Community
Legacy structures as an AI opportunity
Having overcome numerous challenges, digital retail banking is in its best-ever state. Online banking, instant transfers, quick onboarding and a variety of products make banking more convenient and accessible than ever before. However, if we dig deeper, we’ll likely discover that underneath the layer of modernity sit principles, procedures and systems from decades ago.
Of course, some of this is well justified. After all, banks are essential to global economies and stable foundations are critical. However, outdated organisational and operational structures are one of the reasons why banking as an industry is often singled out as the one that stands to benefit the most from wide-scale AI applications.
The data-centricity imperative
At most banks, we’ll find that a lot of activities are not particularly data-driven or consumer-centric. But transformation is inevitable since, time and time again, we’ve seen evidence (such as this Asian bank that achieved a 46% uplift in loan applications through machine learning and external data – see page 11) of how focusing on data and feedback brings benefits to all parties involved. And once that’s complete, the banks will be in a position to start truly benefiting from AI.
Smart automation
The vast majority of current applications of AI at banks focus on data. Classifying information, analysing documentation, looking for simple patterns and so on have been a domain of specialised solutions that excel in fraud detection, basic decision-making and more. At Bud, we believe that this is only the beginning.
If we take a closer look at many of the functions of a retail bank, a lot of them can be simplified to processing some data input, making decisions and generating output. This is how simple tasks work (e.g. accepting authorisation transactions), how operational actions are undertaken (e.g. an affordability assessment for a loan application) and how whole departments run (e.g. financial product marketing). And it happens to be that if sufficiently good data is accessible, many of those functions look ready for smarter automation.
Agentic banking is here
When we talk about smarter applications, we mean agentic AI applied in banking. A lot of banking activities have been successfully automated (it wouldn’t be possible to have contactless payments or open banking without it), but those implementations often lack ‘agency’.
The processes are fixed and do not adapt. In some cases, it’s the right thing to do, but in many more, it’s a limitation which prevents financial institutions from becoming capable of offering truly personalised services to their consumers while boosting operational efficiency.
At Bud, we believe that with the evolution of technology, we’re now in a position to plan for a future where it’s increasingly possible to define objectives and constraints and let AI models find the best strategy to achieve those.
While self-learning models and advanced data processing have been around for a long time, the rapid evolution of LLMs provides a crucial ingredient: the ability to provide an accessible bridge between how humans think (and—not a minor feat!—how humans regulate financial activities) and how data is processed.
Banks that integrate agentic AI into their operations will become more efficient, more competitive and better attuned to the needs of their consumers. The agentic banking platform that we’re building at Bud will enable financial institutions to realise these benefits, make significant cost reductions and get ahead of the competition.
Finally, the same principles apply to the consumer. We have reached the point where it’s feasible to build a deep understanding of individual finances and combine it with models focused on a broad range of objectives. An autonomous agent that, at all times, helps ensure that your financial objectives are met—a continuous control system that makes only good decisions as often as needed—is a great promise to everyone. It remains to be seen if banks will adopt it before a new wave of businesses spot the opportunity and capitalise on it.
This content is provided by an external author without editing by Finextra. It expresses the views and opinions of the author.
Nikunj Gundaniya Product manager at Digipay.guru
11 October
Priyam Ganguly Data Analyst at Hanwha Q cells America Inc
Fang Yu Co-Founder and Chief Product Officer at DataVisor
09 October
Alexander Boehm Chief Executive Officer at PayRate42
Welcome to Finextra. We use cookies to help us to deliver our services. You may change your preferences at our Cookie Centre.
Please read our Privacy Policy.