Skip to content

5 min read
Mambu and Microsoft recently hosted a webinar titled: ‘AI in the lending experience: understanding the risks and opportunities for financial institutions.’

During the webinar, industry experts: Rashmi Sharma, VP of Engineering, Mambu and Christian Becker, Technology Lead Germany – Microsoft Global Partner Solutions, Microsoft sat down with Edward Harding, Director, Global Strategic Partners, Mambu to explore the growth opportunities and typical roadblocks to deploying new technologies such as generative AI.

This is an edited transcript of the conversation which you can watch on demand.

When you think about the different applications of AI, there are internal processes and internal operations, which are impacted by these technologies. What possibilities are you seeing from a front-office perspective?

: In the lending space, we hear that there are a lot of customers who complain their experience takes too long – for example, decisioning on whether their loans will be approved. This is something AI can help with by using the powerful data insights to help deliver a more personalised experience. Likewise with customer service and support.

Christian: Rabobank is a good example of this. They are using Microsoft’s Power Virtual Agent platform to handle around 60,000 to 80,000 calls per month without involving a human agent.

Another example is the German institution, Commerzbank, who have recently announced a generative AI and avatar-based experience for its customers. It combines the best of the personalised experience at a branch location, with the best of online banking, which is available 24/7 on any device.

Both experiences are great because they bring together generative AI at the front end, as well as having a good back-end to make sure the customer expectation is met. This is where the partnership between Microsoft and Mambu is a powerful combination.

How can AI be deployed around operational processes?

: We have some partners leveraging AI for back-office processes such as credit decisioning. One example is explainable AI, which is used for personal loans and credit cards. In this instance, AI provides detailed, understandable explanations of the decisions being made or the actions that are recommended. This is already showing results in terms of higher approval volumes and faster loan cycles.

Another partner of ours has developed an identity verification technology to provide customers with real-time answers to someone trying to open a new bank account. In this instance, if the decision is no, the system provides immediate feedback.

Do you have any thoughts on how organisations who don't have the sort of flexibility can easily adopt these capabilities? What do they need to think about?

: For traditional financial institutions with legacy core systems, adoption can be difficult. Change can take time. In AI, there is a huge growth opportunity with the development and deployment of apps, for example, to choose from to help you keep ahead of the competition.

We believe in the adoption of a composable approach by using technology that is able to speak to standardised APIs to help maintain competitive advantage without having to invest in a large-scale transformation program.

Edward: There are huge benefits – and competitive advantages - in this approach by having an architecture that is flexible, allowing you to experiment with new technologies on an ongoing basis. Build for agility and flexibility, rather than investing in bespoke builds or one solution that will be quickly outdated.

What should banks be thinking about in terms of what they’re currently required to do or what’s coming down the line in AI? And is there cause for banks to be cautious?

: Financial institutions operate in a highly-regulated environment and at the moment technology is ahead of regulation.

The question you need to think about is how far do you want to go before regulation catches up? It’s only right that regulators get involved as financial service providers manage highly sensitive data, which can impact people’s lives.

Christian: I couldn’t agree more. From our side, Brad Smith, Microsoft’s chief counsel is also calling for regulation of AI - simply because we don’t believe it should be the technologist who defines how AI is used – it must be a much broader discussion with governments and broader society.  The technology is so impactful, we as a society must be able to make decisions on how we want to use this - especially in a heavily-regulated banking industry.

On the other hand, we also don't believe we need to stop until there is regulation. Something in between is the right way – small, careful steps by focusing on use cases that don’t have such a high impact. The contact centre I mentioned earlier is just one example of this. Making sure the human is in control is important.

Edward: Fast forward ten years, where do you think we’ll be?

: AI is undoubtedly a major disruptive innovation. Fundamentally the way we work with technology is going to change – akin to life before and after the internet, so I can’t foresee where we’ll be in ten years from now, because of the rate of accelerated innovation. I do think we’ll look back and wonder why we even questioned whether we would use AI. It's going to be just a normal way of life.

Christian: My thoughts exactly. It’s difficult to grasp the exponential growth of this technology. We will either totally overestimate it, or totally underestimate it. Your glass ball is as good as mine, so I wouldn’t make a prediction, but what I end with is that we all need to think about what would happen if we’re seeing this growth, and how do we get ready for it? Can you really afford to ignore it? I think that’s the more interesting question.

Join Microsoft and Mambu for this panel discussion where we explore the growth opportunities and typical roadblocks to deploying new technologies such as generative AI.

Watch on demand
Webinar: AI in the lending experience

Share this post

Mambu Partnerships
Mambu Partnerships