Financial Services industry has always been in forefront of leveraging technology, digitally transforming and AI is no different. We noted AI initiatives in this sector being leveraged by organizations to not just create new products and services, but also help them transform processes, infrastructure and redefine their existing business models.

The areas of focus been around Risk Management, Customer Experience Management (primarily in Customer service, Customer Acquisition) or Business Process Re-engineering. Some common areas where AI is being leveraged for rapid gains or early results include:

Fraud Detection via Transaction monitoring: The ability to analyse large volumes of transaction data and in real-time, flag suspicious activities, be it atypical spend patterns, high-value transactions, irregular transfers etc.

Credit Risk: Analysing an individual’s credit history, and other relevant data points to determine their creditworthiness and assess the risk of default. This is also leveraged by marketing teams to upsell relevant services/products to eligible customers.

Portfolio Risk Assessment: Identifying potential risks and opportunities and making informed investment decisions based on analysis of market trends & data.

Compliance (e.g. AML compliance): An extension of the spend and transfer patterns can help uncover AML-related compliance issues and help avoid financial penalties.

Insurance fraud: Help detect fraudulent insurance claims by analysing data such as customer information, medical records, and claim histories. AI-based systems can detect patterns of fraud and flag suspicious claims for further investigation.

Document processing: Extract and analyse data from various documents such as loan applications, invoices, and financial statements to autofill backend applications, automate manual data entry, and reduce processing times.

Chatbots: NLP Chatbots – which are now more powerful, thanks to Large Language Models like ChatGPT, can provide customers with real-time support and assistance by responding to customer queries, providing information about products and services, and helping customers troubleshoot issues.

Customer Acquisition & Retention: Using predictive analytics to anticipate customer needs, creating targeted marketing campaigns that appeal to individual customers to personalized offers, content, and messaging to proactively offer relevant products and services and increase the likelihood of conversion.

Whilst ChatGPT has created a lot of buzz, the use of AI has been leveraged within Financial Services for a while now. With the advances and the quantum improvements which Natural Language Processing (NLP) together with greater advancement in AI, we can expect to see a lot more use cases around customer services, customer acquisition and more.

In our discussion we see some pointed to personalization and automation as key areas in which they are looking to leverage AI. e.g. delivering more personalised experiences to their customers, tailored to their individual needs and preferences from investment recommendations, financial planning advice, insurance policies, or even personalised banking services like loans, credit cards etc. Automation would continue with the routine tasks that financial services firms perform, such as loan processing and underwriting, KYC etc being looked at as levers to save time & cost, while at the same time being delivered with consistently high quality.

Some hurdles in this race

However, there are hurdles to implementing AI, in the financial sector:

  1. The first is obtaining high-quality data which is critical to the success of any AI model training process. The ability to collect the relevant data from various sources such as databases, APIs, web, and other third-party sources; scrub incomplete, inconsistent, and erroneous data to make it ready for analysis may require expensive processing given the diverse technologies at play.
  2. Secondly, access to talent. The technology has been democratised and largely open-sourced, good talent is scarce. What is perhaps good is that AI is getting easier to implement. ChatGPT is an excellent example of how the technology is being made available in an easy-to-use form.
  3. Additionally, AI systems can perpetuate or amplify biases and inequalities, leading to unfair exploitation and possible social implications. Financial institutions have to consider and ensure that their AI systems are transparent, explainable, and fair.Singapore is amongst the front runners at creating a model AI Governance framework (“Model Framework”) when it launched Model Framework at the World Economic Forum in Davos. The Model Framework provides the frame of reference in translating ethical principles into practical recommendations that organisations could readily adopt to deploy AI responsibly. Singapore has also launched A.I. Verify – the world’s first AI Governance Testing Framework and Toolkit that allows organisations to assess the alignment of their AI governance practices with the Model Framework aligned to industry examples & practices.
  4. Lastly, financial institutions are highly regulated and must ensure that AI solutions comply with various local regulations, including data protection and privacy. The Monetary Authority of Singapore (MAS) has worked with public and private sector organizations to develop principles for the use of AI and data analytics, which could serve as a useful framework for companies looking to implement AI solutions.

The reason to be optimistic

Overall, we see interesting times ahead, and as AI technology continues to improve, we are likely to see that financial services firms will find even more ways to leverage this technology to improve their operations, enhance customer experiences, and even creating a greater impact in the lives of people and organization who use their services.