Quantum Computing is a solution looking for a problem
Spending on Quantum Computing is now an eye watering tens of billions of dollars every year according to research from firms such as Deloitte. The global investment is comparable to NASA’s budget but where the space agency’s plans include getting back to the moon, it is not yet clear what the Quantum Computing “moon shot” goal really is.
There are plenty of companies pursing the vision of a quantum-enabled computing future making announcements every few weeks. But the largest portion of the spend is in fundamental research which confirms that this is a field where the foundations are yet to be built.
The security implications are clear. A Quantum Computer has the potential to realise “Shor’s Algorithm” which undermines the security on which the internet, cryptocurrency and indeed much of our daily lives depends. In the simplest possible terms, Shor’s Algorithm would allow a Quantum Computer to determine the factors of very large numbers. The difficulty of doing so is how the “https” at the start of your web address is encrypting the communication you have with each website, including this one.
It is always easier to sell a crisis, after all John F. Kennedy created the first moon shot to counter the fear of the Russians getting a military advantage in space. I’ve written before that people are right to worry about the cybersecurity risks of quantum technology, particularly Quantum Computing, but that the risk mitigation options may actually already exist (see Preparing for Q-Day using Vernam, Miller and Shamir).
The discussion of Quantum Computing as being a moon shot misses the point. We don’t really know whether it is possible to make an errorless Quantum Computer at all (see The Quantum Computer dream could be killed by information management). A much better analogy is the laser, a device first emerging from laboratories in the 1960s. One laser pioneer, Irnee D’Haenens, described the technology as “a solution looking for a problem”. Lasers, a direct result of the same quantum predictions as Quantum Computing relies on, turned out to be very useful but for applications that weren’t even under consideration as the technology was developed from holograms to range finding.
Today’s technologies in the general quantum technology realm include many challenging and foundational capabilities to support the fundamental building block of Quantum Computing, qubits, the quantum analogue of the digital bit. Qubits can be created using many different approaches but share the attributes of being in multiple states at once, enable so-called entangled interactions where their states are correlated and support logic gates, essential for any calculating engine or computer.
Supporting this technology are capabilities that are interesting in their own right, such as the ability to cool qubits down to near absolute zero, colder than the universe itself. For some of the quantum technologies being developed, these super cold temperatures are required to enable the qubits to retain their quantum properties and to avoid excessive errors creeping into calculations.
If you read the press releases of the various companies and research groups in this field, you could be forgiven for thinking they are all on the verge of building a general Quantum Computer. Such a device would make Moore’s Law of computing power growth look like something from the stone age! In reality, what we have is a range of amazing technologies emerging which promise solutions that don’t even have problems to solve yet.
Arguably the least interesting stream of news is about the general Quantum Computers being built with, to-date, a little more than 100 qubits. These devices from some of the world’s largest companies are displaying amazing science and are pushing the field forward. None of these machines can perform tasks that outperform today’s digital computers and are focusing on being platforms to test ideas. Alongside the billions being spent on the race to their development, the really interesting research is looking for novel ways to leverage their current capabilities.
The next category of application of Quantum Computing research is a group of technologies built around quantum annealing, a specialised application which solves a group of optimisation problems. Where there are a myriad of ways of approaching the same problem, optimisation can narrow down the solution set. This can be particularly interesting in logistics, pharmaceuticals, market research and other problems that are typically solved using linear programming techniques. While none of the current group of specialised devices are solving problems faster than today’s digital computers, they are arguably close.
The third category could combine annealing applications with the nature of qubits, being inherently quantum. If qubits are one level abstracted from the building blocks of every chemical molecule, the simulation of chemicals becomes practical in a way that is simply not possible with a digital computer. Although no practical application has yet been created, this form of machine built on qubits would arguable not be a computer at all but rather be a controlled analogue of a real-world molecule.
The fourth category is communications. Again, using the attributes of entanglement that have been developed to support qubits, but applying them not to calculations but rather to communications promising the possibility of completely secure networks.
Finally, we are seeing the building blocks of Quantum Computing being applied in interesting ways that could never have been anticipated. For example, researchers have built a simple quantum battery that can theoretically be instantly recharged an infinite number times.
Even beyond today’s research would be new approaches to quantum-based artificial intelligence based on indications that real intelligence and consciousness are a quantum phenomenon. We don’t know if a Quantum Computer can be built, but nature has shown us that an intelligent and conscious brain can be. Perhaps biologists and quantum researchers should spend more time together!
All of this suggests that as research into Quantum Computers continues apace, it is worthwhile to at least direct some of this effort and investment into finding new problems that the emerging technologies can solve. The next unicorn will likely be based on commercialising a novel problem rather than creating a new solution.