Databricks: Why Now's the Time for Finserv to Embrace Gen AI

Unsurprisingly, the financial services sector has often taken its time to leverage new technologies. Institutions must adapt to ever-evolving and extremely rigorous regulations and the time and cost it takes to implement a new IT solution. 

But with generative AI proving invaluable for even the most regulated industries, financial institutions now have the opportunity to maximise the value of their data to improve internal processes and evolve customer experiences.

Russ Rawlings, RVP, Enterprise, UK&I at Databricks, says the first step to leveraging generative AI tools, like customised large language models (LLMs), is entering with an optimistic outlook.

Gen AI: Improving productivity in banking by 30%

“The technology has the potential to improve productivity in banking by up to 30%,” says Russ. “Moreover, a recent survey found that 68% of the industry’s leaders believe that having a platform that enables the adoption of emerging technologies is of high importance - so it’s clear that generative AI is rewriting the future of finance as we know it. 

“That being said, if institutions want to capitalise on the technology, they have a lot to factor into their adoption roadmap - from leveraging the right data to complying with regulations to bolstering a workforce’s data intelligence.”

Smaller LLMs: They’re just as mighty

So, what’s the first thing to consider when businesses are looking to adopt an LLM? According to Russ, it’s all about size: “When people use the term “LLM”, they usually think of the huge consumer chatbots that have been at the helm of the AI conversation. But financial institutions need a smaller, more secure LLM."

And whilst the right size model is crucial, it’s also about leveraging a solution that is tailored to an organisation’s needs: “A financial enterprise’s model does not need to know anything about, say, celebrities; this data is irrelevant. The model needs to help organisations with what matters to them, like  determining risk, minimising fraud, or providing more personalised experiences for customers.”

Enter custom open-source models. But for an LLM to be tailored to a specific need, it must first be trained and reasoned on an enterprise’s proprietary data. “Customised models are actually more cost-effective to run due to their smaller size. Not to mention, the smaller and higher quality datasets will result in the model producing more relevant and accurate results”, Russ notes. 

“Imagine how efficient and productive an enterprise could be if its LLM could analyse a consumer’s buying behaviour and flag suspicious or fraudulent actions. Or, if a consumer is applying for a loan, the LLM could use the specific algorithms it was trained on to determine eligibility.”

So, how do these smaller, tailored options fare in comparison to their much larger counterparts? Simply put, it all comes down to the data. 

“The large, general-purpose models are trained on a much larger dataset, often composed of data scraped from the web. This includes all manner of information,  including irrelevant or poor-quality data, which has a huge impact on the model’s output”, Russ points out. 

“For example, it could hallucinate or produce inaccuracies, and in an industry as heavily regulated as finance, this would spell disaster."

Naturally, data quality is at the forefront of financial services. But the question then becomes: how can enterprises safely experiment with the technology, and unlock its benefits, whilst ensuring governance? 

“If institutions leverage their own LLM that is trained on their own data, that model belongs to them”, Russ explains. 

“They control what data goes into building their model, and what data is left out. Plus, they won’t need to share anything with a third party. As such, this approach ensures they comply with regulatory requirements, and can experiment within secure boundaries.”

On a broader level, Russ adds, this can facilitate in-house training of custom models across the industry, empowering every organisation to extract the most value out of a model that is powered by its own private data. 

Data intelligence as the foundation for organisational success

Evidently, a smaller, customised LLM can be invaluable for financial institutions. But it’s not just about generative AI; Rawlings notes that organisations can’t neglect the importance of data. 

“Good AI is always underpinned by good data and a solid understanding across the board. If the financial services sector wants to maximise the value of generative AI, then enterprises need to establish a strong data culture and build data intelligence as part of their overall data and AI strategies”. 

According to Russ, data intelligence looks like all employees - including non-technical individuals - having the skills, knowledge, and understanding to confidently use data. 

“This includes recognising how the enterprise stores and uses its data, as well as using data-driven insights to improve decision-making and innovation.” 

Not to mention, enterprises cannot overlook the need for an airtight data governance strategy to help them adhere to stringent legislation.

“The most effective way that financial institutions can cultivate data intelligence whilst complying with external regulations is by leveraging a data intelligence platform”, Russ explains.

 “This is built on a lakehouse architecture, and acts as an open, unified foundation for all data and governance.” 

Breaking it down further, Rawlings notes that a data intelligence platform is trained on an enterprise’s own data and concepts, so it’s tailored to an organisation’s exact needs. 

“For instance, the platform would be able to understand industry-specific jargon or acronyms, which then leads to more accurate and relevant responses. On the flip side, data intelligence platforms have an equal understanding of natural language thanks to the integration of generative AI. 

“So if, for instance, a non-technical user inputs a query in this format, the platform can still deliver highly relevant output.” 

Giving an entire workforce the confidence and ability to query an enterprise’s data and pull insights can have a huge impact on how the enterprise operates and innovates. But this isn’t the only benefit - as Russ highlights, a data intelligence platform also acts as a secure end-to-end solution, meaning no third-party platforms are needed for data analysis. 

“Keeping everything on one secure platform can help institutions best comply with legislations and ensure data privacy, but it also facilitates a smoother approach to governance and accelerates data-driven outcomes.” 

So, no matter where an enterprise sits in the spectrum of financial services, one thing is clear: every single one must prioritise meeting industry legislations and requirements, whilst figuring out how to simultaneously meet their own needs. 

To Russ, the solution is simple: “A data intelligence platform meets both internal and external needs at the same time. It’s undoubtedly a win-win situation.”

**************

Make sure you check out the latest edition of FinTech Magazine and also sign up to our global conference series – FinTech LIVE 2024. 

**************

FinTech Magazine is a BizClik brand.