Money20/20: Elastic Bets on Context-Driven AI in Finance

At Money20/20 in Amsterdam, global fintech thought leader Dr Efi Pylarinou and Tim Brophy, Principal Solutions Architect at Elastic, had a clear message for banks weighing up the next phase of AI: the real challenge is not the model, but the context behind it.

Together, the pair discussed how financial services companies need to focus less on hype and more on the data, workflows and governance that make agentic AI usable in practice.

From RAG to agentic AI

The vocabulary around AI is shifting fast, Efi and Tim acknowledge. 

Concepts like retrieval augmented generation (RAG) are being reframed as part of a broader move towards agentic AI and what Tim calls “context engineering”.

“We’re maturing as an industry,” Tim says. “People have evolved to start talking about context engineering. This is essentially the same thing as RAG – however the consequences are higher than simply providing an informed chat response. It’s about providing really precise, highly relevant context into an agentic process so that an LLM can make good decisions based on what’s happening in any given workflow scenario.”

For Elastic, this evolution reinforces the importance of unstructured data and real-time retrieval. 

Tim adds: “I think anyone who’s just saying ‘AI-enabled blank’ is behind in terms of some of the messaging.”

Context over models

While much of the industry debate centres on foundational versus domain-specific models, Elastic’s approach focuses on the layer beneath.

“Even across all of these three model types, every single one of those still needs the information about the process that’s running at any given time to give that model what it needs to make a good decision in terms of the outcome,” Tim adds.

Efi points to a widening gap between hyperscalers and the rest of the market.

“What worries me is that hyperscalers not only have that rich data, but they have the resources – human, technical and everything – that they can go ahead and experiment with foundational transactional models,” she says. “But what about the 99% of banks and financial services players?”

Elastic’s answer is flexibility. 

With its way of operating, institutions do not need to commit to a single model strategy.

“You can pick and choose,” Tim responds. “And if you have the resources to train your own model to be so domain-specific that it is aware of your own processes and your own data, that is ideal. But, of course, not everyone has that luxury.”

Start small, scale impact

Despite the scale of AI ambition across financial services, Tim emphasises a pragmatic starting point.

“Customers come to us and ask, ‘Where do I start?’,” he begins. “What I like to say is: think really, really small. Don’t try to boil the ocean with broad AI and have the approach of just deploying AI to everything.”

Efi agrees: “I like the example of taking anti-money laundering and saying that we’ll tackle that, or onboarding a customer – KYC – and really go deep there.”

A practical example is fraud detection, where Gen AI can augment – rather than replace – existing models, as well as human capability.

“A fantastic way to leverage Gen AI in the fraud detection workflow is to shift left and use the AI to summarise alerts, find related alerts and really derive an initial case analysis,” Tim notes.

“This means the human analyst doesn’t have to then do that work themselves.”

Data architecture as the differentiator

As banks explore new AI-driven interfaces, Elastic argues that front-end innovation is secondary to back-end readiness.

Tim shares: “If your data layer is fragmented in silos – with different query languages across all of these silos of data – that process is going to be cumbersome and painful. It doesn’t matter what the interface is.”

Efi continues: “It’s interesting because for consumers and users, it’s all about the interface. But for businesses, it’s all about the behind the scenes. 

“Now, we are going beyond the back office deep into the plumbing. We can’t kick the can down the road on data architecture – it has to really be intelligent.

“That context is needed to serve us better.”

Elastic’s platform combines APIs, dashboards and emerging agentic tooling such as its MCP capabilities.

“The foundations remain the same – it’s all about the data,” Tim says. “We have to think about that data in the sense of speed-of-light retrieval and context generation.”

Compliance and explainability

With the EU AI Act set to introduce stringent requirements for high-risk use cases, explainability and auditability are becoming central to AI deployment.

Efi highlights the scale of the challenge.

“How do you make sure that this is auditable and explainable for at least six months? And the penalties for not complying with this are huge: we’re talking 7% of annual revenues or €35m, whichever is higher.”

Tim notes that there is a risk around the “black box” nature of models.

“There’s an element of this process that you cannot track and audit,” he asserts.

The answer is logging every interaction.

Tim continues: “What is really critical is not only to think about context, but also to think about this workflow and these steps – and logging every interaction, every iteration.”

For Tim, it goes back to RAG in the context of engineering: “What was the question? What were the prompts? What were the search results that powered the prompt? And then what was the answer?”

“I really feel that we’re in a very good position not only to power these agentic workloads, but also to provide the assurance of how these processes run and the explainability of their outcomes. 

“It’s an exciting time and we’re well positioned to benefit from some of these developments this year.”

Efi concludes: “These are exciting, challenging times, but with a lot of opportunity too.”