From fraud detection to risk management and customer service, artificial intelligence will fundamentally redefine the way banks operate and interact with their customers. What are the possibilities offered by this technological revolution? What are the risks associated with its use? Have the Principality's banks already moved into the AI era? An interview with Frédéric Genta, Interministerial Delegate for Attractiveness and Digital Transition, and Robert Laure, Chairman of the Association Monégasque des Activités Financières (AMAF).
Can we say that AI is already being used on a massive scale in the banking sector? Or is it still in its infancy?
Robert Laure: From a technical point of view, it's safe to say that the banking sector is an "early adopter" (1) when it comes to Artificial Intelligence. In fact, the banking industry has been quick to embrace this technological revolution. However, the regulatory and legal dimension is still in the construction phase. In this area, a series of texts are being debated at European and international level. I'm thinking in particular of the IA ACT recently adopted by the members of the European Parliament. This regulation is vital for the banking sector, which cannot afford to go down this road without a safety net. So today, the banking sector is making technical, exploratory and cautious use of AI. It is not being used to its full potential. However, I believe that this will happen very soon.
So you would say that the banking sector is clearly ahead of other sectors in terms of artificial intelligence?
Frédéric Genta: Absolutely, and for several reasons. First of all, the financial industry is immaterial by nature and requires little or no physical space. The purpose of finance is to understand the economy and its returns, and to allocate capital to the best possible places. It is also an industry with a very strong track record, as financial transactions have been digitised for several decades and banks have access to a great deal of data. Finally, it's an industry that has made a success of its digital transition, as demonstrated by the rate of transition from physical branches to digital applications. It is therefore a qualified candidate for Artificial Intelligence. What's more, the Finance and Insurance sector alone accounts for almost 18% of the country's GDP (IMSEE 2023). It is therefore undoubtedly one of the most promising areas for the adoption of AI in the Principality. In this context, banks are constantly looking for ways to improve their efficiency, increase their productivity and provide a better service to their customers, while trying to reduce costs. The implementation of AI tools should make it possible to achieve these objectives. The potential is considerable, but there are also three major challenges: to legislate quickly to align with the European framework and the major international standards, to train employees to encourage the adoption and use of AI, and to develop internally, with the help of experts, the right tools to transform the organisation and its processes.
Are banks already investing substantial budgets in artificial intelligence tools?
Robert Laure: It's part of the IT budget, and yes, it's a significant investment.
In concrete terms, how can this technology be used in the banking sector?
Frédéric Genta: In the financial sector, AI can help banks to reduce costs and increase productivity by automating administrative tasks. Broadly speaking, AI can be used in three areas. Firstly, in risk and compliance management. Machine learning algorithms analyse huge volumes of financial data in real time to identify patterns and abnormal behaviour that could indicate fraudulent activity. This technology, based on what is known as machine learning, makes it possible to identify potentially risky transactions. Based on the principle that the past can be used to predict the future, the technology can be used to guide the creation of "red flags" that are passed on to the compliance officer. The machine will never replace the human being, but it will be an indispensable support. The second area is investment, with AI or by AI. AI can help us to invest more effectively, but given the potential of AI, we also need to think about how we can take advantage of the exponential growth that this revolution will bring.
In other words?
Frédéric Genta: Banks can use AI to implement trading strategies. Algorithms analyse data sets in real time to detect trading opportunities and execute transactions with a speed and accuracy that surpass those of human traders. This integration of AI improves banks' ability to react to market fluctuations and optimise their trading operations. Customers may also find new investment opportunities in AI. For example, should they continue to invest in the "Fantastic 7"(2) that will drive stock market growth in 2023 thanks to AI? Should we invest in China or California, as each region has strong ambitions in AI? Finally, should we invest in 'traditional' companies such as Microsoft or Google, or on the contrary seek out the future Microsoft and Google of artificial intelligence via venture capital funds? These aspects are essential to a financial strategy in 2024.
What is the third area in which AI could be useful?
Robert Laure: In customer management and the way we interact with them. For example, we are seeing more and more chatbots powered by generative AI that answer customer questions in real time. They can provide information about products and services, and recommend personalised financial products based on customers' past behaviour, preferences and financial situation. This practice has been adopted by the JP Morgan bank in the United States, which has also automated customer legal checks. This ability to process massive amounts of data extremely quickly and efficiently is quite spectacular. And generative artificial intelligence, with its ability to produce unique content, is even more spectacular.
In the Monegasque banking sector in particular, is AI also being adopted?
Robert Laure: Most Monegasque banks belong to European and international groups that are already using this technology. So yes, in the Principality, AI has also penetrated the banking sector. Essentially, at the level of risk and risk control. Other tools, I would say more office automation tools, are also being used, such as the famous GPT chat system, or tools that can be used to take minutes of meetings, for example.
There is a lot of talk about AI for various reasons, pointing to the loss of jobs or the loss of humanisation. Do you sense any concern about this among employees in Monaco's banking sector?
Robert Laure: There is bound to be some concern, because we are facing an epochal change that has never been seen before in the history of humanity. The speed with which this technology is being deployed is phenomenal. When you look at the penetration of certain computer applications such as Chat GPT, with 100 million users just two months after its launch, it's spectacular. At one time, when we managed to reach one million users, we were already delighted. Today, we're on the billion-user scale. So I can't guarantee today that some jobs won't disappear. Perhaps in support activities. And even then. Human support will still be necessary. But other professions will certainly emerge as well.
Fréderic Genta: This change is irreversible, and the impact of AI is recognised worldwide. Monaco is of course fully aware of this, and we need to adapt. The aim is to seize every opportunity while guarding against risks, even if there is no such thing as zero risk. For finance in particular, the issues of performance, efficiency and security are critical, as we are touching on a strategic pillar of the Monegasque system, and there is no room for error. There are also many fears about the risk of losing human contact, job losses or the security of personal data. We obviously need to be vigilant, but we also need to be optimistic. Well-implemented AI will enable us to be more competitive and attractive in an industry that consumes very little square footage. AI is first and foremost an opportunity, and we need to take a positive view of it, because the potential is enormous for countries with limited space.
In 2024, however, the US consultancy Forrester predicts that ten banks will be prosecuted for their problematic use of generative AI, in breach of data protection regulations. What are the risks that banks are currently running by using generative AI, and how is this compatible with data protection?
Frédéric Genta: It's important to understand that there is such a thing as generative AI. Using a public chatGPT with bank data is very dangerous. Conversely, using a framed algorithm in a sovereign cloud with a space dedicated to bank data is perfectly feasible. The aim is not to use AI, but to do so professionally and legally. Banning AI in companies will increase the risk of illegal use. Economic players therefore need to start looking at this issue now, or risk incurring economic or legal risks. Success will require investment, talent, method and time. This process of moving towards controlled generative AI will be complex, but essential.
Robert Laure: For banks, the risks associated with the use of generative AI can be multiple: their security and confidentiality, which Frédéric mentions (loss, malicious use, access, sharing, etc); and also their exploitation in terms of compliance (potential bias and discrimination, the right to be forgotten, the RGPD, etc). The protection of this data in relation to generative AI will require voluntary action and measures by banks (security, governance, traceability, training), as well as supranational regulatory developments. However, given its potential, banks will have to adopt the use of generative AI in the near future.
(1) The term "early adopter" refers to a company that uses a new product, innovation or technology before others.
(2) The "Magnificent Seven": the GAFAMs (Google, Amazon, Facebook, Apple and Microsoft) plus Nvidia and Tesla.
credit photo : © Photo Iulian Giurca / L’Observateur de Monaco
