HSBC SG's Tancy Tan on balancing risk and innovation in AI adoption

COO Tancy Tan says the bank currently runs over 100 use cases for generative AI.

Generative artificial intelligence (AI) may be all the rage in the banking industry right now, but for HSBC, it is just another step to add under its growing expertise on the use and application of AI tech in banking services.

“In HSBC we’ve been using AI for almost a decade now across the bank, with over 100 use cases,” Tancy Tan, chief operating officer of HSBC Singapore, told Asian Banking & Finance during an exclusive interview at the Singapore Fintech Festival (SFF), held on 15-17 November 2023.

Tan shared that for HSBC Singapore, adopting generative AI is not just about the new use cases it offers, but also ensuring that they adopt and deliver the new service in a responsible and ethical manner.

“It’s really about balancing risk and innovation. As much as we all like bright, shiny toys, in HSBC, we want to ensure that AI deployments are geared to our risk framework and are compliant with the laws and regulations of each and every country,” Tan said.

Asian Banking & Finance spoke more with Tan to learn more about her thoughts on the emergence of generative AI, and how HSBC Singapore is exploring AI use within the bank.

What does generative AI offer for banks, especially for HSBC Singapore?
If you look at the generative AI technology, that itself offers new possibilities and use cases for HSBC. Computer-generated AI is about creating content, whether the content is about images, text and video. With that itself, you can imagine the number of use cases that could possibly happen, whether it’s optimising operational efficiency, whether it’s creating customer experiences. So AI technology, in summary, offers new ideas for us and as a whole for the banking industry.

How is HSBC Singapore approaching the adoption of generative AI and which lines of business are you focusing on for its use?
Maybe before we even talk about generative AI, I’d like to bring you back to AI. In terms of our AI usage right now, in our firm, it’s a very much mature stage. 

We’ve been using AI technology to create new customer products, and to really redefine the way customers look at banking. So for example, if you visit our booth (at the Singapore Fintech Festival), you would have come across a new product called AI Markets. AI Markets itself as a friendly chatbot, which improves price discovery.

In fact, I just had a quick demo of that. It enhances FX trading, it helps the customer to understand price discovery, to plow through research articles quickly by asking questions to the chatbot, and to make execution decisions. And it’s all facilitated by AI. That is how we redefine the way people think about banking.

Regarding generative AI, right now as we speak, HSBC Singapore has over 100 use cases that are running. The use cases range anywhere from code optimisation to customised marketing for the customers.
More importantly, we continue to invest in resiliency and security, primarily using generative AI or AI s code to improve our fraud detection and financial crimes.
 
With the rise of deepfakes and AI fraud, what can the banking industry at large do in order to ensure the safety of customers’ data?
With most emerging technology, there’s bound to be risk itself. While we see some exciting use cases for AI, at the same time, there’s a lot of focus around resiliency and risk in order to protect our customers’ data, and especially in the rise of security concerns and cybersecurity attacks. That’s why we’re spending a lot of time on the controls of AI and the guardrails around AI usage.

Within HSBC itself we’ve got an established approach to managing such risks. Whenever we roll out emerging technology, there’s a few key areas that we focus on to make sure that we deliver AI in a responsible and ethical manner. First is really around balancing risk and innovation. The second area is achieving a balance between product versus the market fit. All our AI solutions should meet customers’ expectations, and should be well-received internally as well as externally. Last is around ensuring that all AI deployment aligns with HSBC’s principles of ethical use.

Within HSBC, we are committed to growing the AI and data analytics standard. We have been organising a series of activities to make sure that business users understand not just the AI use cases, but the risks that come along with that. For our data scientists, we want to deepen their knowledge in terms of the right use cases, and how to do it in a safe and ethical manner.
 
What’s the outlook for generative AI in the coming year?
You’ll see [generative AI] continue to mature this year. I think in the financial services industry in the next 12 months, we’ll start to see more and more use cases leading to maturity. But coming along with that will also highlight the need for talent. And this is where really the commitment to grow AI and data analytics talent is critical. If not, we can’t push the use cases to realisation. That continues to be the key challenge for all the banks out there.

Stepping back from generative AI and talking more in the broader sense: what can we expect to see from HSBC Singapore in 2024?
Singapore itself is viewed as the launch pad for innovation, so we are quickly picking up digital solutions. Our customers are mainly international customers, so this is why you’ll see many of our solutions transcending across international boundaries.

We’ve announced Project Epsilon, which is a blockchain-based solution targeted in helping the real estate industry streamline their end-to-end processes with their suppliers.