Kia ora koutou
In March we were excited to launch our intern programme, and have particularly enjoyed having our enthusiastic PhD graduates from the MacDiarmid Institute join us in the Office to have a first look at some of the projects that we are not resourced to do, while they learn some skills needed to shift from a research focus to an ‘evidence synthesis for policy’ focus.
At the end of June, Wayne Crump completed his internship on Quantum Computing – our first graduate. This is perhaps the most “academic” topic that any of our interns chose, with this technology a relatively long way from being scalable and likely to impact society. It is also arguably the hardest to explain! The analogy Wayne includes below might help: current computers use ‘bits’ which have a binary state – let’s say one or zero – like a coin that has been flipped and must land as either ‘heads’ or ‘tails’. Quantum computers have ‘qubits’ which exploit quantum mechanical phenomena and are not restricted to a binary state – more like a coin that is spinning and doesn’t land. This science opens up the possibility of vastly more powerful computers, once we have figured out how to manufacture them.
(For anyone whose head is hurting and would like to venture further down this rabbit role – I recommend this brilliant video by IBM’s Talia Gershon)
Wayne gathered together all the resources he could find internationally on how quantum computing might impact on society, with a particular focus on cybersecurity issues.
Download the list of quantum computing resources (PDF, 196KB)
He also wrote an accessible short information summary on quantum computing.
Download the quantum computing summary (PDF, 252KB)
He also wrote a reflection on his time with us, which I’ve included below. Thanks for all your work with us Wayne, and we wish the best of luck with your quantum computing journey.
“Over the past three months, I have been looking into the field of quantum computing for the Office. I found the idea of a new unconventional way of doing computing intriguing, especially the thought of exploiting some of the weird phenomena of quantum mechanics for this purpose. The chance to do an internship at the Office was a perfect opportunity to learn a bit more about the quantum computing space and get a feel of what was happening, while hopefully providing some useful information to people who think about what these things might mean for policy in New Zealand.
The large investments by the US, EU, UK and China show that interest in quantum computing is not insignificant. The large tech companies also have programs dedicated to building these machines. The hope is that quantum computers might help us go beyond our current paradigm of computing, which is limited physically by how small a transistor can be made. In a conventional computer, data is encoded as a series of “bits” which can be in one of two states (on/off, true/false, one/zero). A series of logic gates operate (algorithm) on the bits to manipulate your data in some desired way.
Instead of using bits like a classical computer, you have “qubits” (quantum bits) and apply quantum gate operations. The special thing about a qubit is that it can in a sense be in both one and zero states at the same time. An analogy to this would be a spinning coin, you don’t know if it will land as heads or tails and in some sense, it is in both of those states when it is still spinning. The theoretical power comes from the fact that you can in a sense set your series of qubits to represent many possibilities at the same time, and gate operations you do upon these qubits will operate on all these possibilities at once in a highly parallel way.
The technology at this stage, however, is still young and undeveloped, with current generation quantum computers only having in the 10s of qubits (IBM: 50 qubits, Google: 72 qubits). D-Waves machine is an exception here with 2000 qubits, however, its model of computation is different as it is purpose-built to solve one type of problem. Error is a large problem that limits scaling up to large numbers of qubits. A quantum computer must maintain its quantum state through a whole calculation, however, this quantum state is very sensitive to any noise that can get into the system. This causes errors in calculations, and so current quantum computers can only perform a limited about of gate operations before the errors become too large.
As for what this might mean practically, quantum computers should be a catalyst in the development of new materials, chemicals and drugs as they are by their nature suited to the simulation of these quantum systems. They have shown to be good at optimization problems and there is also research looking into how they can enhance AI capability. One big concern, however, is their theoretical ability to crack our current public key cryptography. Public key cryptography is important in the security of many online systems and is used to encrypt email, authenticate websites and updates and establish secure communication between two parties. The kind of quantum computer needed to decrypt current public key cryptography however probably needs to have at least in the millions of qubits, and also needs to be able to perform error correction since the algorithm for decryption is very sensitive to errors.
Experts I was able to talk to didn’t think this was an issue at this stage as this kind of quantum computer is still likely far off. You can read my information sheet on the topic and also the list of resources on the subject.
Overall the technology is still on the horizon and is currently unproven but it will be interesting to track its development and see if it lives up to expectations. Throughout this work, I was able to meet with several academics who gave their thoughts on what was happening in the space. This was valuable knowledge, as it served as an asterisk on much of what I was reading in government reports and other sources. Much of the media around quantum computing is positive so it is good to get the views of other parties closer to what is going on. It was refreshing also to find that people were in general happy to share their knowledge, which indicates that they care about the policymaking process and desire for it to be properly informed.
Compared to my previous work in doing a PhD, this project required a bit of a mindset change as things needed to be looked at in terms of impacts on policy and industry. This has given me a much broader view of the interaction between science and policy and an appreciation of where researchers can be valuable in policymaking. It was great to have a luxury of exploring a field which holds interest for me and hope to potentially find a research position in.
I would like to thank the Office and the MacDiarmid Institute for giving me this opportunity, and also the academics and industry people who were willing to share their knowledge.”