Name: Bryan Evans | Neo4j
Uploaded: 2024-06-11T16:00:00.000Z
Duration: 12 min 54 s

Clips
News
More from AWS Financial Services Symposium

Bryan Evans

RVP Sales

Neo4j

Three insights you might have missed from theCUBE’s coverage of the AWS Financial Services Symposium

Cloud spending in the financial services industry has been on a growth curve, and Amazon Web Services Inc. has gained momentum in this sector. For financial services companies, broader cloud adoption has led to a heightened focus on data and new ways to organize and manage vast amounts of information to create intelligent solutions using AI.How the financial world is leveraging data to transform operations and increase customer value was a central focus of theCUBE’s coverage of the AWS Financial Services Symposium. TheCUBE, SiliconANGLE Media’s livestreaming studio, provided insights through interviews with company executives, customers and partners. (* Disclosure below

Neo4j enhances fraud protection through knowledge graph databases

Artificial intelligence is a double-sided coin for security, proving a boon to both attackers and protectors, but Neo4j Inc. is continuing to to harness AI when protecting financial service organizations from fraud. The company creates knowledge graphs, which store networks of data, specific to an enterprise, allowing them to find fraud that might otherwise go undetected in an organization.Neo4j’s Bryan Evans discusses the company’s collaboration with AWS on a graph database in cloud

play_circle_outline Neo4j's approach to GenAI initiatives and its use of knowledge graphs.

play_circle_outline Scaling Data Insights in Financial Organizations: Neo4j's Graph-Driven Augmentation Engine for Enhanced Outcomes

play_circle_outline Neo4j's future plans: managed graph database, collaboration with Snowflake and Databricks for graph analytics.

Info

Bryan Evans | Neo4j

Bryan Evans

RVP Sales Neo4j

>> The AWS Financial Services Symposium, presented by theCUBE.

>> Good afternoon technology fans and welcome back to New York City. We're here at AWS Financial Services Symposium with a power-packed day of coverage. My name's Savannah Peterson, you're watching theCUBE. More importantly, I am joined by a fabulous guest from Neo4j today. Brian, welcome to the show.

>> Thank you for having me. Happy to be here.

>> Hey, it is a pleasure. You're new to the Neo4j family.

>> I am.

>> Congratulations.

>> Thank you.

>> Very exciting. Neo4j is widely adopted here in the financial services space, and you're on the sales end. Can you give me a couple of the use cases or things that you're seeing out in the wild?

>> Yeah, of course. Yeah, so we've been working with financial services organizations for a while now. We've been one of the leading solutions that they use for use cases that can be broad, but more specifically fraud, money laundering, customer 360 use cases, also identifying cyber attacks where we're able to identify bad actors as they digitally interact with a financial services organization and also as they transact online, we're able to identify those fraud patterns. More specifically, what we're seeing a lot of right now is use cases around GenAI initiatives and how Neo4j can fit into those use cases.

>> Let's talk about that. What are some of those use cases?

>> Yeah, so the power of LLMs can be massive with financial services organizations, but if we're being honest, there's been a slow uptake in GenAI apps that are actually in production in financial services organizations. And that's due to three main reasons. The first is lack of accuracy on responses that you get from LLMs, lack of explainability and lack of context, all three of which are big problems for highly regulated organizations. So the approach is grounding, that's really what we're trying to do in the GenAI stack. And so what we're doing is we're building a knowledge graph specific to an enterprise and their data set that will augment how the LLM interacts with the end user, provides answers and gives them back to the end user. So the workflow would be, you ask the LLM a question, the LLM will then interpret that question into code that will query the knowledge graph, which again is an external connected network of a financial organization's data set. The knowledge graph will then return back specific explainable and explicit answers, the LLM will then provide the response back to the end user in plain English. So this is commonly known as RAG, Retrieval Augmented Generation. So that's really what we're doing. And knowledge graphs can really drive high levels of accuracy, deep explainability, context, reduce hallucinations, and that's why a lot of leading service providers are recommending knowledge graphs as really an essential part of the GenAI stack.

>> And this is a big shift from legacy systems, yeah?

>> A big shift, yes. So we've been working with financial services customers to augment their existing systems for a while now. So let me give you an example, maybe back up a moment. So we've been working with financial organizations for a while. We've got over 75 of the Fortune 100 that we work with now, and a growing list. And so years ago, a financial services organization, this predates my time, obviously, but they came to us and said, "Hey, we know that there's more fraud in our network than what we're able to identify. How can we work with you guys to help see that, identify it and mitigate it?" And so what we did was a couple of things. We put three pieces of critical information into a knowledge graph. We put their customer information into a knowledge graph. We put their product hierarchy into a knowledge graph and their credit card transactions all into one data set where we track the data itself, but also the relationships and patterns within that data. What that ended in was finding a lot of fraud that was otherwise unknown within the organization. But a really cool byproduct of that use case was because we had customer data, product data and credit card transactions in the knowledge graph with Neo4j, it turned into a product recommendation engine as well to help drive revenue growth for the financial -

>> Which makes a lot of sense, but what a nice little silver lining if you're going to be making that investment.

>> That's right. That's exactly right.

>> Clever. Is that type of approach and solution synonymous across those 75 of the 100, or are you seeing different approaches depending on the institution?

>> Yeah, so it's similar from a perspective of, we're looking to augment existing systems. So graphs are already everywhere if you really think about it. So financial transaction systems, IT networks, social media networks, search engines, public transit systems, these are all examples of graphs that already exist today. And our belief in what we're doing with a lot of companies is providing deep insights from structuring data. And so a lot of organizations, they want to be data-driven, they aspire to be data-driven organizations, but many companies are missing half the value of their data because they're not able to track and keep records of relationships and patterns within data sets. That's what we mean when we say graph database. That's what a knowledge graph really is, where we're bringing in data from disparate systems and keeping the data in the relationships among data points as equally important data sets. So this is what's really driving a lot of insights and outcomes at scale that companies otherwise would not have seen before.

>> Yeah, wow. So that's pretty interesting. I got to ask this question because now I'm curious, to the 25 that aren't working with Neo4j, what do you want to say to them?

>> Jump on board.

>> Yeah. Yeah, I was going to say, I feel like that's kind of an obvious choice-

>> That's right....

>> at this point. One of the interesting things as we evolve, and it's a conversation we're having across verticals with GenAI, is trust. And especially when it comes to financial services, we're thinking about a bank, how do you ensure that these experiences you're designing are trustworthy?

>> Yeah, it's a great question. So it goes back to the retrieval augmented generation strategy, the grounding strategy where we can really tell the LLM, "Provide me the answer back that I'm looking for exclusively based on the context that I'm providing you." So we can really put some fencing around what the LLM is responding back to you. And again, it's based on a knowledge graph from the enterprise's data set that we've built that's going to augment the response and provide that deep explainability and context. And that's how we're able to drive trustworthiness because in a regulated industry, most of our customers, they can't afford to have inaccurate answers from an LLM, never be able to explain.

>> No, it's stable stakes, it's not-

>> That's right.

>> There isn't the level of forgiveness that we see in certain other verticals. Yeah, I don't need the bank messing with me. That's the last-

>> A hundred percent....

>> that's the last thing I need. Looking forward, what's next for Neo4j?

>> Yeah, so great question. We're excited about a couple of things. We've been working on our managed graph database in the cloud for a number of years, AWS is a key partner for us in that offering. But what that does and what that unlocks for our customers is speed of deployment on Neo4j, speed of ingest, much greater levels of scalability and spinning up multiple databases for specific customers. So that's a huge part of our growth over the next five to 10 years. Something additional that we're really excited about as well, also a big part of the AWS network is with Snowflake and with Databricks, we're working with them to provide analytic data and relationship data on data sets that aren't yet in Neo4j. So we're putting graph analytics-

>> Oh, cool....

>> on top of external data sets to drive insights and actionable outcomes from information and data that's not yet in Neo4j, which is a huge part of our growth story, I think.

>> Yeah, I could definitely see... I mean, that feels like it opens up an entirely new market of opportunity. Oh my goodness. Taking off your Neo4j hat for two seconds, what excites you most about where we're at, our Gen AI future and what could come?

>> Yeah, I think what excites me most about Gen AI is unlocking, again, insights that we didn't think was possible before. It is an amazing enablement and training tool. It certainly can be for people coming into an organization like myself. The productivity benefits that you can gain from Gen AIs is massive as well. We use that within my sales organization. So we use AI tools to perform tasks that can help us be faster to respond and eliminate a lot of the basic tasks that a salesperson would do to give us more time to work with customers to understand what they're trying to do and accomplish.

>> Yeah. I mean, I think you're absolutely accurate. When you're talking with customers, I'm curious, are they... I mean, the financial services industry is not known for its rapid technological adoption, lovingly. And so do you feel like we're in a new era now where this industry is adopting tech a little bit faster than they have, say, historically, given the GenAI revolution?

>> We feel so, yes. Yeah, what we're seeing is a huge uptake in interest and engagement around, "Hey, we've got these GenAI initiatives that we're working on. Here's the plans that we want to build, how can we work with you and how can we get these into production faster?" And where, again, knowledge graphs are a key component of that story is we provide the explainability, we provide the accuracy. We can help drastically reduce hallucinations for LLMs. So yes, we're seeing a huge uptake in the speed at which financial services want to bring to market their productionized versions of GenAI applications.

>> Awesome. Are there any use cases, since you do get to interface with a lot of customers in your role, are there any use cases, that you're allowed to share with us, that have surprised you perhaps or been interesting?

>> Yes. Yeah, a couple. So I think the most typical that we see right now are customer support use cases, know your customer, customer 360 use cases. Those are typical, but also some of the more cutting edge ones have been, specific to outside of financial services, some of the manufacturing use cases that we've seen around GenAI applications has been really interesting.

>> Can you give us an example?

>> Sure. So if you've got a manufacturing system that can have no downtime, and you've got all these manual documents of how to repair a certain piece of machinery, how do we feed that into an LLM so that a maintenance worker can ask the LLM a question? And again, we've built the knowledge graph based on the customer's documentation to provide the 100% accurate, specific explainable answer back to the maintenance person who can get that machine back up and running to reduce downtime, to reduce any loss of revenue, and just overall help to keep the business moving. So that's specific to manufacturing, but in financial services, we're seeing customer 360 use cases, customer support use cases, some really interesting bleeding edge use cases around investment banking, which is very cool as well-

>> I'm excited about that....

>> with externally available information, but also proprietary information as well feeding into the LLM. So those are very cool to work on.

>> Yeah, that side of it is going to be really fascinating, I think in the next couple of years, quite quickly, as everything ramps up. What do you hope to be able to say, when I interview you next time, let's say a year from now, that you can't say yet today?

>> I would love for you to ask how many GenAI applications do you now have in production with your customer base?

>> And you're able to say, "Many, many, many."

>> That's what we're really driving towards. Exactly.

>> That is awesome. Wow. Well, Brian, thank you so much for taking the time today. This has been an exciting conversation. Congratulations again on your new job. Say what's up to Steve for me.

>> I will.

>> And from the whole CUBE gang, we really appreciate it. And thank all of you for tuning into our absolutely power-packed coverage here at AWS Financial Services Symposium in New York City. My name's Savannah Peterson. You're watching theCUBE, the leading source for enterprise tech news.