This article was originally published in Fintech Herald and can be accessed here.
How are regulators seeking to deal with AI in the financial sector?: The advantages of AI in the banking sector
Artificial Intelligence (AI) is complex and constantly evolving. Still in its relative infancy, it has the potential to turbo-charge financial services and transform the way that they are delivered to customers. This is already happening, at an embryonic stage, as AI provides them with more informed and tailored products and services. Internally, it also delivers process efficiencies, enhanced cybersecurity and an increased risk mitigation for financial institutions.
To appreciate the impact of AI, and how far it deserves the moniker of a fourth industrial revolution, it is necessary to consider what AI is, what it can do, and how regulatory challenges can be met by financial service providers. Implemented properly with appropriate safeguards, AI can create significant benefits at both an institutional and customer level. As regulators increasingly focus on AI in financial services, further development of AI regulation is an inevitable corollary.
So how are regulators seeking to deal with AI in the financial sector? As more businesses adopt intelligent systems, it is necessary to determine standards so that their use is both responsible and safe. Accordingly, governments are moving globally to propose laws and regulatory bodies that will affect companies and financial institutions beyond existing legislation, which is still designed primarily for the previous generation of technology. In trying to strike the right balance, the challenge is to ensure that sufficient protection is afforded to those who use AI systems while emerging regulations do not limit the pace of technological evolution and impede the competitive advantage or progress of companies investing in AI.
Key drivers for the creation of potential new AI laws include:
- The limitations of existing data protection laws;
- The ethical implications of AI and protection of individual consumers and customers;
- Mistakes made by AI systems, e.g. facial recognition technology, autonomous cars or weapons; and
- The potential use of AI for criminal or other nefarious purposes.
It is useful to examine where AI and machine learning (ML) are already being used in financial services. Until a few years ago, hedge funds were the primary users of these technologies in finance, but we have recently seen ML applications spreading to other areas, including banks, fintech companies, regulators and insurance firms. From accelerating the underwriting process, portfolio composition and optimisation, model validation, Robo-advising, and market impact analysis, to offering alternative credit reporting methods, different use cases of AI and ML are having a significant impact on the financial sector.
The rapid deployment of machine algorithms to automate time-consuming, mundane processes is now widespread across the financial services sector, offering customers a far more streamlined and personalised experience. Uses vary, but a few key areas where continued development of ML is prevalent include the following:
- Financial Monitoring – enhanced network security, training systems to detect flags such as money laundering, which can be prevented.
- Making Investment Predictions – enables fund managers to identify specific market changes more effectively than traditional models.
- Process Automation – allows financial institutions to replace manual processes by automating repetitive tasks.
- Secure Transactions – detect fraudulent activity by analysing multiple data points which can go unnoticed by human intervention. It also helps with the process of false rejections that improve the precision of real-time approvals.
- Risk Management – ML techniques can enable banks and financial institutions to significantly lower their risk levels through analysis of huge volumes of data sources. Unlike traditional methods, which are invariably limited to essential information such as credit scoring, ML can analyse significant volumes of personal information.
- Algorithmic Trading – trading is an excellent example of effective ML usage in the finance sector. Algorithmic Trading (AT) has become a dominant force in global financial markets.
By deploying ML-based solutions and models, trading companies can closely monitor trade results and news in real-time to detect patterns that can drive stock prices up or down. This allows them to make better trading decisions. Machine learning algorithms can also analyse hundreds of data sources simultaneously, giving traders a distinct advantage over the market average. Other benefits of AT include:
* Increased accuracy and reduced chances of mistakes;
* Trades can be executed at the best possible price;
* Human error is likely to be reduced substantially; and
* Automatic and simultaneous checking of multiple market conditions can be enabled.
Other factors also come into play; the first of which is customer data management. The enormous volume and structural diversity of financial data from mobile communications, social media activity, transactional details and market data creates a significant processing challenge, even for financial specialists. Integrating ML techniques to manage such large volumes of data can deliver both process efficiency and the additional benefit of extracting real intelligence from data. AI and ML tools, such as data analytics, data mining and natural language processing, help to get valuable insights from data which, in turn, generates higher business profitability.
When banking and financial institutions use ML algorithms to analyse both structured and unstructured data, decision making can also become easier because of access to better information. Customer requests, social media interactions, internal business processes, and discovering trends (both useful and potentially dangerous) can be used to assess risk and help customers make informed, accurate decisions.
Another benefit is customer service enhancement. Using an intelligent chatbot, customers can get all their queries resolved: finding out their monthly expenses, loan eligibility, or affordable insurance plans, for example. The capacity of AI and ML models to make accurate predictions based on past behaviour makes them a great marketing tool. From analysing mobile app usage, web activity, and responses to previous ad campaigns, ML algorithms can help to create a robust marketing strategy for finance companies.
If the forecasts are right, the use of AI will increase substantially over the next decade and beyond. As part of that, the future of finance will be heavily influenced by emerging fintech with technology setting the stage for ever increasing competitiveness between banks and financial institutions.
Big global banks have already been investing heavily in AI for several years and that will continue to grow apace. It can be a challenge for smaller banks to keep up with the latest developments and some are understandably concerned about the implications if they do not. Facing limitations on capital expenditure and budgets to match the technological capacity of larger financial institutions, another wave of consolidation might therefore ensue.
IBOS expects more will happen in the financial sector as banks adapt to the next set of technological changes. Recent competition in the banking space has been made more exciting by AI and allied technologies, as well as by ML powered advancements. Overall, AI in banking has an exciting future. Leading banks that have prioritised strategic technological advancements with significant investment into AI applications will be able to serve their customers more efficiently, to set new performance standards and to increase revenue.