Banks are currently facing challenges on many fronts, which is why they are looking forward to the AI revolution, hoping that it will help them get back on track. On global capital markets, many banks are quoted at a discount of up to several dozen percent to book value, and approximately three-quarters achieve a return on equity that does not cover this cost (McKinsey & Company data). Banks also face intense competition from fintechs. To cope in this challenging environment, they must learn to use artificial intelligence to rebuild their operating models.
AI is a chance for success
The amount of data available in financial services enables the use of artificial intelligence in banks' digital strategies. AI is already helping banks reform core business functions (operations, finance, risk management, product development and sales). In the future, it may also open new sources of revenue that are difficult to imagine today. According to a survey conducted by The Economist Intelligence Unit – 77 percent. banking executives believe that the ability to unleash the potential of artificial intelligence will determine the success or failure of banks in the coming years.
Banks were among the first enthusiasts of artificial intelligence. In fact, AI systems have been used for decades to improve risk management processes, mitigate losses, prevent fraud, retain customers, and deliver efficiency and profit gains. However, until the COVID-19 pandemic, the use of artificial intelligence in financial institutions was random and focused on selected areas. Nowadays, more and more banks are taking a comprehensive approach to implementing artificial intelligence. McKinsey also predicts that the motivation is, of course, financial: AI has significant potential to create added value. Globally, banks can gain USD 1 trillion thanks to it. annually. In turn, Business Wire estimated the value of artificial intelligence in the global banking market at USD 3.88 billion. in 2020 and estimates that by 2030 it will be USD 64.03 billion, which means a CAGR increase of 32.6%. in 2021–2030.
“Traditional” artificial intelligence
AI-based automation allows banks to simplify operations and reduce costs by identifying manual, mechanical tasks performed by employees and replacing them with computers. This way, they can focus on more difficult activities that require human intervention. In turn, machine learning techniques (a subset of AI) allow people employed in banks to perform tasks that were previously beyond their control. For example, prediction and recommendation models can suggest a good time to invest in specific assets and warn of a downturn in a given market. Thanks to this, banks can better manage their investment portfolios. By analyzing huge amounts of data, they discover patterns that would be elusive to humans. This facilitates more accurate and faster decision-making. Going further, pattern recognition provides insight into processes, which is helpful in an area that is the pillar of banking – lending. Currently, many banks rely mainly on credit history. Meanwhile, artificial intelligence can verify how a potential borrower handles money, even if he or she does not have a solid credit history. Additionally, analyzing data for patterns is used in counteracting fraud and cyberattacks, as well as in financial forecasting and risk management.
GenAI comes into play
The year 2023 was marked by transformative generative artificial intelligence. The GenAI (generative artificial intelligence) fever has not spared banks that are trying to take a step towards implementing this technology. According to a report by McKinsey & Co, on an annual basis, GenAI can increase banks' operating profits by 9-15%. The potential comes from the ability to create content based on the analysis of large amounts of data, including text, images, video and code. In practice, the strength of GenAI will be hyperpersonalization – and this is the key word – for the user. Thanks to natural language processing, it can draw conclusions from unstructured data sets, such as historical interactions with bank services. In this way, it can also become an excellent pro-efficiency solution, while improving user experience in banking and financial services. Virtual assistants will analyze customer data to understand their preferences and needs and then provide the necessary support to users by answering their queries in real time and in a natural, conversational style. Employees – investment bankers, analysts and advisors – will continue to perform supervisory functions to ensure that results meet expectations in terms of accuracy, precision and regulatory compliance. They will be equipped with tools that allow them to personalize investment strategies and assess transaction risk.
Banks are also likely to apply GenAI to legacy AI applications to improve their performance, as the digitization and automation of customer-facing processes generates a digital data footprint that the tool can use to fine-tune the services it offers. There is a consensus among experts that when thinking about GenAI, banks should not limit themselves to automation, process improvement and cost control. It can influence operations related to customer contact and revenue in a way that has not been possible with previous “incarnations” of artificial intelligence. Still, banks are prioritizing automation in back-office functions. The vast majority are waiting for further development and testing of GenAI before they prioritize its use in front office operations.
Due to the prevalence of sensitive and confidential data in banking, artificial intelligence, especially GenAI, poses certain threats to both banks and their customers. As lawmakers around the world regulate AI, banks may face financial penalties if their applications lead to incidents such as customer discrimination or data leaks, or fail to meet security and privacy requirements.
Banks are working internally or with external partners on GenAI solutions, but they do so with their usual conservatism. Risk management with this type of tool is, by its very nature, still in its early stages. Sooner or later, however, banks will need to develop effective controls as GenAI technology matures. The key conclusion is that in a highly regulated industry like banking, there must be a balance between the speed of implementation of AI solutions and the need for appropriate regulation.
As for the users of banking applications themselves, it is worth mentioning the risk that they will overly rely on AI tools without having sufficient knowledge about their mechanism of operation. As a result, they may not know what is happening to their money or use products that they do not understand and which may prove to be unfavorable for them.
Over time, banks should develop a comprehensive business vision that includes a full portfolio of innovations and respond flexibly to changes in the face of the rapid development of artificial intelligence. There is no doubt that the use of AI will simplify the user experience and reduce the complexity of banking operations. Despite the inspiring prospects that AI technology opens up in terms of improving the profitability of banks and the quality of customer service in financial institutions, its implementation into banking products is associated with challenges, some of which are still unconscious.
Financial Observer – open license / Observerfinansowy.pl
