6 vulnerabilities technology is creating in the financial system

You’re scrolling through your feed when you see a post from the CEO of your bank alerting customers to a major problem. What do you do?

In a world of increasingly convincing deep fakes and bot-generated fake news, fabricated information presents a real threat to the financial services sector. If you, and multiple other customers like you, react and withdraw your money, then what started as a fake problem could soon become a very real run on the bank.

Like with so many other sectors, technology is undoubtedly influencing and reshaping financial services. It has boosted financial inclusion and improved security and the customer experience. But it has also created new risks, which without action could create major systemic problems.

The World Economic Forum launched the Technology, Innovation and Systemic Risk initiative in 2021 to explore how technology is increasing risk in financial systems, and how it can also be used to mitigate this risk. The group’s research reinforces the need for a greater understanding of how technology-driven risks can emerge and spread.

Here are six of the risks it has identified which are fuelled by technology.

The technology and financial barriers to producing machine learning algorithms are rapidly falling away. This makes the proliferation of deep fakes and synthetic botnets a growing problem. For example, botnets are being used to attempt to induce bank runs, by spreading fake stories of customers unable to withdraw funds from their accounts. They are also creating ‘flash crash’ events, by mimicking trusted financial sources and posting about fictional market-moving events, for example using the Twitter account of a central bank chief to call out ‘problems’.

The increasing sophistication of deep fake technology is also making it harder for fraud detection software to identify attacks.

The cryptocurrency market is largely isolated from traditional markets because of low institutional exposure. However, as regulation starts to emerge, this may heighten confidence levels and there may be a spike in interest from retail and institutional investors.

But with limited visibility of leveraging and capital reserve data, investor deposits could be vulnerable. Meanwhile, security weaknesses could also pose a separate risk which may affect the health of crypto markets.

If a cryptocurrency exchange can’t meet customer withdrawal requests, panic could spread to other crypto exchanges and create a sharp drop in cryptocurrency values. This may lead to institutional investors having to unwind funds invested in traditional markets in order to repay debt and cover their losses.

It also creates a credit risk for banks if individuals investing in crypto lose a significant volume of their savings and are unable to meet their bank loan repayments and mortgages, for example.

Buy now, pay later (BNPL) platforms are becoming increasingly popular, giving consumers easy-to-access finance with low credit approval processes. BNPL payments surpassed $120 billion in 2021, and are projected to grow by a quarter over the next three years.

However, the ease of access also creates opportunities for overborrowing, with this type of finance particularly appealing to customers who may not be eligible for loans through traditional channels.

Barclays reports a quarter of BNPL users are concerned about their ability to repay their BNPL bills. And because of the limited reporting requirements of BNPL debt, there is also limited visibility of consumers’ total debts.

BNPL providers package this debt and sell it to investors as securitized assets. If the BNPL market continues to grow and a hard economic climate means more people default on their payments, there could be a spiralling effect on the financial system. This could be similar to the large-scale defaults that drove the mortgage-backed security crisis in 2008.

Where do we get our market intelligence? We all know rumours and speculation can spread like wildfire on social media and through messaging channels like WhatsApp. And financial information is no different.

In recent years we have witnessed the growth of ‘meme stocks’, where asset prices have become disconnected from the underlying value of a company and have been driven by speculation on social media.

Algorithms on platforms like Twitter and Reddit amplify stock volatility by creating echo chambers where investors communicate with others holding similar views, reinforcing speculation.

This meme stock purchasing could cause wider problems if investment firms start to liquidate their holdings to cut short their losses amid volatility, creating wider market destabilization. And retail investors may also face a liquidity crunch as markets restabilize, threatening their ability to pay back loans and commitments elsewhere in the financial ecosystem.

“Technology-driven risks in the financial system are inherently complex given how heavily interconnected today’s global financial system is,” says Drew Propson, Head, Technology and Innovation in Financial Services, World Economic Forum.

“Mitigating these risks, then, will require collaboration from actors across the ecosystem, including financial services executives, policy-makers and regulators, and efforts will need to extend beyond borders. Fortunately, technology itself can be an extremely valuable mitigation tool, particularly in the areas of communication, where new technologies can facilitate rapid reporting and essential information transfer.”

Real-time sensor-generated information has become broadly incorporated into investment firms, helping them make investment decisions, identify market inefficiencies, and predict future market moves.

More than half of the datasphere is predicted to be generated by sensors by 2025, including from satellites, CCTV, smartphones and devices connected to the Internet of Things (IoT).

However, IoT botnets, which tamper with sensor feeds, are being seen by criminals as a low-cost high-reward method of disrupting commodity markets for financial gain. For example, research suggests that deploying just 50,000 botnets to activate thermostats and air conditioning units could impact a region’s power grid and influence market prices.

Alongside this, there is also a growing black market for fake or corrupted sensor data.

If multiple investment firms share data from sensor devices which is compromised in some way, then a series of misinformed trading decisions could be made. Another risk arises if industry professionals have conflicting information, which could lead to scepticism about market intelligence and create instability in the markets.

Central Bank Digital Currencies (CBDC) are being explored by a number of countries as a way to improve payment efficiency and expand access to the financial system.

One of the methods to implement CBDC relies on distributed ledger technology (DLT), which uses cryptographic methods to store transaction data sourced from multiple entities. However, the involvement of multiple parties expands the number of points of attack available to malicious actors.

For example, a cyberattack could be launched on a CBDC by targeting a security vulnerability in one of the participating institutions, causing disruption to a country’s payment system. This would affect global trade, as well as hitting confidence and causing financial losses.