NYSE Media Week CXO Series, theCUBE + NYSE Wired | Sonny Singh, Oracle

Clips
More from NYSE Media Week CXO Series, theCUBE + NYSE Wired

Sonny Singh

Executive Vice President and General Manager

Oracle

play_circle_outline Ensuring Security and Efficiency: Oracle's Approach to Fraud Detection in Financial Cloud Services

play_circle_outline Cloud adoption challenges for banks

play_circle_outline Potential for AI agents to transform business outcomes

play_circle_outline Oracle's approach to showcasing AI use cases and cost structure.

play_circle_outline Maximizing Success with Data, Cloud Tech, and Financial Services: Interview with Sonny Singh

Info
Transcript

Sonny Singh, Oracle

Sonny Singh

Executive Vice President and General Manager Oracle

At NYSE The Cube, Sonny Singh, executive vice president of Oracle Financial Services, discusses challenges faced by banks in modernizing technology infrastructure. Singh highlights struggles with legacy systems hindering adoption of new technologies like AI and cloud computing. He emphasizes importance of security and compliance in banking industry amidst increasing cyber threats. Singh explains how Oracle's cloud infrastructure addresses challenges, offering cost-effective solution with focus on security and automation. He discusses potential of AI in revolu... Read more

explore Keep Exploring

What organization was Oracle's first customer and what was their focus on when it comes to security? add

What were some of the reasons why the banking community initially had reservations about adopting cloud technology? add

What capabilities does Oracle's stack have in terms of unlocking and utilizing data from applications, metadata, and business logic through Fusion, and how will multiple agents work together to create business outcomes utilizing this data? add

What factor is crucial in ensuring that customers are inclined to adopt AI use cases within business applications? add

What is the reason for the company's cost advantage and ability to help customers leverage data effectively? add

bolt Powered by CUBE AI

Sonny Singh, Oracle

search

>> Everybody welcome back to our CXO series here at NYSE The Cube, plus NYSE Wired. Two days a wall-to-wall coverage with CEOs, CISOs, Chief executive officers, and more. I'm really excited to have Sonny Singh here who is executive vice president and general manager of the Oracle Financial Services business. Sonny, great to see you. Ringing the bell this morning, how'd that feel?

>> Oh, it's always exciting. Firstly, really glad to be here. It's always exciting to come back to the NYSE. And this morning, we actually rang the bell to celebrate our customers in financial services solutions that we have launched, and I was delighted to bring some of our customers along for that. We had Toronto Dominion Bank, we had the Navy Federal Credit Union, we had Citizens Bank, we had Guardian Life, and other customers that we were able to interact with. And by the way, just a year ago, we were celebrating our 10th year anniversary for our listing on NYSE. So this is right on the heels of that. So confluence of great events.

>> Yeah, Safra was here last year ringing the bell as I recall.

>> Correct, yes, absolutely.

>> And of course we just saw you guys out at Oracle Cloud world.

>> Yeah.

>> It seems like quite some time ago, but so much has happened since then. But it was really amazing show. I mean, there were many, many tens of thousands of people there. I would guess 25,000 plus. Larry of course spoke, Safra spoke, you had customers up on stage. It was quite remarkable. I was there as the analyst, and we went through the full stack.

>> .

>> And it was great. And one of the most interesting pieces of that stack was what's happening at the application layer, and then of course specifically your world of banking. So what do you see happening in banking?

>> Wow. I mean, it's one of those things that you always hear about banks just being technology companies with a banking license, right?

>> Sure.

>> So we have a tremendous amount of intersection in the way we think about technology and the way banks think about technology. I think probably one of the industries where that is the most relevant component of a bank's infrastructure, if you really think about it. Outside of a checkbook, a credit card, and a debit card, there's really no other physical product in banking, right? Everything is technology. So we fit in there like a hand in glove kind of arrangement. So we are seeing so many things continuing to evolve. The fact that banks have this huge reliance on technology, made them a very early adopter of technology. Unfortunately, that came with the downside, and the downside is that they built a lot of their applications multiple decades ago in the '70s and '80s. And this was the time where the internet wasn't a commercial idea, you didn't have any mobility or mobile technologies, you did not have any social technologies, you did not have internet of things, IoT, any of the transformational technologies that we take for granted now, those didn't exist at the time. So now we are here looking at AI, and AI was just a research paper back then. So a lot of these banks are sitting on legacy infrastructure that was built at a time when none of the fast-moving transformational technologies were available. And so they have a huge challenge ahead of them to modernize. And this is one of the key things that we are seeing. The second thing we are seeing with banks, is because of that, their competition is changing. They're getting their key services around payments and lending, eaten away by fintechs, by big techs, and a whole new brand of banks, neobanks and over-the-top banking organizations. So we are seeing all of this happen. Again, I'm giving you a long answer. The other thing we are seeing is regulations and compliance. These are things that are just, they keep them up at night, including cybersecurity. A lot of their investment is going in stabilizing those areas. So the opportunity to invest in transforming the bank gets diminished. So these are all of the things that we are seeing banks deal with right now.

>> It's funny, you say the things we take for granted. I'll give you a quick little story. My credit card just got hacked, like today. And I was down in South Carolina at a wedding this weekend. Of course my credit card was all over South Carolina, so somebody obviously grabbed it. And I know this bank, I know they're running Oracle. But you think about fraud detection 10 or 15 years ago, you'd have to go through your statements, maybe you'd find it, maybe you wouldn't. I got a text, I got an email, I got a phone call, spent a little bit of time on the phone with them, I'm having a new card sent out, they're blocking that. And so I know it's Exadata and Oracle database, and Oracle application. So thank you...

>> ....

>> by the way, for protecting me. But we take that for granted, the time to catch fraud now, like you said, in the distributed world, it gets compressed down to nothing. So you've got to know your customer. A lot of challenges. You've got to know your customer, you have to be able to respond to them. You have to have an infrastructure to be able to verify who it is and make sure that that transaction actually is fraudulent, and do so in near real time. Those are big challenges. How do you see banks adopting things like cloud? Because early on, they were very much sort of against cloud. Actually some tried to build their own cloud and realized maybe that's not such a good idea. And even Larry, early on, people misunderstood I think what he said about cloud early on. Well, it's just servers and databases, et cetera. Now you guys have a world-class cloud. How are banks leaning in to OCI generally and cloud?

>> So there's a lot packed into that question and I'll try and unpack it in sequence. So you mentioned security and fraud detection, and I'll come back to the cloud leveraging that a little bit. So firstly, at Oracle, we are obsessed. Thank you for passing on your thanks. But at Oracle, we are totally obsessed about security. I don't know if you know this, but our first customer was actually the CIA...

>> CIA. Oh, so you know....

>> and other three-letter intelligence agencies that nobody talks about. And so we got grounded in the notion of mission critical and highly, highly super-secret confidential information. So we had to build for that from the get-go, which means that you need an obsession with everything from making sure that the infrastructure is secure, the data is secure, both in rest and at motion, the applications that use that data are 100% secure. And the most important thing which people always forget, is the vast majority of security problems come because of human error. So one of the things that we'll see AI doing is create a lot of automation in the management of this infrastructure. And that automation will actually reduce. And we are at the forefront, for all of our data centers are leveraging autonomous technologies. This means that we minimize human interaction, we minimize the potential of human error. So that's the whole idea of security. You mentioned fraud and the move to cloud, so let me hit that just very quickly.

>> Please.

>> So in my view, there was generally a little bit of reticence from the banking community on the cloud. And they had reasonably good reasons for that, right? They had the notion of data security and privacy that the regulators looked at very tightly. Then you added on data residency requirements. There are a number of issues around letting the data of your citizens, including their financial information, travel around the world, or travel in areas that are away from where they can personally secure it and be accountable for it. So cloud had some headwinds. So what we saw move early to the cloud was just infrastructure as a service. So we saw test and development environments, peak loads for non mission critical applications. That was the first thing to go. The next thing was other applications like human capital management, financial management. So that was the progression. Now we are seeing that core banking applications, things that are core operating technologies for the bank, they have started to move to the cloud. And I think there's a number of things driving that. In COVID we saw the number of digital customers go 3X, and number of digital transactions go 6X. Now if you see that kind of volume, you can't track what you just experienced this morning, which was fraud. How do you have enough human beings to monitor and then investigate this huge volume of transactions? And the bad actors use this kind of advanced technology to perpetrate those criminal activities. So banks have to adopt large-scale computing environments, artificial intelligence, to start to combat those issues. And for that, they need to start to leverage cloud technologies to be able to burst the capacity, etc.

>> There's a lot of chatter in the community. So those companies that really only play on-prem, talk about how the cloud is so expensive. Those companies that are only playing in the cloud, talk about how the cloud is so cheap. You're in both. And so, one of the things people may not understand about Oracle, is every single data center in Oracle, its entire cloud, is identical. The hardware, the software, the middleware, the processes, they're all the same. So why is that important? That's important because that drives costs down, it drives efficiency. You guys have the Autonomous database and now it's 23ai, and so you're driving a lot of automation there. How much of a factor is cost and how much of a delta do you see? Is it pretty comparable? I'm sure there are other considerations, especially when you get into AI. I'm interested in what you're seeing with... You know? Everybody talks about large language models and then there's small language models that are specific to your data, you don't want to share it with the LLM vendor.

>> No.

>> So two questions. What are you seeing on cost, cloud versus on-prem? And how are customers thinking about their private data? Are they putting it into the cloud? Do they want it to be on-prem? How do you see that?

>> So that's actually a superb question, and let me answer that in sequence. So firstly, I believe cost is a factor from two different angles. In general, when you look at a cloud provider who's doing things at scale, and at a kind of factory setting, if they're not able to drive down your cost, it's kind of meaningless, right?

>> Yeah.

>> I mean, cloud gives you probably two or three key benefits. One is the speed at which feature density becomes available to you. All these big cloud companies are innovating, and they can roll out that feature to you a lot faster than you actually bringing that over into your data center, deploying it, testing it. You offload that. Secondarily it's the cost factor. The scale that a cloud provider can bring you is definitely something that you can take advantage of. I think Oracle has two or three very key advantages in this space. One is, when we built our second generation cloud, we used some very unique technologies. And that has allowed us to deliver compute and networks, and storage, at a very, very aggressive price point. And as we move into the world of AI, where running these large language models, running inferencing from that large language models is super expensive, even a fractional change in cost makes a huge difference to your customer. So we are very much at an advantage in that regard, given our, what we call next generation cloud infrastructure. Secondarily, our obsession with security. I mean, we are definitely kind of the hallmark of security in this industry. The third thing is we have mission critical workloads. Core banking, core retail merchandising, core telephony, core applications for health. The most demanding workloads are testing our software, testing our cloud infrastructure, and that actually makes us uniquely differentiated in the marketplace, because we get to experience our customer's problem because we are building not only infrastructure for them, but actual business capabilities.

>> Yeah, financial services, healthcare, and three letter agencies are the three probably most...

>> Demanding....

>> intense when it comes to security.

>> Yeah, most demanding. Yeah.

>> Your folks sent me over a note last night that Oracle named the leader for AI for the risk and finance industry. And so I'm interested in how banks are thinking about that. Our data shows a couple things. I'll share it with you, see what you see. We have a survey partner, ETR based in NYC. They show that 45% of customers tell us that, to fund GenAI they're stealing from other budgets. It's probably not as acute in the financial services industry, but broadly it is. And then the second thing is, their number one concerns are things like privacy, security, and legal, and compliance. It's almost like that used to be an afterthought and it was bolted on. Now you can't really do much without getting sign off from say, legal. What are you seeing in financial services?

>> Oh, financial services has that problem in orders magnitude, to be honest, especially the regulatory element that you referred to. And we are building to that. So I'll give you an example. So when you're in the debt collection business, so you're lending, you need to collect on your debts, so the FDCPA, which is an act that was created specifically for the collection of debt, absolutely outlines the type of language and sentiment that a debt collector can use when trying to collect debt. And there are very heavy fines for violating those policies. So how did this happen earlier? You had to monitor every call and make sure that a supervisor was watching a debt collector until they went up experience curve. And even then they had to spot check. Now with Generative AI, you can actually monitor every call and start to identify sentiment or tone, and see if it is in violation of the regulations. And you can provide feedback very quickly to the collection agent and say, "This is where you could have gone wrong," you can create a training program for them, and supervisors can look at multiple agents and see how their overall groups are functioning, what kind of incentives are required. So this is a classic situation where GenAI solves a real problem. I mean, the others are making sure that you get more personalized products and experiences when you're interacting with the bank. You could be doing this in financial crime, in souping up your investigation capabilities. So the number of applications are tremendous, but understanding the models, making sure you're following fair lending practices, fair collection practices, this is going to be a part and parcel of deploying these models. And regulators will continue to take a keen interest in all of this.

>> I went down to your little coffee area this morning, was talking to some of your customers. I mean, these are business people. They're not geeking out. I mean, I love to geek out, but I love the intersection of technology and business. Because that's what it's all about, how do you apply that technology to create a business outcome. But I'm super excited about the future and here's why. And I want to geek out a little bit and then get your reaction in terms of the business impact. So if you think about Oracle and its stack, other than semiconductors, pretty much has everything. And of course you have strong partnerships with many semiconductor companies. But the idea of, and I heard this at CloudWorld, the ability to take all the data that's locked inside of applications, the data, the metadata, the business logic, through Fusion, it's harmonized, you guys just did some hard work over the years to do that. But now I can serve that up to agents. We heard at CloudWorld, you promised to have 51 agents, many of which are in your space, which is super exciting. So a lot of people talk about co-pilots or maybe single agents. We're talking here about multiple agents working together, acting on that harmonized data to create a business outcome driven by top-down goals. You might have metrics that say, okay, we want to grow market share, so we're willing to take some price concessions, or we want to maximize profit so we don't want to cut price, et cetera. Whatever it is, top down, but then connect to that backend data to create a bottom-up outcome, agents working together in harmony. It's a long-term vision. It's not here today, but you've got the pieces to do that. How are you talking to customers about the future and applying AI? Are they thinking in those terms? Do they see the potential to completely reinvent their businesses around new workflows and drive new levels of productivity? Or is it more kind of day-to-day battles? What are you seeing in the field?

>> I don't see how they cannot be thinking about it to be very honest, I gave you just one example, which is, how do you manage fraud and the investigation of fraud if your transactions are growing 6X? I'll give you an example in terms of agents. We just release an adversarial agent to the market today as a matter of fact. And what the adversarial agent does is it acts like a bad actor and tries to break your money laundering protections, right? So it comes in behaving like a bad actor, and it tests all your scenarios, its settings, and it tries to get through them, and it finds all the weak points and gives you the opportunity to go address them. Now, this is what I mean by, how can you not be thinking about this, because trust me, the people who are trying to break into your system, they're not just criminals, they could be state actors. They could be people with unlimited funding and unlimited access to the best technology, the most transformative technology of today. So that's what you're up against. And you're protecting your customers, you're protecting citizens in the country that you're operating in. And so I can't see how you don't think about this. So from our standpoint, the best way we are doing this is to actually leverage our business applications, our core operating application, build in the AI use cases that I just described, and showcase them to customers, and make it easier for them to adopt. And the cost structure around this, if you don't have a single-minded focus on that, you can make the cost of inferencing so expensive that your customers will be disinclined to use it. So for us, the cost advantage that I mentioned earlier, is of paramount importance. We want to make sure that customers not only can leverage the data locked inside their enterprise, but all external data, and come out with outcomes at a price point that makes the ROI useful. And the reason we can do that well, is because of our next-generation cloud and our experience with all these business applications.

>> You got REDstack stocks at an all-time high, you've got all the pieces, of course financial services is a key harbinger of the future. Sonny Singh, thanks so much for coming on theCUBE.

>> My pleasure.

>> Great to have you.

>> Thank you so much for having me here.

>> You're very welcome. All right, keep it right there, for our next guest this is Dave Vellante for theCUBE at NYSE with NYSE Wired. We'll be right back right after this short break.