Generative AI is the next big thing in banking and FinTech. Just as mobile completely transformed the industry a decade ago, Generative AI is on track to do the same. Technologies like Dall-E, GPT-4, LLamMa, and PaLM-E have left the realm of research and gimmicks and become products like GitHub CoPilot, Midjourney, Adobe Firefly, and the most well-known, ChatGPT. 

While there are still limitations to generative AI, especially in a sector that's built on trust and security, it has the potential to revolutionise the industry. In this article, we'll explore how generative AI can be used in banking and FinTech to strengthen the relationship with users.

User Research on steroids 

User research is an essential part to banking and FinTech, and large language models like GPT-3 can serve as a powerful bridge between data and language. These models can comprehend, transform, and synthesize human language effectively, making them capable of scaling and speeding up user research tenfold. 

Traditionally, when analysing customer behaviour at a large scale, researchers tend to rely on quantitative metrics. The familiar 5-star scale, engagement rates and multiple choice questions. Useful methods, no doubt. But limited in depth of insights that can be extracted. 

The real value can be found in open-ended customer feedback. By analyzing a huge batch of open answers with Generative AI, you can effectively - and rapidly - extract rich insights, cluster information and identify patterns and trends that were previously hidden in your data batch. Additionally, given the ad-hoc nature of extracting these insights, you can take a more dynamic and responsive approach to user research as you swiftly adapt inquiries and explore new areas, based on the insights obtained from these large language models. 

Let's put this to the test. In this example, we used app store reviews of a banking app in order to interrogate the dataset:

Augmented personalisation 

There are already numerous ways in which we apply personalisation within banking apps to create a more engaging and user-friendly experience. Generative AI will only augment this personalisation. For starters, it will help us to personalise content for different personas more efficiently. Language, tone of voice and context play crucial roles in forging a personal connection and ensuring an exceptional user experience. 

While Generative AI tools won’t completely replace copywriters, they can certainly shift their focus from content creation to validation. This way, product designers can leverage AI-generated content themselves, leaving copywriters to fine-tune and perfect it. As a result, more time can be spent to accommodate additional personas, ultimately leading to increased diversification and a tailored experience for a wider range of user personas. 

Moreover, Generative AI will enhance personalisation in face-to-face interactions. For instance, before a bank clerk meets or calls a client, they can use Generative AI-powered language interfaces to swiftly review previous communications, such as emails, call transcripts and message threads. By doing so, the clerk can quickly identify the user’s current products and decide how to best steer the conversation to achieve maximum efficiency.

About chatbots & expectation management

For a long time, the holy grail of digital client interaction in the banking sector seemed to be chatbots. While AI-powered chatbots have promised to revolutionise customer service by providing quick, efficient, and personalised responses to various inquiries, we aren't there yet. 

One of the main challenges is that the GPT technology is an advanced autocomplete and not a knowledge base. It is trained to predict the most probable, not the most correct. This means that it tends to hallucinate, that it can be easily tricked. However, this does not mean that we should dismiss the potential of integrating this technology into the banking sector.

Just as with previous AI technology, we should stay transparent to the user about what they are interacting with.

Just as with previous AI technology, we should stay transparent to the user about what they are interacting with. By being transparent about the nature of the conversation, chatbots can help users set the right expectations and understand when to trust the information provided. Providing sources and links to support responses can also allow users to validate the information being shared.

Additionally, while AI chatbots can be the default flow for customer interactions, they should not be the only option. It is essential to provide users with a way to opt out and retain control over their banking experience. The world is full of exceptions that AI chatbots may struggle to handle, so it is important to take these into account to ensure we do not leave anyone behind.

Conclusion

We’re only seeing the tip of the iceberg when it comes to Generative AI in banking and FinTech. User research, personalization, and powerful chatbots are just a few ways that this technology will influence the industry. We're excited to uncover further advancements in AI that don't just automate the human touch but rather augment and empower it.