Some heavy-duty human brainpower is focusing on the job of creating and marketing a spectacular new form of computer brainpower.

Quantum computing (QC) might turn currently unsolvable math problems into snaps, but could also destroy trusted infrastructure such as bank card security. It also has the potential to make early entrants in this field very, very rich ... if their high-risk investments pan out.

Top business and technical minds in QC descended on the University of Calgary recently to celebrate the launch of the institution's Institute for Quantum Information Sciences (IQIS). They join researchers in places such as Harvard, Stanford, the University of Waterloo - and, increasingly, the private sector.

"The folks at Waterloo are focused on trying to build a quantum computer.We're trying to figure out how to keep the world safe from it if they succeed," laughs IQIS director Barry Sanders.

Tom Keenan, Business Edge

Alex Lvovsky of U of C's quantum institute at work in lab.

His point is echoed by Chip Elliott, principal engineer at Boston-based BBN Technologies. "I think that there are a lot of indications that the ground (of computer security) may be disappearing below our feet," he says. "So it might be good to have a plan B available."

He's referring to the very real possibility that someone with a QC could compromise security that's based on calculation-intensive math problems. Put a current computer to work on a toughie like factoring a 1,000-digit-long number into two primes, and it will run through its bits (0s and 1s) for decades. QCs use "qubits," which can take on many values all at once, so they might solve this problem in a flash.

QCs follow the wacky rules of quantum physics, including the Heisenberg uncertainty principle: If you touch a piece of quantum information, you destroy it. That sounds bad, but it can actually lead to unbreakable security since it renders eavesdropping impossible.

Elliott's quantum network project is funded by DARPA, the U.S. government research agency that gave us, among other things, the Internet as we know it. He won't disclose the budget, but says it's already a technical success.

"We have six nodes up and running 24 hours a day, seven days a week, between BBN, Harvard University and Boston University."

What do they do with the world's first (known) quantum network?

"Quantum cryptography by a variety of techniques," says Elliott. "We then try attacking it to see how well it actually holds up in practice."

Cryptography, the making and breaking of codes, is the current sweet spot for quantum information science. Bob Gelfond, CEO of one of the few companies that's actually selling quantum devices, knows that for sure.

He was one of the early investors in Amazon and says he was drawn to quantum information science because it gave him "a once-in-a-lifetime opportunity to spawn a new company that could be present at the birth of a new technology."

Gelfond founded privately held MagiQ, and easily raised $6.9 million US in seed funding. He got friend and Amazon CEO Jeff Bezos to invest. Then, Gelfond says, they evaluated MagiQ's options. Acquire intellectual property? Collaborate with a bigger company? Build something? He chose the latter path, and they are currently selling the QPN Security Gateway to people who really, really need to keep their secrets safe. It provides point-to-point security for fibreoptic lines up to 120 kilometres long. It offers, they claim, absolute and "future-proof" security.

Gelfond explains that all existing communications lines are vulnerable to intrusion. Even fibre-optic cables can be tapped by bending them and siphoning off some of the photons. He says there are even techniques to steal the information without physically touching the fibre.

With his QPN system, any attempt to read a photon in transit destroys the information in it. The communicating parties will be alerted that something is wrong, and the eavesdropper will not get any useful information.

Gelfond insists that this is a qualitatively different kind of protection from anything else on the market. For example, even if the eavesdropper can't break an intercepted non-quantum transmission, he could store it away and hope that more powerful computers will allow decryption in the future. That's simply not possible with quantum cryptography.

Gelfond is coy about his customer list, but admits that the company is far from profitable. His next step is to try to interest big customers and also data-transmission carriers. Yes, he has competitors, notably a Swiss company called Quantique, which also makes quantum products. In a nice touch, they offer free "true random numbers" (as opposed to pseudo-random numbers generated by conventional computers) from a website. If you happen to win a lottery using numbers from it, they ask you to donate half the proceeds to the site.

BBN's Elliott has simple advice to Gelfond and others trying to sell quantum devices for security. "Most information-security technology follows a very predictable path," he says. "It's first used by governments, then by very large financial institutions, then by enterprises, then by people like you and me.”

Gelfond also needs to look for customers with deep pockets. With a current selling price of around $70,000, he says, these things aren't going to be showing up in Future Shop any time soon.

With more than $1 billion invested since 2001 and the opening of a $43-million photonics fabrication centre in Ottawa last year, the future is looking bright for Canada's photonics industry.Speakers at the Calgary conference praised the university's vision in setting up IQIS, and professor Raymond Laflamme of the University of Waterloo's Institute for Quantum Computing welcomed the U of C researchers into the fold, as did Greg Luoma, chief technical officer of defence contractor General Dynamics Canada. His company is working closely with IQIS.

"We do the science," says IQIS Director Barry Sanders, "and they are going to develop the products. They have the customer base to sell to.”

In fact, one of the company's senior researchers is doing a PhD under Sanders.

No credible quantum information scientist is willing to give a firm date for "the first quantum computer.”

They usually punt the question with "your teenager won't be asking you for one soon" or "perhaps someone has built one already and isn't telling us.”

But the consensus is that the QC's day will come. It's pretty hard to believe that the best business and technical brains are chasing thin air, even if it sometimes looks that way.

In true Heisenberg uncertainly principle fashion, we have no way to be certain about the future of quantum computing until we actually touch it.

Web watch: http://www.iqis.org/ quantum.bbn.com

http://www.magiqtech.com/

(Tom Keenan is a professor at the University of Calgary and an expert on technology and its social implications. He can be reached at keenan@businessedge.ca)