FinTech Connect: FS must adopt quantum encryption before it is too late

By Sunniva Kolostyak

Financial services must adopt quantum computing as soon as possible to ensure the highest level of cybersecurity, according to a quantum expert speaking at FinTech Connect.

Mark Jackson, Scientific Lead at Cambridge Quantum Computing, says it is important that the financial industry moves now as it could be too late to adopt new encryption methods in a couple of years.

Jackson explained that while existing encryption methods are working well, they will have to be revised with the advent of quantum computers which in the future could have the ability to decrypt the mathematics used today.

“This is being considered a threat, and so there is already a lot of effort underway to advance this,” he said.

A number of mathematical routines are therefore being developed to withstand hacking from quantum computing – such as the use of truly random numbers – however, the development and subsequent implementation will take time.

Quantum computers leverage quantum mechanical phenomena, such as superposition and entanglement, to manipulate information. They rely on quantum bits, or qubits, which are able to store a 0 and 1 at the same time, unlike the binary bit.

“Another reason that you should start making this up right now is because there are rumours that hostile governments or groups are archiving things right now from the internet,” Jackson continued.

“In a few years, when quantum computing technology catches up, they will be able to undo the encryption which exists right now. And so, it could be embarrassing or even harmful to you if any of that information from today is uncovered in a few years. We really do encourage you to start thinking about upgrading right now.”

More famously, quantum computing is set to bring another dimension to data analysis, picking up patterns that would be missed by normal machine learning.

One example is the Monte Carlo simulations, where a financial algorithm is used to predict a future price using systematic randomness. The more simulations run, the more accurate the answer is expected to be.

“There has already been work which suggests that quantum Monte Carlo is quadratically faster than classical Monte Carlo – so what that means is, to get the same level of accuracy that you would get by running a million simulations on a classical computer, you only have to run a thousand simulations on a quantum computer,” Jackson said.

“There are all sorts of other things, like in optimisation that shows up all over such as in arbitrage or in portfolios. And there are optimal quantum optimisation algorithms which are exponentially faster – it’s even better than quadratic.”