In recent times, generative artificial intelligence technologies have become increasingly ubiquitous within the banking industry. finews.asia takes a look at some of the use cases being implemented.

In the past year or so, generative artificial intelligence (GenAI) has been all the rage, including in the banking industry where many players are rapidly exploring opportunities for deployment. 

The benefits are well-documented, most notably from improved productivity. According to a report by McKinsey Global Institute, GenAI could add the equivalent of $2.6 trillion to $4.4 trillion in value annually across 63 use cases analyzed. Banking is expected to be a major beneficiary with $200 billion to $340 billion in new value created.

Internal Use Cases

In a report by Capgemini, 21 percent of financial firms in 2024 said they had enabled genAI capabilities in some functions and locations while 3 percent have enabled it in all or most. This is quadruple the level of 2023 when just 6 percent enabled any genAI capabilities at all. Thus far, the focus for practical application has been primarily placed on internal usage to support employees.

OCBC, for example, rolled out GenAI access to all 30,000 employees globally in October 2023 and it developed its own ChatGPT-based capability. Today, 4 million decisions per day are powered by AI and this is projected to reach 10 million by 2025. 

At DBS, over 20 GenAI use cases are being implemented, including a new virtual assistant to help deal with customer queries. The so-called «Customer Service Officer» uses voice telephony and speech recognition capabilities to transcribe queries in real-time, make live searches to retrieve information as well as conduct post-call documentation and pre-fill service request fields. 

External Exploration

External use cases are also being studied, though with more caution compared to internal usage due to the involvement of clients or other stakeholders. 

At Julius Baer, its Singapore-based innovation lab developed a large language model use case for clients to access all types of content such as market views. The project is now being incorporated by the head office into the bank’s strategy for scaling. Through a sandbox, Standard Chartered has explored external use cases such as custom market creatives that target specific clients based on their behavior.

Data Quality

Despite the optimism, there are also downsides in the deployment of GenAI technology. One risk is the quality of data and the effect it may have on the output.

«Company-wide AI success hinges on getting a handle on data quality,» said Pascal Wengi, APAC managing director at Avaloq. «It is also essential to proactively prevent errors and biases to ensure that the outcomes remain fair, compliant and reliable. Financial institutions using AI must be committed to maintaining the integrity and effectiveness of AI-driven solutions by continuously evaluating data quality and monitoring for errors and biases.»

Governance Risk

The use of governance frameworks can be useful to manage such business risks. At DBS, a framework called «PURE» (purposeful, unsurprising, respectful and explainable) has been employed as effective guardrails in addition to a senior-level, cross-functional committee to deliberate on ambiguous cases.

«As companies continue to experiment with the use of AI, it is imperative for business leaders to carefully consider governance frameworks to manage material risks around using AI,» commented Lim Him Chuan, DBS group head of strategy, transformation, analytics and research. «Venturing into AI without considering its potential ramifications […] would be reckless.»

Job Displacement

As is the case with every trend of significant technological adoption, there is a wave of new jobs created as well as jobs lost. According to a Citi report, the financial sector is at the greatest risk of experiencing AI-driven job displacement with 54 percent of jobs within banking and 48 percent within insurance face high potential for automation. 

«As companies adopt AI, it is equally important to ensure that employees are not left behind. AI will inevitably change the nature of work, including the types of jobs and skills that will be in demand,» Lim added.

«Upskilling, now more than ever, must be an urgent priority for all organizations. While workers can take individual ownership of acquiring these new skills […] companies also have a responsibility to prepare their people for this impending reality.»