Please turn your device

How Artificial Intelligence Is Changing the Face of Banking

You’ll see a lot of artificial intelligence talk on our blog. The reason? This new breed of technology has imposed its will on every facet of the business world, it’s virtually impossible to ignore it.

Today’s angle on AI technology will run us through the different ways it’s altering the banking playing field. Banks have enjoyed a long tenure of uninterrupted dominance, monopoly and unquestioned reign. The most recent economic crisis coupled with the rise of financial technology have opened the door to change and disruption. The entire banking sector is being “attacked” from multiple angles and the main artillery unit behind this attack is artificial intelligence.

Chatbots & Retail Banking

Do you recognize any of the following services?

  • Opening a savings account
  • Authentication
  • Credit card cancellation

You do? Great.

We’re not magicians but we guess that when answering this question you took a deep breath and exhaled in desperation. We’ve all been there. The long queues at the bank branch, the uncomfortable conversations over the phone with your “dedicated bank specialist”, the awkward silences and infuriating questioning that leads to no clear-cut answers.

Enter AI in the form of chatbots.

Chatbots are one of the fastest growing applications and use cases of artificial intelligence. In the case of retail banking, chatbots take on the role of a bank employee and perform tasks in a faster more efficient manner. How are they able to do that? Through ‘semantic engines’.

As explained by Croud.com:

Semantic search is a deeper, more comprehensive search technology that allows chatbots to “understand the intent of a searcher’s query within a specific context. By learning from past results and creating links between entities, a search engine might then be able to deduce the answer to a searcher’s query…”

The more you interact with a chatbot, the more it learns about your preferences, chat history and intentions, making it able to personalize your banking experience.

Never discount the importance of the overall experience. It matters more than banks have cared to believe. When interacting with your bank, getting the answers you need is only half of the job. The other half rests on how you get those answers.

Chatbots allow customers to engage in a stress-free, information-rich, instant conversation that gets them from point A to point B in the most effective and efficient manner possible. The personalization of the experience is not only a user experience gimmick. Upon gathering enough data, advanced chatbots can offer insightful and helpful functions such as tracking your spending habits, help you manage your budget or provide a credit score.

Personalisation is a big component of what the future holds for retail banking with or without chatbots. Just have a look at what Poland’s first fully online bank mBank has managed to achieve. Using SAP Predictive Analytics software, the bank has managed to anticipate and serve its customers future needs and requests, offering them a truly personalized experience.

Using data they already have on customers, the bank is able to extract insights and provide them with offers that are relevant to their specific preferences.

“SAP Predictive Analytics has allowed mBank to discover individual customer preferences and identify the next, best activity for our marketing efforts. Now we are able to initiate more direct conversations, resulting in a better understanding of our clients on a personal level.”

– Bartosz Witorzenc, Strategic Initiatives Manager

Retail Banking Department, mBank S.A

Chatbot benefits do not stop there.

Take online forms for example. Tedious, repetitive and at times, confusing. Chatbots can auto-fill user information, provide swift answers and guide users through form completion in the blink of an eye.


What about accessibility? Chatbots are at your service through a multitude of online channels. From the bank’s website or app to social media handles and messaging apps, chatbots are available within a single click.

The rise of chatbots and their real-life use is not mere conjecture. Numbers are beginning to back the theories behind chatbots, showcasing that people, especially younger generations, are positively responding to this new breed of AI systems.

A recent survey by Humley, an AI powered cognitive assistant platform showed that “two thirds of those surveyed felt that an AI-powered chatbot would be useful in assisting them and 44 percent would rather communicate with a chatbot than a real person, assuming it could answer their questions as reliably as a live agent.”

ubisend, a technology company that creates AI-driven chatbots also released a chatbot statistics report with interesting findings. The report references data and statistics from a pool of highly credible sources such as HubSpot, Drift, Juniper Research, Adobe, Mindbowser and Business Insider. Some of the most eye-catching numbers are:

  • 48%of consumers would rather use live chat than any other means of contact
  • 35% of consumers want to see more companies using chatbots
  • 57% of consumers are interested in chatbots for their instantaneity
  • 47% of consumers would buy items from a chatbot
  • 90% of bank-related interactions will be automated by 2022

Still not convinced? How about an MIT & Genesys Technology Review reporting that 80% of respondents saw measurable improvements in customer satisfaction, service delivery and contact center performance.

These numbers are a testament to the growing popularity and use of chatbot technology. AI is infiltrating the ranks if retail banking in a very meaningful and measurable way, leaving marks of huge potential and promise.

Risk Management

A recent study by McKinsey Global Institute partner Michael Chui and McKinsey alumnus Sankalp Malhotra, reported that reported that AI shows significant value in a lot of business areas. Risk management ranks high with 51% of respondents claiming it’s one of the key areas AI can have significant impact.

How much of an impact? Just have a look at the latest GARP/SAS survey. The numbers about the biggest benefits expected from AI over the next three years tell you the whole story about the relationship between AI and risk management:

  • 78% Faster insight from data
  • 77% Reduced manual tasks
  • 77% Improved decision-making
  • 73% Higher productivity
  • 66% Lower operating costs
  • 66% Product quality/customer experience

Artificial intelligence and risk management are a natural fit. In what sense? Artificial intelligence possesses cognitive capabilities such as data mining, machine learning, and natural language processing that perfectly complement the unstructured data associated with risk management.

A 2015 International Data Group study estimates that roughly 90% of data generated today is unstructured, making the need for AI even bigger. Data nowadays are much more than numbers in rows, columns and spreadsheets. What AI has managed to do is go through the data and make inferences that help decision-makers anticipate and proactively manage risk.

Financial fraud detection is a very good example of how AI affects risk management.

Traditional fraud approaches have always been characterised by a level of simplicity that ended up causing more problems rather than give solutions. Let’s use a wire transfer as an example. A bank would set their “fraud indicators” at a specific amount (e.g. $20,000) and any transfer exceeding that amount would get flagged and investigated.

The result? Too many false-positives that caused the bank time, resources and money to double check and cross-reference.

How does artificial intelligence help in this scenario? Cognitive analytics will not only help with identifying fraud but learn from the adjustments and alterations made by human operators. Once a false-positive has been “fixed”, AI will be able to take that parameter into account next time, becoming “smarter” and smarter with every single fraud instance and scenario.


A stellar example of how far AI has progressed  in the world of anti-money laundering, KYC and compliance is the growth of fintech companies of this specific niche. Just look at Arachnys, a cloud-based solution for assessing financial crime customer risk and CDD, KYC, and EDD compliance. The London-based company is harnessing big data in a way that cuts costs, increases turnaround time and makes life easier for compliance departments of major financial services institutions.

This 10-minute podcast interview with the company’s CEO, David Buxton, is eye-opening on how cognitive data and artificial intelligence are shaping the future of finance and how banking institutions are managing risk internally.

Apart from risk management within banks, AI has had an impact in the investment and hedge fund branches of the banking sector.

Human bias is one of the biggest factors that move the needle in the hedge fund and investment world. Ulterior motives, error judgments and subjective forecasting have always been the blind spots that have hampered those industries.

Artificial intelligence offers a solution through its advanced learning algorithms, machine learning capabilities and problem solving capacity. People that want to invest can simply submit their risk-tolerance levels and investment preferences and the algorithm will give them a portfolio suggestion that’s not skewed by extenuating factors.

Workforce

Talking of chatbots and how they can automate basic banking functions, one can assume that AI will eventually replace people that are currently tasked to do those jobs. The reality is that using AI in the banking sector will not only enable overall efficiency, innovation and growth within the workforce, but will allow the redeployment of employees in higher-value areas of the business.

This is not a case of ‘man vs machine’ but rather a case of man aided by machines to perform at higher frequencies. The combination of AI and the human brain can become the biggest unique selling point and competitive advantage for banks going forward.

AI’s ability to interpret huge numbers of unstructured data coupled with human creativity and emotional intelligence can provide solutions to complex challenges, develop new products and services, and break into or create new markets. Contrary to popular belief, the banking workplace of the future does not see people losing their jobs but rather utilise their skill sets in more meaningful ways.

According to an Accenture Survey Report, Banks that invest in AI and human-machine collaboration at the same rate as top-performing businesses could boost their revenue by an average 34% and their employment by 14% by 2022.

The real challenge for banks will be integrating AI in the workplace and blending man and machine in the most effective way.

Gone are the days where technology worked in isolation and people supplemented the remaining work. AI needs to infiltrate the workforce from the top-down. Embracing and ultimately applying AI technologies needs to become a core part of the organization and be instilled in the culture of banks.

If numbers are any indication of what’s to come, it seems like banking executives and employees are indeed seeing the value deploying AI in their organization. The aforementioned Accenture Survey used a sample of 100 CEOs and top executives and more than 1,300 bank employees to reach some interesting numbers.

  • 77% of banks plan to use AI to automate specific tasks to a large or very large extent in the next three years
  • 51% believe human-machine collaboration is important to achieve their strategic priorities
  • More than two thirds believe AI will improve workforce productivity
  • 28% of the employees surveyed reported that they work with intelligent technologies for more than 50% of their time

The numbers reveal a very positive appetite and attitude in welcoming AI into the banking workplace/workforce. People see the value but are they prepared for the changes this will bring? Some jobs will be lost, others will morph into something new whilst we will also witness the birth of new positions with hybrid responsibilities.

Can we interest you in some guesses of what these new roles might actually be? How about a new position for a Machine Learning Engineer? A Bot Designer? A Deep Learning Scientist?


Roles will evolve and the fancy names are not the only interesting thing about them. The scope and use of human skills will become more meaningful when augmented by intelligent technologies.

Mono-skilled roles will become multi-skilled roles, generalist roles will transform to specialised roles and data-oriented roles will gain a slither of creativity. Take a sales support specialist or loans application manager for example. The introduction of strong AI will help these roles develop requiring skills that are centered analysis, evaluation and establishing deep relationships with customers.

The bulk of the administrative work will be done by AI, allowing the human talent to employ and grow its decision-making abilities.

A prime example of such an application is KEB Hana Bank from South Korea. Their eMortgage service is a real-time, fully digital mortgage. It takes 1.5 days from enquiry to funds release. Augmented reality is used to understand the property and neighbourhood, using the personal data from the applicant and market rates to make an official offer.

Just think of how many manual processes AI managed to overcome and resolve in this specific example and scenario, freeing up the human employee to make insightful decisions on the mortgage.

The organizational structure within the company will also have to change. We’ve talked about the organizational structure revolution in the workplace in one of our previous blogs and the pairing of AI with banking shares some of those similar traits. The structure will now be more flat and AI will take away a lot of the repetitive work. Employees will therefore be empowered, given more autonomy and decision-making power.

Employees will be more free to experiment, be themselves, feel valued and valuable and not be restrained by the administrative, mundane aspects of their job role. Processes and communication within the bank will be redesigned to accomodate fluidity, moving away from traditional functional and organisational constraints. The idea is to build an agile workforce where computer science and humans create new sources of value throughout the bank.

One of the most important factors in moving the workforce into the AI era is education. Employees need the resources to be able to develop their skills and not only work with ai-powered computer systems but also become familiar with the entire digitalisation spectrum of the banking services.

Banco Santander has been at the forefront of this movement, showcasing immense interest on the AI front. The Spanish Bank has been going through a technology transformation over the past couple of years trying to adapt to a changing banking sector. The bank has invested in various artificial intelligence (AI) startups, but has also taken steps to spread AI knowledge within the business.

Its digital academy does not only offer employees the resources to master new methodologies and acquire new skills but it also has a deeper purpose: changing the culture and mindset of workforce.

Recap

How will AI change the face of banking? In ways that are countless and unfathomable to the human brain. The important thing here is to understand the significance, importance and true potential of the term artificial intelligence.

AI is not merely a new technology. It is a force that will impact life as we know it the same way the car, the Internet and the airplane have in the past. The entire foundation of the banking sector is changing and with it all of the departments and branches of the business.

The keys to successfully transitioning the banking sector into the AI era is embracing it at the top level of the organisation, set the right tools in place to make the transition and understand that a machine learning algorithm is more powerful when coupled with a trained and skilled employee.