By Lucas Coope
Introduction
As we know, AI is having a growing impact on our lives daily as it develops in the world around us, and we learn how to use it to our advantage. This is no different in the finance world, as it is already a stronghold in large financial institutions and is making its way down the chain. Since its uptake, AI has rapidly expanded throughout the finance world, from governments, regulatory bodies, and banks to everyday SMEs. AI is now a central part of the finance world.
Governments globally have adapted their processes system-wide to include AI, which is no different in the monitoring and regulation of the finance sector. AI tools help to ensure banks comply with financial regulations with their capacity to process vast amounts of data and comb through them to identify suspicious activities, potential fraud, or other breaches. The tools will allow for an improved system and a fairer market. AI is also used by governments and large institutions such as banks for forecasting; by analysing macroeconomic trends, it can identify financial risk, improve tax collection systems, and aid fiscal policy decisions.
There are many other ways in which financial institutions use AI. One example is JPMorgan; later this year, they plan to expand the use of a tool dubbed 'Moneyball', which analyses decisions made by portfolio managers (Masters, B. and Schmitt, W. (2024)). Tools such as this can split opinions as they are intended to improve processes and allow improved analysis; however, they may place people under increased scrutiny in the workplace. Examples such as predictive models and data analysis demonstrate the deep integration of AI in the financial sector. It's only a matter of time before we experience the widespread influence of these major institutions on our financial interactions. Research conducted last year analysing 160mn US jobs found that “service sectors such as legal and financial are highly susceptible to disruption by AI, although job replacement is unlikely” (Murray, S. (2024)). The disruption mentioned is more likely to relieve us of tasks and streamline processes to allow us to improve productivity. There are some examples we use every day and do not consider the use of AI in them, such as credit scoring and customer experience (AI chatbots, etc.).
How may AI help Brokers?
In a word, streamlining. AI will allow for a much smoother and simpler process, with its capacity to process vast amounts of data much quicker and better than any person. By enhancing data analysis, brokers will be able to see more clients and process them faster, all with improved decision-making. As I mentioned before, AI is used in credit scoring, and it can consider this and other factors, such as transaction histories and behavioural data, in minutes. By using AI effectively, brokers will be able to focus more on the customer's needs and find a specific tailored package to fit each customer well and ensure the best outcome for all. Just as large institutions use predictive tools to analyse risk, brokers will be able to put similar tools to use when analysing a client's past behaviour; they may be able to predict the future needs of said client and create a bespoke package to fit just that. Furthermore, something you may like is the sound of AI purpose built to be used on routine tasks such as filling out applications, conducting credit checks and verifying documents, freeing up time for more critical decision-making tasks. It may also assist in the later stages of an application when it comes to satisfying conditions; often at this point there is confusion and lots of correspondence between brokers and lenders while trying to clarify and finalise the deal, this can often be a slow and time-consuming task. However, with the help of AI, brokers may be assisted in this process by using AI to clarify what is needed and how to obtain certain data/documents, this would massively speed up the process by removing the endless emails and slow correspondence leading to a quicker pay out.
How may AI help Clients?
In a way, help to a broker is help to their client, and consequently, this section will mostly just show the flip side benefits of the broker's use of AI. Significantly, AI will improve the processing speed of an application; due to the data processing capabilities, the time taken to consider each client will be much less, allowing them to receive results quickly. There are possibilities for its use in underwriting based on complex regression analysis, which could improve the assessment of credit risk for applicants (Koide, M., (2022)). Another benefit is that results achieved with this method will be better fitted to their needs; as previously mentioned, AI will assist in creating bespoke packages for clients, giving them a better financial outcome, and improving their business prospects. Furthermore, AI will remove any potential human bias in the decision-making process and focus on objective financial data; this opens opportunities for diverse groups and creates a fairer and more equal market.
In addition to the application process, AI may also help improve general business functions and eligibility for finance. For example, AI-driven accounting software can help with bookkeeping by handling complex tasks, ensuring accuracy, and improving tax compliance. Another function may also be improving cash flow management by predicting trends, automating processing and optimising payments. Finally, AI may be able to improve access to credit for people without a strong credit history; instead of using this as a determinant, it may be able to consider alternative data such as sales performance, customer reviews, etc., thus further increasing the equality of the market.
How may AI affect Lenders?
There are both positive and adverse effects to the introduction of AI into the lending process as you can guess. Firstly, lets focus on the positives, now as before there will be a similar theme to the previous two sections here as, generally, some help for one party is a help for all. On a whole, AI will just boost general productivity by increasing the speed and quality of data analysis, for example processing applications; it will be able to scan a document and highlight any points of interest before any work is done. Another form of the data crunching will be the credit scoring and analysis of alternative data sources to help create a better picture of the applicant. Therefore, using AI in the reception and processing of applications will decrease the area for mistakes and simplify the process for the salespeople/underwriters. Finally, AI will be a massive help with administration (with routine tasks being its forte) and so things such as fraud detection and identity verification can be done in seconds.
Although it seems like sunshine and daisies so far, it's not all great, at least not for lenders. The ability of AI to analyse vast amounts of data and, importantly, alternative data sources will allow larger banks to enter markets traditionally served by 2nd tier lenders, increasing competition. Furthermore, as good as it is, AI is quite expensive to implement purpose-built models, further disadvantaging the smaller lenders in favour of large institutions that have already begun to use it. Another large issue which may arise is ethical and transparency issues; as we have spoken about, AI could use alternative data and methods, which may lead to a different outcome than a typical lender would’ve come to. However, if the lender can't explain this decision, it could result in a world of problems, from simple explanations to issues with legality and regulation. Correctly interpreting machines' decisions is vital to be fully transparent and ensure firms stay true to their values and regulations (Eluwole, O.T. and Akande, S., (2022)). Finally, this may be a long way off yet, but we may reach a point where we become over-reliant on AI, which will only exacerbate the above issues as well as it’s use in underwriting. Analysing the eligibility of clients is not all simply down to numbers, there must be a certain level of instinct and intuition which is something we may lose in AI. Thus, until AI becomes emotionally and characteristically intelligent, us humans can keep our jobs, as we cannot simply rely on regression analysis and credit/sales performance to accept/decline a proposal.
Summary
As we go through this increasingly complicated information age, we need to move and adapt to the times, and AI is the next big thing; ChatGPT is just one example that is now world-renowned. Statista published an article predicting the market for AI tech to be over 1.8 trillion dollars by the end of the decade (Thormundsson, B. (2023)), so getting to know it now will be a huge advantage. As with everything new, it starts from the top and trickles down, and as we have seen, the trickle has already begun as AI has worked its way into all different types of finance and banking processes. Therefore, the way I see it, it is only a matter of time before we communicate with artificial brains daily.
Reference List
Eluwole, O.T. and Akande, S., 2022, July. Artificial Intelligence in Finance: Possibilities and Threats. In 2022 IEEE International Conference on Industry 4.0, Artificial Intelligence, and Communications Technology (IAICT) (pp. 268-273). IEEE.
Koide, M., (2022). AI for good: Research insights from financial services.
Masters, B. and Schmitt, W. (2024) ‘AI is promoted from back-office duties to investment decisions’, Financial Times, 2 June. Available at: https://www.ft.com/content/3d82ea9f-f040-47aa-9b9d-0be9decdbb14.
Murray, S. (2024) ‘AI in finance is like “moving from typewriters to word processors”’, Financial Times, 16 June. Available at: https://www.ft.com/content/c35ce925-d7b3-4920-a431-c4ca1aa33503.
Thormundsson, B. (2023) Topic: Artificial Intelligence (AI) worldwide, Statista. Available at: https://www.statista.com/topics/3104/artificial-intelligence-ai-worldwide/#topicOverview.
Comments