Potential of AI in Revolutionising Financial Services in the UK
Date: 29 June 2023
Author: FINTECH Circle
Have you ever wondered about the incredible possibilities that come with the explosion of AI tools and ideas? It’s like a gold rush of inspiration and limitless potential!
In times like these, I can’t help but think of a famous quote by Bill Gates about making predictions for the future: “People often overestimate what will happen in the next two years and underestimate what will happen in ten.” So, let’s dive into the amazing world of AI and its potential impact on financial services in the UK.
Even the legendary “Godfather of AI,” Geoffrey Hinton, has raised concerns about the speed at which AI is advancing and what it can achieve. One of the reasons for this accelerated progress is the development of Large Language Models (LLMs) that can learn faster and with less data and training.
In the past, AI required a significant commitment of resources and training just to reach basic levels of pattern recognition. It was slow and error-prone, making it a risky venture for businesses. Moreover, past learning approaches often led to unpredictable outcomes.
But why are we witnessing this incredible boom in AI possibilities now? Well, it all comes down to LLMs learning faster, which is quite hard to measure and explain. The power and fear surrounding AI stem from its remarkable efficiency, which isn’t yet fully understood.
Researchers at prestigious institutions like the Massachusetts Institute of Technology, Stanford University, and Google have discovered that modern LLMs are taking shortcuts in a way similar to human understanding.
For example, if you know that fire is hot, you can assume that an electric stove top is also hot by connecting existing information. This groundbreaking evidence shows how AI is making connections in a manner akin to human cognition.
All these advancements in AI sound incredibly exciting, but are they mature enough to be applied in the financial services industry today or in the future?
Well, it’s true that much of the enthusiasm arises from the unknown nature of what AI might bring and why. Financial services, on the other hand, thrive on stability and easily evidenced outcomes, making them somewhat skeptical of unpredictability.
The Financial Conduct Authority (FCA), the regulatory body overseeing our industry, is keen to embrace innovation but has also raised concerns and considerations. For instance, when it comes to insurance, using additional data and Machine Learning (ML) or AI to determine premiums for customers should only result in a better deal for them compared to what they would have received without these additional services.
As an AI expert, I’m thrilled about the potential AI holds for revolutionizing financial services in the UK. While there are still challenges and considerations to address, the advancements in AI and LLMs have opened up a world of possibilities. It’s an exciting time, and I can’t wait to see how AI shapes the future of financial services!
Revolution or Evolution? Let’s Take a Closer Look.
When it comes to major changes in the Financial Services industry, I often find that they can be categorized into two broad types: evolution and revolution. Evolution tends to be subtle, happening gradually beneath the surface. It brings about changes that influence how we work and opens up new opportunities for customers. On the other hand, a revolution is more abrupt and disruptive, even if it was predicted. Take the introduction of the new Consumer Duty or the Retail Distribution Review (RDR), for example. These were revolutions within our industry that demanded businesses to change how they operate, impacting expectations and ways of working.
Reflecting on Bill Gates’ quote, while the revolution in AI and LLMs is currently driving an evolution in Financial Services, with new opportunities and innovations emerging, we must ask ourselves: Is this a long-term trend? Will this evolution eventually lead to a revolution? Is there a point in time where we all need to make a change, if only to keep up with the pace of progress?
To gain a better understanding of where we stand on the spectrum between a pending revolution and a growing evolution in our industry, we need to explore the opportunities at hand.
Of course, peering into the crystal ball and predicting the future is always a risk, as both Bill Gates’ quote and the unpredictable nature of financial markets remind us (remember, investing carries financial risks, so make sure you understand them before investing any money!).
However, it’s clear that the potential of AI and LLMs in Financial Services is immense. We can already see exciting developments in areas such as customer service automation, fraud detection, risk assessment, and personalized financial advice. These advancements have the potential to enhance efficiency, improve decision-making, and provide tailored solutions to customers like never before.
So, whether we’re on the brink of a revolution or in the midst of a gradual evolution, the opportunities presented by AI in Financial Services are hard to ignore. As an AI expert, I believe that embracing this technology will not only keep us at the forefront of innovation but also enable us to better serve our customers and adapt to the changing landscape of the industry.
The Opportunities: Unleashing the Power of AI
One of the most promising areas where AI can make a significant impact is in data analysis and insight generation. Traditionally, number crunching has been a prime candidate for automation, and AI has proven itself to be exceptional at processing and detecting patterns or outliers within consistent data blocks.
As the speed of AI processing continues to increase and the cost decreases, we find ourselves in an exciting sweet spot. Now, we can swiftly process large, structured data sets with minimal training.
For instance, financial statements, which often have slight variations in format and approach, can be efficiently processed by AI to extract relevant context, saving valuable time. It’s important to note that AI isn’t meant to replace financial analysts but rather to support them by enabling bulk processing across the entire market.
Instead of a team of analysts with limited bandwidth manually reviewing a fixed number of companies and reports, AI can help identify interesting or unexpected findings, allowing analysts to focus their time and resources on areas that hold the greatest potential value.
This leads us to another captivating area: insights. My personal ambition is to witness the delivery of genuine and impactful insights to anyone who seeks them. There is a distinction between merely presenting data and truly uncovering an insight. Knowing that you fund your account at the beginning of each month is surface-level data, not an insight.
However, discovering that your deposits from August to November consistently outperform other months over a 5–10 year period provides an interesting and unique insight. It’s these types of “unknown unknowns” that drive focus and further investigation.
While AI may not have all the answers readily available, it can serve as a guide, assisting in directing efforts toward fruitful areas of exploration, much like the data processing capabilities I mentioned earlier.
The real power of AI will be unleashed when we combine these opportunities with others. Having insights is the “what,” but delving deeper to understand the “why” is when we enter truly powerful territory.
By leveraging AI’s capabilities in data analysis, we can identify patterns and correlations. But the real value comes from connecting these insights with additional sources of information and knowledge. This multidimensional approach enables us to gain a deeper understanding of the underlying factors driving those patterns. It helps us uncover the root causes, detect emerging trends, and make more informed decisions.
Imagine the possibilities when we merge AI’s data processing prowess with expert human analysis, domain knowledge, and a creative, inquisitive mindset. Together, we can unlock the full potential of AI in generating transformative insights and driving innovation in the financial services landscape.
So, as we navigate the evolving landscape of AI in Financial Services, let’s embrace the opportunities it presents, from enhanced data analysis to the discovery of meaningful insights. By combining AI’s capabilities with our expertise, we can go beyond surface-level observations and delve into the profound “whys” that will shape the future of our industry.
Get ready for a collaborative journey where humans and AI join forces to uncover the hidden treasures that lie within our data-rich world. The future is bright, and together, we can achieve remarkable advancements in the realm of financial services powered by AI!
Summary & Extracting Data: Unleashing the Power of Unstructured Information
Data analysis encompasses a wide range of areas, from structured data sets like financial reports to complex and unstructured information found in emails and various forms of communication. Understanding patterns in communications and identifying what effectively conveys a message is vital for achieving Consumer Duty, Good Outcomes, and Fair Value.
This is where AI showcases its capabilities in delving into the unknown, summarizing, and comprehending unstructured language-based information. Regardless of our opinions on emojis and the increasing use of images and videos in modern communication, all these forms of expression constitute unstructured data that carries meaning. Complex aspects like sarcasm, which may be universally recognized, aren’t uniformly understood and identified. Even in governmental contexts, individuals documenting conversations often seek confirmation regarding the intended tone.
Here lies the importance of investigating the accuracy of the shortcuts and internal learnings employed by LLMs. Do these models truly comprehend tone and intent across various forms of communication, be it written, spoken, in video format, or represented through images?
While ensuring communication quality is one example, the applications go beyond that. AI can assist compliance teams in their role as the second line of defense by monitoring vast amounts of communications, identifying tone and intent, ensuring fair treatment of customers, detecting and appropriately addressing complaints, and ensuring adherence to expected service levels.
Looking ahead, as AI models continue to improve, I envision applications in areas such as Know Your Customer (KYC) and Anti-Money Laundering (AML). Instead of relying solely on standardized questionnaires and limits, analyzing customer data can provide a more holistic understanding, offering a bespoke experience. Traditionally, bespoke solutions have been deemed costly and inaccessible to many companies. However, embracing bespoke operational and regulatory outcomes can actually reduce costs and provide customers with a better, more balanced experience, departing from a one-size-fits-all approach.
AI can assist teams in providing oversight by offering summaries of previous conversations, aiding financial advisors or agents in preparing for meetings. On the flip side, AI can also help customers by recalling past discussions, providing updates on documents, and ensuring that crucial matters are top of mind during infrequent interactions, such as annual meetings with a financial advisor.
Understanding and extracting insights from unstructured data is not only a demanding human activity that can deliver immense value but also a requirement for even the simplest tasks we perform daily.
By leveraging AI’s capabilities in deciphering unstructured information, we can unlock new levels of productivity, efficiency, and personalization. The potential for extracting meaning from various forms of communication is immense, opening doors to enhanced customer experiences, improved compliance, and more effective decision-making.
Embrace the power of AI and unstructured data, and let us embark on a journey where our interactions, whether written, spoken, or visual, become a source of valuable insights, creating a future where simplicity and effectiveness go hand in hand.
Conversational Flows & Assistance: The Rise of Dynamic Conversations
As we delve deeper into the realm of unstructured conversational data, we encounter an emerging field within AI. It involves not only processing and understanding data but also mimicking and generating dynamic conversations.
While chatbots are not a new concept, recent advancements in the field, coupled with cost reductions, are paving the way for exciting possibilities. Live chat, which has already undergone its own evolution, may be on the brink of a revolution.
Imagine an AI that can handle phone calls, comprehend tone and intent, possess contextual knowledge of help documents and escalation points, and even emulate human voice and expression. This could redefine the entire first-line support experience.
Companies that currently emphasize connecting customers directly to a human representative who can understand their needs in their own words might face disruption. While the idea of interacting with more switchboards pretending to be human is unlikely to thrill many customers, there may come a day when we are oblivious to such a change.
The concept of storing an FAQ document in an AI’s “brain” is not new. However, even with the current iterations of LLMs, we are witnessing the remarkable speed with which these solutions can scan through information and understand connections without requiring customers to go through a predefined set of questions.
How often have you encountered information that was documented differently than what you initially expected? AI can bridge that gap and provide relevant answers based on understanding the context.
While the notion of a human mimic on the other end of a call might be unsettling, there’s no denying that AI will continue to advance everyday communication. Presently, your email likely suggests ways to complete your sentences and provides prompts like reminding you about the attachment you mentioned or adding a standard disclaimer. We are witnessing a silent and steady evolution of AI in how we compose and review our communications.
As we move forward, AI will play an increasingly prominent role in refining our conversational experiences. The ability to understand and anticipate user needs, offer tailored suggestions, and assist in generating coherent and effective communication will become more seamless.
While the prospect of AI-driven conversations may evoke mixed emotions, there’s no denying the potential for increased efficiency, improved accuracy, and enhanced user experiences.
So, brace yourself for a future where AI-powered conversational flows and assistance become integral parts of our daily interactions. With each passing day, we move closer to a world where AI seamlessly integrates with our communication practices, making them more efficient, personalized, and productive.
Let’s embrace this evolution, while ensuring that human values, empathy, and ethical considerations guide the development and deployment of these AI-driven conversational tools.
Check-Ins & Behavioral Analysis: Tailored Assistance at the Right Time
Regular check-ins are crucial for security, compliance, data updates, and maximizing the value of a service. However, determining when and how to conduct these check-ins can be challenging.
Imagine having an AI system that understands what needs to be done, when it’s due, and even has insights into your behavior to find the perfect timing for prompting you.
We already see a glimpse of this in email campaigns, where AI analyzes when people tend to open and engage with their emails, optimizing the timing of message delivery. But this analysis of data is only going to grow more sophisticated. It needs to go beyond mere engagement and delve into deeper behavioral analysis.
Human behavioral analysis often focuses on outcomes—for instance, whether you funded your account after seeing a banner about the end of the tax year. However, we aspire for something deeper. Understanding the “why” behind our actions is always the ultimate goal of any analysis, and this is where AI has the advantage. It can explore vast amounts of data, test combinations, and identify patterns that we may not even consider exploring.
While it is commonly believed that people behave differently during a full moon (which has perpetuated the use of the term “lunatic” from the Roman goddess Luna), it is not scientifically true.
However, with better behavioral analysis that incorporates both structured and unstructured data around us, we can gain insights into our own behavior and how we consume new information.
Simple concepts like the fact that decision-making might be quicker in the morning than after lunch can make a significant difference in our lives. But for a company to offer a tailored service that takes these unique aspects of our behavior into account may have seemed unbelievable to the regular customer in the past.
However, this is becoming less like science fiction and more of a reality. Being prompted not just when it makes logical sense, but when we are truly ready for it, can have a profound impact on our decision-making and overall experience.
Imagine receiving personalized prompts and guidance precisely when you need them, based on your individual behavior patterns and preferences.
This level of tailored assistance, driven by AI’s analysis of comprehensive data sets, can empower us to make more informed choices, uncover hidden insights about ourselves, and optimize our interactions with financial services. While it may sound futuristic, this personalized assistance is within reach, and its potential to positively impact our lives is immense.
As we navigate the evolving landscape of AI-driven behavioral analysis, let’s ensure that ethical considerations, transparency, and user control remain at the forefront. The goal is to leverage AI’s capabilities to enhance our understanding of ourselves, provide timely support, and improve our financial well-being. Together, we can unlock the full potential of behavioral analysis in shaping personalized experiences that truly resonate with each individual.
Financial Advice: AI as a Wealth Manager or Financial Advisor
The question of whether an AI can effectively manage your finances and wealth is a topic that sparks curiosity and debate within the financial services industry. Some businesses have already started using AI to make significant decisions, with one notable example being a listed company that appointed an AI as its CEO.
This raises the question: Could an AI make better, more rational decisions when it comes to managing your finances?
However, finance is not solely about the money in your wallet. It encompasses a much broader spectrum, including how you earn, spend, share, vote, and interact with your money. It is deeply personal and intimate, revealing aspects of an individual’s life that few other things can match. Your bank statement tells a story unique to you.
Considering the various opportunities we have discussed thus far, and recognizing that there will likely be countless more in the future, there is no technical reason why an AI couldn’t make financial decisions on your behalf.
It is worth noting that professional financial analysts have an approximately 50% success rate, essentially equivalent to a coin flip. So, from a purely statistical standpoint, an AI’s rational decision-making capabilities could potentially outperform human analysts.
However, I believe that the personal nature of wealth and finances is a barrier to completely handing over long-term financial decision-making to an AI. The cost of financial advice can be high, but the value and peace of mind it provides often justify the expense for customers.
Furthermore, people are not consistently rational when it comes to finance, including myself. The unpredictable and emotional nature of the stock market further reinforces this notion.
Ultimately, the question is not whether AI has the capability to serve as a wealth manager or financial advisor, but rather whether individuals would be willing to fully embrace and rely on AI for their financial decision-making.
The personal nature of finances, the role of ego, and the desire for human interaction and guidance may prevent many individuals from completely entrusting their long-term financial decisions to AI alone.
As AI continues to evolve and demonstrate its potential in the financial services industry, striking a balance between AI-driven automation and human expertise may provide the best outcome.
Augmenting human decision-making with AI-driven insights and analysis can lead to more informed financial decisions while preserving the human touch and understanding of the personal aspects of wealth management.
In the end, the choice of whether to utilize AI as a wealth manager or financial advisor remains a personal one, driven by individual preferences, comfort levels, and the value placed on human interaction and expertise in the realm of finance.
Fact or Fiction: The Future of AI in Financial Services
The role of AI in transforming work processes and the financial services industry as a whole is undeniable. Currently, there is a competitive race to integrate AI and find the most effective applications for it.
However, it is more realistic to expect incremental improvements in analysis and processing tasks in the near term. The focus will likely be on operational efficiency and enhancing existing customer experiences rather than revolutionary changes.
Looking further ahead, around five to ten years into the future, we can anticipate more advanced AI models being used to assist with directing queries, supporting front office staff, and providing first-line support and sales.
These models will likely provide enriched summaries, highlight key focus areas, and offer better clarity in customer-centric language and terms.
The ongoing development of AI will undoubtedly continue to generate excitement and open up new possibilities within the financial services industry. However, what is crucial is the maturing of processes and understanding surrounding AI.
As the technology becomes more familiar and integrated as another tool in the industry, rather than an unknown force, that is when we will witness the true potential of AI shining in the context of Financial Services in the UK.
In summary, while AI will continue to drive innovation and offer new opportunities, its full potential in financial services will be realized when it is seamlessly integrated into existing processes and workflows, providing tangible benefits and enhancing the overall customer and staff experience.
About FINTECH Circle
FINTECH Circle is a global platform of more than 260,000+ Fintech entrepreneurs, investors, finance professionals, academic & government representatives, and solution providers. The company launched Europe’s 1st Angel Investor Network providing seed capital to the best fintech startups in the UK.
FINTECH Circle also runs courses, webinars & innovation workshops for finance teams and C-level executives and publishes fintech thought-leadership titles.
You might also be interested in
