventor receives a message from her future self, telling her that she is going to invent a time machine, and giving her the details of its construction," Brun said. "She duly builds the machine and demonstrates it. When she is old and famous, she sends a message back to her younger self, telling her that she is going to invent a time machine, and giving her the details of its construction.
"This situation is self-consistent, but still very strange. The information on how to build a time machine appears out of nowhere," he explained.
While this may seem "quite bizarre," Brun said such computers "don't defy logic, only common sense."
What happens if the computer is approached with a problem that would take so long to solve, the wormhole or the universe would end first? Or what happens if the wormhole can only send information backward for a short amount of time? Steps in the program break such incredibly difficult problems into smaller and smaller issues, until each is reduced enough to be solved within the computer's lifetime, at which point the answer is sent back in time.
Of course, the existence of such a computer depends on the unproven existence of wormholes. Even if wormholes do exist and can be harnessed, they could cause space-time fluctuations that destroy the wormhole.
"If there are fluctuations, however small, then they would wipe out any answers," said theoretical physicist David Deutsch of the Center for Quantum Computation at the University of Oxford in Britain. "Wiping out the fluctuations is in itself a computation that involves an arbitrarily large amount of work. There has been work on whether these loops are realistic, and everyone disagrees."
Brun believes these strange conclusions are less likely to lead to supercomputers than they are to show the existence of wormholes is even more unlikely.
"I think its implications are very interesting, but mostly for theory rather than practice," he said.
Brun described his program in research submitted to the journal Foundations of Physics Letters.
(Reported by Charles Choi, UPI Science News, in New York.)