Hosted by top 5 banking and fintech influencer, Jim Marous, Banking Transformed highlights the challenges facing the banking industry. Featuring some of the top minds in business, this podcast explores how financial institutions can prepare for the future of banking.
How Generative AI is Changing the Face of Financial Services
Generative AI is pushing financial institutions to rethink how they work in exciting ways. Early adopters are already seeing the benefits. But, before financial institutions can reap the rewards, there are considerations and lessons to build on. Prime among them is a warning: the most dangerous thing about AI is assuming that it only delivers productivity.
In this episode of Banking Transformed, we are joined by Anu Sachdeva, global solutions & service line leader for banking and capital markets at Genpact, a global professional services firm focused on delivering outcomes that transform businesses.
Anu and I talk about the opportunities and change that generative AI can bring to financial institutions, the four key considerations for financial institutions and other enterprises to follow, the steps toward integrating gen AI most effectively, and the use cases that offer unexpected possibilities.
This episode of Banking Transformed Solutions is sponsored by GenPact
With 25 years of experience leading digital transformations at Fortune 500 companies, Genpact empowers financial institutions to develop, deploy, and scale gen AI solutions. Fast. At Genpact, we do more than rethink banking operations; we're democratizing AI so firms can anticipate customer needs, deliver seamless experiences, reduce financial crime, and grow revenue. Learn more about our AI solutions.
Hello, and welcome to Banking Transformed, the top podcast in retail banking. I'm your host, Jim Marous, Owner and CEO of the Digital Banking Report and co-publisher of The Financial Brand.
Jim Marous (00:20):
Generative AI is pushing financial institutions to rethink how they work in exciting ways, early adopters are already seeing the benefits.
Jim Marous (00:30):
But before financial institutions can reap these rewards, there are considerations and lessons to build on. Prime, among them is a warning. The most dangerous thing about AI is assuming that it only delivers productivity or only can be applied in a risk and fraud area.
Jim Marous (00:50):
Today, we are joined by Anu Sachdeva, Global Solutions and Service Line Leader for Banking and Capital Markets at Genpact, a global professional services firm focused on delivering outcomes that transform businesses. Anu regularly writes on digital disruption banking, designing customer experience-led operational models, and is a frequent speaker for industry associations.
Jim Marous (01:14):
Today, Anu and I talk about the opportunities and change that generative AI can bring to financial institutions, the four key considerations for financial institutions and other enterprises to follow, the steps toward integrating AI most effectively, and some of the use cases that offer unexpected possibilities.
Jim Marous (01:36):
It's amazing how many conversations there are around generative AI and ChatGPT. Even though the technology from a public standpoint was introduced just a little bit less than one year ago. The stories of generative AI success are beginning to catch the attention of financial institutions globally.
Jim Marous (01:55):
Morgan Stanley has used open AI, ChatGPT to organize its wealth management knowledge base, and both Goldman Sachs and Westpac have used it to help their developers write code. But you know what, we're still only scratching the surface of the opportunities before us.
Jim Marous (02:13):
So, Anu, before we get into a little bit about ChatGPT and generative AI, could you introduce yourself and give our listeners a quick look at your extensive background as well as describe what Genpact does for the financial services industry?
Anu Sachdeva (02:30):
Absolutely. And thank you very much, Jim, super excited to be here today. In my role as global service line and solutions leader, I'm responsible for our go-to market and execution of all our future-forward data process and tech solutions, and we generate all the big business outcomes for our clients.
Anu Sachdeva (02:51):
In my past role, I was the head of North America segment leadership, and in that role, I led several large relationships, created multimillion dollar contracts for Genpact, and doubled down the footprint across various financial institutions.
Anu Sachdeva (03:09):
I grew in commercial and retail banking, that was the early part of my career. And there, I was responsible in several roles across sales operations, Six Sigma, and that's where I really developed my deep domain.
Anu Sachdeva (03:23):
Now, a little bit about Genpact. Genpact is global, publicly listed professional services firm. And in Genpact, we deliver business outcome that actually transform industries and shape futures.
Anu Sachdeva (03:38):
Our global Fortune 500 clients, they come to us, they come to us for digital-led innovation. They come to us for digitally enabled operations that dig very deep into data and data from insights into action.
Anu Sachdeva (03:51):
Gen AI is all a part of that, as we are all talking about, and it is a deep domain and industry experience that we have built over the past 25 years that has led for us having that meta intelligence and algorithms which are so crucial for enabling generative AI. I'm very, very excited about it.
Anu Sachdeva (04:15):
Genpact has over a hundred thousand employees from New York to New Delhi. We actually are passionate about driving the purpose of the organization, which is relentless pursuit of a future that's great for everyone. And we are the number one business services provider for several companies.
Jim Marous (04:39):
Well, it's interesting, Anu, that AI is not something new to the financial services industry. In fact, financial institutions of all sizes have used AI for decades actually in preventing fraud and looking at risk options.
Jim Marous (04:54):
But when you look at generative AI and you look at ChatGPT, it really opened our eyes to what was possible. And as I said, this is only about one year ago that became more of a public domain type discussion. What do you find the most exciting about what has happened in the last year?
Anu Sachdeva (05:14):
Actually it's very interesting, Jim. Look, with generative AI, and I think all of us kind of got introduced to it with ChatGPT as you just mentioned earlier in the show. The power of generative AI came into our hands, we could actually, all of us - and this ignited conversations, this actually brought people together to look at different innovations.
Anu Sachdeva (05:35):
And what I'm super excited about is how it can actually lead to newer solutions, newer products. To be honest, I almost think about generative AI similar to how electricity came into our lives several years back. And if you think about it, what it did was it actually decoupled the energy from source to use. And what it led to was newer factory models, newer products, and newer usage. I think of generative AI in exactly the same way.
Anu Sachdeva (06:08):
This is actually a way for us to create newer designs, newer solutions for our customers, those personalized experiences, actually helping containing risk in operations in our supply chain. So, opportunities are tremendous, Jim.
Jim Marous (06:26):
So, I know that Genpact is working with a number of global banks and financial institutions of all sizes to solve business challenges for their customers and partners using AI capabilities. Could you share some of the use cases of how generative AI has been applied, and where have you seen the greatest interest?
Anu Sachdeva (06:47):
You are right, Jim, we have been working with several of our clients across various parts of the value chain and applying generative AI. And I can share with you a few examples where we have applied both in the non-financial services world as well as the financial services world.
Anu Sachdeva (07:04):
A few examples which come to my mind, a software development team with 1,800 plus custom applications, they use different programming languages. But imagine how the change request process would be - very, very time consuming. With generative AI we were able to automate the code delivery and accelerate the testing by almost 60%. That's a great example of using generative AI in actually software development.
Anu Sachdeva (07:33):
The other example which comes to my mind is also even at Genpact internally, we are leveraging generative AI in our own payables process, where we have been able to make it far more efficient almost by 35 to 40%, and responding to the payment queries faster, thereby, enabling the process to be far more smarter.
Anu Sachdeva (07:54):
So, there are a number of examples which come to my mind, and coming back to our financial services world, which is where I'm seeing several applications of generative AI and where we are testing, some of them are very futuristic, but we are the innovation hub for our clients, and we are testing generative AI across the value chain within banking, and couple of ideas and areas.
Anu Sachdeva (08:17):
I spoke about that in one of the Finovate sessions as well is in underwriting, in commercial banking, where we are actually testing to reduce the cycle time it takes to underwrite. Where we are crunching reams and reams of documents and how you can actually create so much of intelligence from there, and help reduce the cycle time of bringing customers on board.
Anu Sachdeva (08:40):
So, those are a few examples which come to my mind, and there are several more which we are testing right now.
Jim Marous (08:47):
So, you very much like me, do a lot of speaking. You get in front of a lot of groups and I know there's a lot of buzz around generative AI.
Jim Marous (08:54):
And you delivered recently a keynote at Finovate fall on the topic as well as wrote an article, or was interviewed by FinTech Futures to get your perspective on generative AI. So, as an expert, or at least right now, as much as an expert as any of us can be after only one year under our belt, what in your view should banks consider as they start their generative AI journey?
Anu Sachdeva (09:19):
And this has come up several times and that's actually a real challenge, I would say, as banks get onto the journey, because there is so much of information out there and there are so many ideas, how do you prioritize? I think that's the first thing which the banks are actually struggling with. How do you prioritize what use case to take? And more importantly, how do you make sure that that's effective?
Anu Sachdeva (09:44):
So, the first area, or the first, I would say learning that I want to share with our customers is be very specific about what outcome that you want to drive from generative AI. It's extremely important to use generative AI in end-to-end solution, end-to-end outcome versus making it a point solution. And that, I think this is the biggest mistake sometimes our clients are thinking very narrow, that's one.
Anu Sachdeva (10:15):
The second area which is very critical is integrate generative AI as a part of your overall technology stack, that's another big learning. Combine it with cloud, with robotic process automation. Close to our banking world, if you think about it cover and monitoring, create your models using generative AI, use your workflows to monitor them. So, there are several areas where it can actually integrate across the technology stack.
Anu Sachdeva (10:41):
I think data becomes a very, very critical component, and data orchestration, how do you ensure that the data is available, it's clean, it's labeled well? I think data is going to be the most important aspect as banks look at getting onto that journey.
Anu Sachdeva (11:00):
And then process is equally important, Jim. And sometimes, a lot of times, we overlook the importance of having a standard process, very well-defined with what KPIs you're going to measure.
Anu Sachdeva (11:13):
And one interesting example, which comes to our mind, I was talking to Chief Data Officer, was they were unable to actually work on a use case because they were very, very split on what KPIs should they target for. And I think that's an important aspect to keep in mind.
Anu Sachdeva (11:30):
And lastly, but not the least is the change management which is needed. How do you bring your employees? Because look, the biggest question on everyone's mind is, "How will generative AI impact me as an employee in the organization?" So, bringing employees with you onto the journey to make sure that they feel a part of it, and also helping them upskill and get trained onto generative AI is going to be critical.
Jim Marous (12:01):
You unpacked so much there, and we're going to get into each one of those topics. One thing that's interesting, and you mentioned it, that when we talk about generative AI and we talk about ChatGPT, we tend in our mind to think it's one thing, but the reality is there's so many different platforms.
Jim Marous (12:17):
I realize that in my own life, and I don't use it nearly as much as most financial institutions do. The reality is there's so many platforms that are specialized in certain areas of generative AI and to find the partner organization or a partner platform that's going to work.
Jim Marous (12:31):
But more importantly, you talked about the importance of data. And so many institutions use the fact that their data is not market-ready, let's call it that. Market ready out there because they just think there's too many silos, it's not integrated well, it may not be clean.
Jim Marous (12:51):
I think what's important is for organization to realize this is why you use partnerships; partnerships like your organization, Genpact, that can help make that data usable, make it so it can be deployed against a number of solutions.
Jim Marous (13:05):
But generative AI being most important, certainly in our discussion today, that's one of the major challenges. But with any innovation, comes challenges, and especially with something as transformative as generative AI, what challenges besides the data element, have you seen the marketplace with organizations that are early movers?
Jim Marous (13:27):
What things have been ... Has it been leadership? Has it been the overall culture of the organization? Has it been that organizations may not know their why, they just want to do something without really knowing what their destination's going to be? If you put it in a general context, what have you seen as being the most important challenge that organizations have to realize they're going to have to face?
Anu Sachdeva (13:50):
Jim, this is such an important question and I think you are absolutely right. As I meet up with so many customers, and I'm sure you have heard in your conversations as well. For early adopters, there's a lot out there. So, how do you make sure that you choose and you are selecting, first of all - and what are the characteristic of those partners? I think there's a little bit of a lack of clarity.
Anu Sachdeva (14:15):
And the way I see it, there are four different kinds of partners which are there in the market. They are the democratizer, which are the big tech, where the large foundational models are, whether it's the Microsoft of the world, or AWS, Google, so many of them.
Anu Sachdeva (14:34):
And then you have those solution creators, and these are the solution creators who actually build those business outcomes on top of these foundation models.
Anu Sachdeva (14:45):
The third category of partners are those technology architects, and these are those architects, experts with domain who will enable fine tune these models using prompt engineering or several other different kind of ways, but bringing in that contextualization of that domain where this application will be in their specific industries or in their specific organizations.
Anu Sachdeva (15:11):
And then lastly, to your earlier point, are the data specialists, and these are the partners who will enrich data, who will make sure that data governance is in place, who will make sure that data is correctly labeled.
Anu Sachdeva (15:25):
So, there are so many different kind of partners out there. And I think as Genpact, we obviously play between being the solution creator or being the technology architect and data specialist, and we work with all the democratizers.
Anu Sachdeva (15:38):
But I think for early adopters really, some of the challenges are, first of all, how do I select, how do I bring these different components of partners together? How do I ensure that my data is ready?
Anu Sachdeva (15:53):
And I think data, as you and I were talking about earlier, becomes a very critical thread to enable some of these models. Some of them are having challenges with having trained data specialist who will enable generative AI applications.
Anu Sachdeva (16:11):
A few others that I've met, they are having challenges in target operating model internally. And one story I have here is it almost to them, some of them seems like a cottage industry because every function wants to create their own models, their own solutions, their own proof of concept.
Anu Sachdeva (16:31):
So, what is that right target operating model within the organization? Who should own the models? How should you run it? How do you create that machinery within the organization and bring it together.
Anu Sachdeva (16:45):
And lastly, I think where I have seen some organizations struggle is really in getting senior leadership buy-in because they're not able to justify the cost. And that's really where I was talking about this whole linkage to outcome versus thinking of generative AI as a point solution. They are very narrowly defining the role it can play. So, those are a few things which come to my mind.
Jim Marous (17:11):
So, almost everybody's at the starting gate right now, and it's not an easy track because there's so many tracks you can go. You can use it as a content creator, you can use it as a way to get better resource management. You can use it for back-office solutions.
Jim Marous (17:26):
So, if we look at the next 12 to 18 months, where do you see some of the strongest implementations happening? And what are you seeing right now as we look at AI solutions and generative AI solutions in consumer and commercial banking specifically? What have you seen as being a good startup use case?
Anu Sachdeva (17:49):
So, Jim, I have seen across the spectrum, I would be very transparent that most of the applications that I have seen so far are more internal facing. What I mean here is they are more targeted towards improving employee experience, whether it's employee experience in the contact centers, which is largely where I've seen banking and financial institutes starting their journey.
Anu Sachdeva (18:17):
Because this is where a lot of pin points are there, a lot of unstructured data is sitting in, and they want to increase and improve the experience which the end customers have. And that can only be done by really improving the employee experience and also enabling them with information.
Anu Sachdeva (18:35):
So, ensuring that next best action is created and support the customer service agents with the right information so that they could deliver a great end outcome. So, that's one big use case, which I have seen predominantly where most organizations are at least starting off from.
Anu Sachdeva (18:54):
The second area, again, little bit internal facing, is more in their own contracting, in their own legal departments, where again, there is reams and reams of data and paper sitting in. How do you extract information, how do you simplify, how do you make sure that you are able to get more information faster.
Anu Sachdeva (19:14):
And then there are some areas where some of the large banks are already testing, trying it, and I think Morgan Stanley announced in their financial advisory where they have already tested and tried and implemented generative AI.
Anu Sachdeva (19:30):
But a few others are testing it in functions like marketing, where they are looking to expand the market size again, by getting as much information from the data, from the customer 360 interaction data, so that they could make much smarter choices for their customers.
Anu Sachdeva (19:52):
We are actually helping, as I said, in a few use cases that we are trying for our customers. Delinquency management is another area where I think there will be a lot more, which will come in. So far, it's largely been in customer experience, but delinquency management, as the market softens, I think there will be more activity we will see there.
Anu Sachdeva (20:14):
We are already trying and testing there using greater insights from data to control delinquency. So, those are a few areas where I'm already seeing some action.
Jim Marous (20:25):
So, it's interesting, as I mentioned earlier, one of the early applications of AI - not generative AI, but AI in general was in risk and fraud management. Because it was using data to find patterns, things that stood out as being good ways of identifying risk.
Jim Marous (20:41):
So, Genpact has recently announced the integration of Amazon Bedrock into its riskCanvas financial crime suite. How is Genpact using advanced generative AI capabilities to really change the way we look at financial crime management?
Anu Sachdeva (20:59):
Jim, this actually is probably the most important area in my view as we look at the future applications of generative AI. How do we make the world more safer? How do we make financial institutes more safer? And I think that's really the application of how to cut down on fraud. How do we stop this money laundering, which is happening there?
Anu Sachdeva (21:22):
And let me back it up a little and share some statistics around it. There is roughly about $1.6 trillion of illicit funds which transact across the globe, and unfortunately, only 1.6% of that ever get caught or get remediated. So, this is a very massive problem and a big risk for us, as financial services partners, banks, institutes, and which has to be solved.
Anu Sachdeva (21:55):
And look, banks have been at it to solve it. They have been trying to remediate it, but they have also ended up paying a lot of fines for lack of controls in these areas as high as 8 or $9 billion within a year, last year or two years.
Anu Sachdeva (22:11):
So, how we are looking to solve it, and again, I'm super proud that as Genpact so close to our purpose of really making this world safer, we have been able to leverage generative AI in what is called transaction monitoring, which is a very important part of anti-money laundering.
Anu Sachdeva (22:33):
So, as any transaction happens, if you and I make a transaction on a credit card or in our wallet, any of these transaction would go through what is called a transaction monitoring process, where some keywords are matched. So, that matching takes a lot of time because there are several permutation and combinations which have to happen. That's one part of this being a very tedious process.
Anu Sachdeva (22:56):
The second part of this being very, very tedious is that in the interest of keeping it, they save banks and institutes, they create a lot more transactions, which they pull out as what is called false positive. So, that increases the load on the analysts who are working on it.
Anu Sachdeva (23:16):
And guess what, in the meantime, while this activity of matching and cross-checking and narrative writing is happening, these rosters are probably going and making transactions in several different institutes. So, the way we actually helped in solving this, we partnered with AWS.
Anu Sachdeva (23:37):
So, we have our own proprietary platform called riskCanvas. Using AWS bedrock generative AI large language models, we have been able to cut down the time it takes to write these narratives. We have been able to actually do this work in minutes where it was taking several days to do it.
Anu Sachdeva (23:57):
And the biggest thing also is this actually helps in explainability to regulators, which has also been a very big pin point. So, very happy and very, very proud to share this with our audience, and this is now a solution which is already running.
Jim Marous (24:14):
So, I'm going to flip things around a little bit. And let's say you are working for a financial institution, and you have an idea of what generative AI is, what its capabilities are, what ChatGPT is, but really, your organization is just starting off, and at least in my experience - I was at Money20/20 in Amsterdam a few months ago. I was in at Sibos in Toronto last month. And there isn't one organization solution provider that's not talking about how they're using generative AI.
Jim Marous (24:46):
If you were a banker, how would you evaluate the providers out there to find out how many are really using generative AI and ChatGPT and other solutions, and how many are just talking about it and not really ready to put it to use yet? How do you sift through all this when everybody's in the learning process?
Anu Sachdeva (25:09):
No, I think great point. And I would probably start with domain. Do you have the right domain in the area that you are actually trying to solve that problem? And I think that will probably differentiate one partner from the other. Look, we all will have access to the foundation models. We will all have access to what sits on top of these foundation models.
Anu Sachdeva (25:35):
A lot of it is now going to be an open source, but how do you use it? How do you personalize it? How do you fine tune it? And for fine tuning also, Jim, the big part of fine tuning is the contextualization, and where will the context come from if you don't have that domain? So, I come back to domain as being one of the biggest criteria which companies should look for.
Anu Sachdeva (25:58):
The second would be the ability to actually look at data. I think that's the other big aspect as partnershow are we creating data discoverers in the organizations? Do they have the right team who's supporting? And do they have the right mix of engineers who will be able to bring in their expertise along with the industry experts who know what they're looking for? So, it's a combination, I would think, which will have to come in play.
Jim Marous (26:31):
So, it's interesting. You talked about it earlier that the need to find talent. We know there was already a need to find talent when you talk about digital banking transformation. But now with generative AI, there's even a greater need to find talent that's going to understand and be able to translate, talk to actions.
Jim Marous (26:50):
You're also going to have to train existing employees on how to best use generative AI to build better solutions, both from a revenue standpoint, a customer service standpoint, and a back-office innovation and transformation standpoint. But how do financial institutions deliver that training, so they have a direct impact on the return on their investment? How can generative AI help this happen?
Anu Sachdeva (27:18):
So, actually, Jim, we went through this journey ourselves, and I can share with our audience how we went about doing this. Because look, in our organization, which is 110,000 plus, large organization, and our biggest asset, our jewel are our employees, our people. So, we went on the journey almost five to six years back when we stepped into the AI world.
Anu Sachdeva (27:46):
So, before even generative AI, we came in as very strong process experts, but we realized very quickly that, look, our process experts now need to be data experts. They have to be data discoverers who should be able to understand the deep data sitting in so that we could create those insights for our customers and take it to action. So, we brought in what we call as a data bridge program, which essentially trained our process experts to become data experts.
Anu Sachdeva (28:19):
So, we currently roughly have about 80,000 of our 110,000 employees who are data-trained. That helped us step into our next phase, what we call as AI journey. So, now, we took our data experts and created a program, a very curated training program, and we call it Genome, which actually has a learning path for each of the employee.
Anu Sachdeva (28:46):
If Anu Sachdeva wants to tomorrow become a data expert in risk and consulting, I will have a curated path for that learning. So, we actually are right now in the process of training our teams who are now AI experts. And this is an open platform, which is open for our employees for cross training, upskilling, because we think it's extremely important to have your organization ready so that we could create that momentum and knowledge management for our teams.
Anu Sachdeva (29:18):
And this is something which we offer to our customers as well. And it was something which was very, very critical for us, even during pandemic because we were able to train our mass employees onto data and several other processes.
Anu Sachdeva (29:35):
So, that's how as Genpact, we have been able to handle this. And I think this is going to be knowledge management training, cross training would become very critical, and that is closely linked to employee retention and employee satisfaction as well.
Jim Marous (29:51):
So, it's interesting. There's not a discussion that goes on where people aren't concerned about whether AI, new technology, or even more importantly, generative AI may actually replace the need for humans, or maybe there'll be a mix of humans and technology.
Jim Marous (30:09):
If I was to look it up in a generative AI query around, will generative AI replace humans? I could get various answers that go both sides of the equation. What is your perspective on how AI may or may not replace humans?
Anu Sachdeva (30:29):
So, Jim, I always think that as humans, we have a tendency to overreact, and this is not new. This has been happening over hundreds of years if you dial back into history. At the time of Socrates with reading, he thought it's going to atrophy memory.
Anu Sachdeva (30:46):
Or when newspapers came in, it was assumed that, hey, we will stop interacting with people and we will not be able to exchange ideas, or not even gossip around. When television came into the lives, I even thought that people are going to stop doing intelligent activities, well, look, none of that happened.
Anu Sachdeva (31:05):
I mean, so in the same light I think that generative AI actually is going to help us. It's going to actually help the task get better. So, in my view, it's going to be human in the loop. It's not replacing humans, and most predominantly, in the banking and financial services world, for sure, where there is so much of regulation, we think that human in the loop is going to help in making the decisions more empathetic in bringing in ethics, in bringing in that emotional question, in bringing in that decision-making, which will actually make these technologies far more creative and innovative so that we could actually make life simpler and give humans more superpower to perform their work.
Jim Marous (31:58):
I see it also as a supplementary or complimentary role. Businesses change, roles change. I see it enhancing the humankind situation, but it's going to take work on humans to be ready for that. They can't stay still hoping that it will stay out of the way. You're going to have to learn along the way. So, what makes you most excited as you look at the future generative AI in banking?
Anu Sachdeva (32:30):
I feel things are going to actually improve in experiences, what we can deliver for our customers. They'll be far more personalized experience. The interaction that we have had with our banks is going to drastically change. It'll be far more intelligent, personalized, there will be newer products, which you and I can't even think about right now.
Anu Sachdeva (32:55):
The way I feel and see the world is going to move away from just being reactive into being more predictive and being more proactive. And that's how the solution set are going to be, and that's at least as an organization, as Genpact, we think about it.
Anu Sachdeva (33:10):
It's go be all about newer designs, newer set of ideas which will come to the market. Yes, there will be regulations, yes, there will be those areas which we all have to think about as we roll out some of the products, but the future is going to be very exciting for us in banking.
Jim Marous (33:30):
It's interesting, I talked about the horse racing analogy saying that we're almost all at the starting gate. Some maybe have a little bit of a lead in the process, but for those that are at the starting gate, trying to look around, see what the playing field's going to look like, figuring out what to do next, how do you think financial institutions should start their journey? What must they do today to become future ready for tomorrow?
Anu Sachdeva (33:56):
I think, first and foremost, employees. First and foremost, make sure that organization readiness is there. And for that readiness, as an organization, you investing in training, cross-training, upskilling employees. That, to me is the number one criteria.
Anu Sachdeva (34:14):
Second area, which I would say is, think about how your data is set up today. Just maybe try and look at how you can create some proof of concepts, pick up areas which can actually help customers. You can look at interaction data for customers, how do you personalize the experiences?
Anu Sachdeva (34:36):
So, I think there are opportunities where it can be tested and tried rather than having to jump into completely a client-facing generative AI application. There are several areas which can be tested right away. There are partners like us who already have some of these ready and some of these are already being tested. Talk to those partners, learn. I think this is an opportunity for all of us to collectively learn out there to understand what more can be done.
Anu Sachdeva (35:04):
So, this is actually a great way for banks and financial institutions to also come together and see how do we make this more transparent for the end customers as well, because those questions will also come in.
Anu Sachdeva (35:20):
And ask us, ask the partners like us, how are we making sure that these models are transparent, these models have been tested. So, I would probably put in both the sides that in your due diligence, ask some of these things to your partners.
Jim Marous (35:38):
So, Anu, finally, on a personal note; when you look at generative AI, while there was inklings of what could happen - until ChatGPT really was introduced in I think it was November 30th of last year, it was all about what we could dream of as opposed to what we can apply.
Jim Marous (35:55):
We've all had to learn very quickly what's possible. And every time I open a new tweet or LinkedIn article, or some research paper, it makes me know how little I know about what's going on.
Jim Marous (36:09):
So, on a personal note, Anu, how do you stay on top of what's going on in the field of generative AI at a time when you have a lot of other responsibilities as well that don't have anything to do with generative AI? How do you balance your day and how do you continually learn what you have to know next?
Anu Sachdeva (36:28):
So, Jim, I actually immerse myself into anything which attaches itself to generative AI, all the way from music to medical, to my current world of banking and financial services. My daughter sings and we are right now recording for her. And I was fascinated when I read about this K-pop group, I don't know whether you knew that. There's a K-pop group, which is completely gen AI-driven, or AI-driven.
Anu Sachdeva (36:57):
So, the opportunities which are out there are so enormous that every day I get up to learn. Of course, in Genpact we have several ... I was telling you about our Genome learning platform. So, I encourage myself to learn and my teams to learn a new, but I think it's all up to us.
Anu Sachdeva (37:18):
I think it's the ingredient of curiosity, which I try to put in every day as I come into work, that what new am I learning today? What is it new application that I'll walk back today? And I try to then learn and share, because that's another way to ensure that you're learning correctly-
Jim Marous (37:35):
Get the dialogue going.
Anu Sachdeva (37:37):
And ask questions. So, there are a lot of terminologies I'm not familiar with, I reach out to my team members and learn from them. So, it's a great way to actually exchange learning, learn something new every day and stay very, very curious because we will know a new thing. I'm sure every second a new thing is getting introduced. So, we'll always be able to learn something new every day.
Jim Marous (38:06):
That is such a great suggestion. And it's something that I take for granted many times. I think you probably take for granted the issue of curiosity, wanting to learn more, wanting to continually evolve because evolution and change is something that is not in our blood.
Jim Marous (38:25):
It's something that we resist overall, because who wants to be told that, "Oh, by the way, what you learned yesterday is no longer relevant, and you've got to keep on changing?" I mean, I'm no youngster, but I realize that I have so much fun having conversations with people that are using it in different ways.
Jim Marous (38:43):
And a real quick story, I was at a restaurant this last week the waitress said that she just got out of school, she's trying to build - a journalistic background she has, but she's trying to write some pieces of fiction. And so, we got done with dinner and I pulled her aside, I say, "By the way, how are you doing?"
Jim Marous (39:01):
She goes, "Really well." I said, "Okay, so in your pursuit of writing some fiction tales, some books, how much are you using generative AI in your process?" She goes, "It is saving my life, because you no longer have to have a writer's block because you can always ask someone or something smarter than you about what you're trying to write about and get ideas." She says, "I don't write from that, but it keeps me going. It keeps my mind moving."
Jim Marous (39:30):
And it's such a great opportunity, but there are so many that find this difficult, especially in banking, where when I started in banking, one of the reasons why I like banking, people recommended it, is that, well, things move kind of slowly, it's pretty stable business, there's not a lot of change. That was something that was an asset back then, it's no longer an asset.
Jim Marous (39:50):
So, Anu, thank you so much for being on this show. I really appreciate your enthusiasm, your knowledge in the area of generative AI, but also, your recommendations as to how people can move forward, because it's so easy not to move forward, but it's really the death nail because things aren't going to stop simply because we want them to.
Jim Marous (40:10):
And part of that learning process is also understanding the risks that are involved. We didn't talk about that too much, but the reality is, you have to be aware to know what you don't want to know, or what you don't want to pursue as much as what you do pursue. So, again, thank you and your team from Genpact for doing so much for the industry in the area of generative AI.
Anu Sachdeva (40:31):
Thank you very much, Jim.
Jim Marous (40:33):
Thanks for listening to Banking Transformed, the winner of three international awards for podcast excellence. We appreciate the support we've received to make this endeavor a success. If you enjoy what we're doing, please take some time and show some love in the form of a review.
Jim Marous (40:47):
Finally, be sure to catch my recent articles on The Financial Brand and check out the research we're doing at the Digital Banking Report.
Jim Marous (40:55):
This has been a production of Evergreen Podcasts. A special thank you to our senior producer, Leah Haslage; audio engineer, Chris Fafalios, and video producer, Will Pritts. I'm your host, Jim Marous.
Jim Marous (41:07):
Remember, it's time to embrace change to make you future ready for the opportunities ahead of us.