�������ϲ���̳

As the payments sector continues to embrace cutting-edge technology, the sophistication of frauds and scams attacking the sector is also advancing, creating challenges for regulators. 

As covered by �������ϲ���̳, regulators and financial institutions (FIs) around the world are in an arms race with scammers, with attacks via artificial intelligence (AI) tools such as deep fake videos, robo-calls, account hacking and impersonations all increasing in frequency. 

In response, organisations are themselves using AI to automate and improve identity verifications and to analyse transaction patterns, greatly improving security measures. 

And the next phase of technological development — quantum computing — is on the horizon. 

The hyper-fast processing speeds that quantum technology promises bring great opportunities, but are something of a double-edged sword. 

On the one hand quantum computing will improve AI capabilities, fraud detection and credit risk analysis; on the other, it will render existing cryptography methods redundant and enhance techniques for fraudulent activity and cyber-attacks. 

Quantum computing

Quantum technology attempts to harness quantum physics in areas such as computing and security.

The fundamental building block of quantum computing is a qubit, a quantum bit equivalent to what is currently used in digital computing, but with additional properties such as "superposition" and "entanglement" that make up a quantum processing unit (QPU). 

If harnessed properly, a QPU could allow for a new range of algorithms that could solve complex problems much more efficiently than can be done with traditional computing technology. 

Quantum computing has yet to be fully realised outside specific conditions because of cooling issues and because of "universe noise". 

Isolating existing quantum computers from the rest of the universe is a complex engineering problem, so existing quantum computers today are stable only for fractions of a second.

They can only run problems for a relatively short amount of time before the "noise of the universe" overwhelms the quantum devices and it breaks down. 

A lot of the research in this space is trying to figure out how to suppress and mitigate these errors and how to make the quantum algorithms run longer to solve bigger problems. 

James Mitcheson, chief architect at Citizens Bank, said that quantum computing offers immense potential value and should be carefully considered in a hybrid approach alongside traditional computing and AI such as large language models. 

“We're going to be breaking down the problem into pieces that are using the right tool for the right problem,” he said.

Regulators and payments organisations will need to monitor the progress of quantum computing because once it comes to market, it could have a significant impact.

Optimisation

One area quantum technology can potentially revolutionise is "optimisation" — improving strategies, technology and processing logic to reduce cost and fraud, improve authorisation speed and rates and generally provide a better service to end users. 

Sources suggest that technology supporting optimisation is the closest to the market of any quantum advancements. 

“We're looking at problems where we can see an optimisation-based approach potentially delivering commercial value, the idea of optimising machine learning algorithms,” said Steve Flinter, AI and quantum computing engineer at Mastercard. 

“We've also explored the idea of how to use optimisation to do things like solve cross-border net settlement problems where, for example, Mastercard has 20+ settlement currencies around the world. 

“When you try to imagine what a real-time cross-border multi-currency settlement system would look like you have to think about it not just settling through a single currency like USD, but potentially across lots of different currencies,” he added. 

“Potentially with digital currencies in the future that problem becomes exponentially large. We think quantum computing might potentially be able to deliver benefits in that kind of environment.” 

Quantum technology could also affect payments in areas such as fraud management, portfolio optimisation and credit risk measures. 

However, there are significant challenges with implementation, even beyond the physical constraints. 

“Some of the hardest challenges have been less about access to technology, and more helping our product managers and leaders think about what problems within their space might be a good fit for quantum technology,” said Flinter. 

FIs will need to identify where quantum computing could add value, and which products, services or problems are best suited to a quantum approach. 

With great power comes great responsibility 

Regulators and payments organisations must be aware that alongside the potential for optimisation and fraud detection come new risks in terms of cryptography and security. 

The existing standard for encryption is a public key cryptosystem known as RSA that assists with data transmission and is used on most of the data found within a computer. 

The system is based on algorithms using prime number generation that essentially regenerate faster than computers can crack the code. 

Quantum computers would theoretically be able to crack RSA, but could also allow for new kinds of encryption that would be "uncrackable" to prevent security breaches and loss of reputation. 

“Our public key cryptography systems may be compromised by quantum devices, we've spent quite a bit of time trying to understand what the likely timeframe for that is,” said Mastercard’s Flinter. 

“It's going to be a decade or more before there is a sufficiently powerful quantum computer available to do that.”

Chinese researchers claimed to have worked out how to solve RSA with existing quantum technology in 2022, but the claims have proven unfounded.  

Losing RSA encryption could disrupt secure online payments and authentication systems that would disable digital wallets. 

Blockchain technology could also be at risk from quantum computers, putting cryptocurrencies and stablecoins in jeopardy.

The industry is poised for a signal to say that the technology is maturing past the noisy version available today and to a place where its promises and fears are realised. 

Regulation

This all highlights the need for greater regulatory protection to ensure that consumers and merchants remain protected throughout these advancements. 

As covered by �������ϲ���̳, not only is AI regulation currently lacking in the US, the Federal Reserve has indicated that any rules must be carefully considered and not rushed to ensure they do not affect innovation. 

This caution is echoed by technology providers, which are keen to ensure that any regulation does not stifle their development opportunities and that it takes into account the differences between developers.

“One of the things to really recognise about the space is that it is still highly unevolved, so there's a lot of things that are uncertain and a lot of science undeveloped,” said Jason Kwong, chief strategy officer at OpenAI. 

“If there's going to be more regulation, then it's got to be sensitive to the size of the company or the amount of resources that the developer has at its disposal. 

“One of the great strengths of the American economy is its private sector entrepreneurship formation. You don't want to go so far that you choke that off.” 

He added that it would be wise for regulatory exemptions for smaller companies looking to innovate. 

Regulatory sandboxes, for example, provide opportunities for companies with certain revenues or customer bases to take certain risks to experiment. 

“You want to ensure that there's continued innovation,” he added. “If you pass a bunch of rules that are very prescriptive in terms of the technology you can risk locking companies into a particular technology path, and they may not actually continue to innovate.”

In this context, it seems unlikely that the incoming Trump administration will seek to regulate quantum computing. 

Given the lack of appetite for regulating AI at this stage, despite its widespread use, regulating a conceptual technology is going to be low on the agenda.