How Generative AI is Transforming the FinTech Industry

Speaker A: Hello everyone. Welcome and thank you for tuning into Credix podcast series Fintech X. I am Avandeep Singh, engineering manager at Credix, which is India's largest supply chain finance platform and I'm happy to be your moderator for today's discussion. So today's topic of discussion is how generative AI is transforming the fintech industry. We will all agree that in today's rapidly evolving digital landscape the intersection of artificial intelligence and finance is actually becoming very critical. As uh financial institutions strive for greater automation, efficiency and personalization, gen emerges as the transformative force. This tech not only streamlines processes but also enables the creation of financial solutions tailored to uh individual needs. Also advanced predictive capabilities and robust security measures of Gen makes it very essential for navigating the complexity of modern finance. Thus understanding how generative AI is reshaping the financial industry or fintech industry if I may say is not just relevant but imperative for staying ahead in today's dynamic financial ecosystem. Well that said uh, allow me to introduce today's distinguished, distinguished panel of speakers who bring a wealth of experience and expertise to share uh, uh their knowledge on this topic. First allow me to introduce Mr. Ashish Khandelwal. He is the founder and CEO of ANQ Finance which is a digital banking platform that harnesses the power of decentralized finance to deliver next generation financial uh services. We also have with us Mr. Pramod Khaturia who is founder and CEO of Easyloan which is India's first digital home loan marketplace. A very warm welcome to two of you to Fintech X powered by Credix. Uh now without further delay let's delve into our discussion. I would like to uh first invite Mr. Khandelwal to share his thoughts around how does generative AI differ from other forms of artificial intelligence and what makes it particularly suited for transforming uh the fintech industry.

Speaker B: Thanks Amandeep. Thanks for hosting me. Uh today definitely we are uh in an era of uh artificial intelligence, right? And uh generative uh AI has taken everyone uh you know uh ABAC with the capability it you know uh offers on the table right now coming to uh artificial intelligence per se and generative you know AI in specific if you look at uh, especially in the financial services uh usage of artificial intelligence have been there for uh, uh you know almost at least for a decade right? And every financial uh services institution have been trying to build uh and bring artificial intelligence capabilities to streamline processes, build efficiencies uh reduce uh human uh error uh and lot many other pieces. Having said that if you look at uh. All those artificial intelligence were primarily based on either supervised or unsupervised learning anyway but end outcome of them were fairly restrictive. Just to give you an example uh uh. If uh you look at from an unsupervised learning standpoint I think most of the pieces were getting uh uh stuck to a level of uh doing a segmentation uh doing some uh specific data uh set basis pattern and converting them into an outcome which was fairly predetermined and predefined. Versus if uh you look at a generative AI I think it has just changed uh the way you uh look at the uh uses of AI and uh. Not only it's bringing the unsupervised uh learning to a different level altogether but at the same time adding a layer of context later. Uh again uh just to give you an example, right. And uh, you know giving example of yesterday, right when we were doing some kind of uh year end closer activity and some bit of reconciliation from different, different invoices. Not that those invoices you know could not be uh you know done uh uh with the earlier version of AI uh or the OCI or OCR or other technologies which are there in place but every time you have to specify with a preset uh uh parameters. Right? But yesterday when I was doing it uh. I had a bunch of uh invoices of different uh uh nature from different uh vendor for different services and so on so forth. I put that uh into generative A and said I need this information. Give me this summary. So effectively what it did uh. It just uh uh helped me streamline the entire uh process without training any kind of model without uh you know doing a predefined uh level uh of uh you know what is the end outcome which I'm looking at. So this is one you uh know aspect of it. The second aspect if you look at even from the supervised learning standpoint, right. Uh it was a very very uh predictive you know uh task. Right. Which you could do for uh example credit scoring for detection. Again uh to a limited set of you know data set again you have to uh you know do lot many massaging and uh play with a very limited number of variability. Right. Uh with the generative uh you know AI uh even for the supervised you know uh uh pieces. Right. It has you know started creating a content, right. It has started getting labeling of the data which essentially mean it gives you far more understanding of the data and far nuanced understanding of the data. So uh. Put put together if you look at you know from a uh analytical output standpoint, right uh uh traditional AIs ah have been there uh and will continue to be there but uh generative AI has put the uh pattern recognition or the discovery uh aspect to a different level. And similarly uh from an output standpoint it has given far more context uh associated with any of the uh end outcome. Now if you look at uh in the context of the fintech, uh see uh financial services are all about you know managing the risk and managing cost effectively right While using the technology. Now the moment you are able to uh bring in the efficiencies, right and uh, the moment you are able to remove uh uh or reduce the error to the final level and bring in the explainability, uh you know what it does is uh, it gives you better control in terms of risk management, in terms of process, uh you know, management in terms of uh you know uh, um better answering or faster you know responding to the uh changing regulatory uh you know environment, customer experience, customer service. I think uh you know uh, if you look at the implications are far wider and uh, uh because of you know the entire uh evolution which we are seeing or witnessing in the generative uh you know, AI space it is bound to create a hyper personalized new experiences. Something which everyone in the financial services have been dreaming for you know ages and have been working actively for ages. Again the nature of the industry, right uh given that financial services have to be uh you know managing the cost and risk effectively and responding to the regulatory and the customer needs with the generative AI right Coming into the picture, giving you added layer of context, ability to deal with a large you know unstructured structured you know data form and uh making you or helping uh you know us understand you know customer much better. It is bound to um, you know impact across the life cycle of the financial services product.

Speaker A: Yeah, I absolutely agree with you. Uh, so Pramod, uh, what are your, your views uh on this?

Speaker C: So hi, hi everyone. And Amandeep, it's a very very interesting topic that you kind of chose for today's discussion. Um, I think uh. Gone are the times when there was this saying that computers uh are uh sort of good at following instruction but not reading your mind. And since that time to now and as we sort of stumble into the future things have completely changed. Uh and so much so that uh a uh computer is able to think like human, uh feel like human and behave like human in so many ways. And I, I think and that's where the generative AI Part starts sort of kicking in so quickly. Uh a very very quick reflection on the evolution of whole of, of uh the entire AI piece. I think uh, uh about 1980s is when we started talking about uh data driven approach, um or machine learning. How this chunk of data can start uh, making sense for processing of information. Right. And then come 2000 and there was this big discussion about big uh data, the availability of data from various uh recordings, various sources uh and how is it that data can interact with each other and give us more insights analysis. And I think all this was progressing and this is the nature of tech and I don't think there's any permanent pit stop here. It's only today uh that the uh time and uh space allows us to read so much into information and I'm sure in next five years the uh artificial intelligence will start uh, surprising us even more in so many positive ways. Right. Um, I think uh, when we talk about generative AI or essentially artificial intelligence, uh this so much more to achieve we just kind of still scratching the surface, uh our ability to right now look into data sets um and look into uh various predictive models, uh our ability to forecast, create uh, you know and bring about non existing data into realm is something which are generative AI which is essentially a subset of AI starts producing uh and unlike uh traditional AI which primarily uh analyzes and predicts based on existing data, generative AI learns patterns and structures within the data to generate new uh outputs. Very interesting topic. Look forward to do that. We ourselves have been scratching surface reading into our customer database for the last uh three three and a half years since the time we started and we decided to be one of the early flag bearer uh on the secured lending space wherein we are trying to bring a matchmaking based on artificial intelligence. So look forward to this discussion uh very interesting topic like I said.

Speaker B: Sure.

Speaker A: Okay, so uh, I was just curious, any specific application of Gen AI that uh, uh you're currently uh if I mean your picks, uh, if you, if I can say that you feel are revolutionizing the fintech landscape. So any, any particular application that you you know have felt. Okay, this is something that you uh, know is turning tables or uh, this is something that can you know grow into something big.

Speaker C: Yeah. So I think in the fintech space first of all anything and any aspect of business, any aspect of output which requires data and which can uh strengthen the quality of output can be strengthened with the quality of analysis of data uh will be of greater importance and which is where I think first up is uh any level of credit uh uh understanding of a customer. Uh you know because there is so much of information available. And uh. It's extremely intriguing how artificial intelligence and gen AI can uh sort of bring all these pieces together and make a very robust uh underwriting or a credit uh analysis of a customer. Number two is I find a high ah level of intervention on the investment side. Both, both whether it is a stock, mutual funds or otherwise. How uh the level of detailing. Uh you can go with the existing algos on past data. Your uh ability to look at behavior of uh you know financial instrument and have uh generative AI play and develop models around that. And third thing is your personalization on the financial uh advisory or the products. Uh every customer is different. Uh one tries to bring certain uh bouquet of services. However the personalization can only happen once. Your understanding of past data and those many Personas of customer you can build together and really make uh some predictive uh suggestions on what would be the desired output for a particular customer. So I think the personalization uh in financial advisory is a great uh opportunity. Uh so these are three things which kind of appeal to me a lot more. Uh and obviously there is a very wide uh use case for gen throughout fintech.

Speaker A: Yep.

Speaker B: Okay.

Speaker A: I completely agree with this actually.

Speaker C: So

Speaker A: uh, just addressing one pain point in uh uh you know our financial industry and that is the regulatory compliance. So how do you think uh genai can play a role here uh in mitigating risks or ensuring that you know the financial products are adhere compliances. So ashish, if you can take this up.

Speaker B: Sure Mandeep. And interesting uh because as I said uh earlier as well, right. Regulatory, right landscape which is ever changing. Uh you know I see. And uh. We started using uh you know somewhat uh you know in a very uh cautious manner. Right. Uh, ah you know the entire generative uh you know a output uh is from the perspective if you look at. Right. Uh. Uh. See regulatory documents are generally uh you know quite big.

Speaker A: Right.

Speaker B: And uh, you know uh. It requires lots of interpretation. Right. Which essentially mean uh somebody has to go through the. The last set of document, uh understand the meaning as is and the uh intent of the meaning. Right. Which is there. And also understand you know what are the implications it might have uh in your existing uh scheme of things right now adding all of these, you know three things together. The fourth aspect which becomes very, very important from a regulator uh you know uh scenario especially for the fintech, right. Where uh you know uh. I think overtime expectation has become more like uh you know, regulations ah will come out today and possibly, you know, in couple of days. You have to just uh, abide by the same. Right? Unlike uh, uh, you know, earlier scenarios where regulations generally used to come. Uh there used to be sufficient time, you know, for somebody to act, uh, you know, make these changes in the system and so on so forth. Now uh, combining all of these together, right? Uh, uh, number one, uh, given the fact that generative AI are good in terms of uh. Processing a large amount of, you know, data in one go, it helps you, you know, summarize uh, the entire regulatory requirement in one go. That's number one, right? Number two, it helps you understand the intent and the real meaning. Right? Uh, and uh, to a large extent, you know, it's more like, you know, you're talking to your legal compliance. You can talk to your you know, generative AI, you know, uh, interface, right? And understand what exactly you know, could uh. It mean, uh, in terms of uh, implications, uh, you know, and uh, what are the changes you uh, know which you need to bring in, if at all you have to bring in, uh, to meet you uh, know those needs. So that's the second aspect. The third aspect is uh. See, whenever we create any kind of uh, you know, uh. Flows, right. Uh, be it a process flow or anything else sometimes and uh, uh. Uh, you know what happens, uh, you uh, just miss out, right? What, what should be the right uh, you know, uh, piece or process which essentially should ensure that you know, regulatory wise you are compliant. Now all of these places, uh, generative AI, uh is helping a lot because it can help you identify, understand and uh, mitigate those pieces. This number one aspect, number two aspect, the second uh, important piece is uh, about uh, uh, monitoring and uh, figuring out what could be the uh, uh issue. Because see, when we build systems, uh systems are bound to have some kind of uh, uh. You know, bugs or errors in between, right? That could mean anything, right? Uh, despite you know, best effort. Now when uh, you look at the large amount data because uh, sometimes you have the small, small you know, events which you generally might uh, overlook. But if you you know, use generative AI and that's where we use for example uh, you know, generative AI from backend standpoint, uh, to analyze large amount of logs and the uh, you know, data to understand what exactly is happening and if at all there could be any kind of uh, you know, incident which needs uh, uh, immediate action, right? So uh. It also help you in terms of you know, complying with those regulatory, uh. You Know requirements. So this is second aspect, the third aspect, uh, uh, and very important aspect is the regulatory reporting, right? Because uh, sometime uh, you know what happens. Your data is not only about the data which is there in your system, right? It includes the data which is you know, there outside your system. It could be uh, some of the management, uh analysis or auditor summaries or even the partner uh ecosystem related information. Now all of these are coming together generally uh for anybody, any human being, ah to do that would be time consuming. But generative AI, what it does, it helps you in terms of putting it into the right form format, uh uh, and in a logical uh uh flow, um, much faster and much efficiently. So yeah, so it helps you in terms of you know, understanding the regulatory uh requirement, uh figuring out what needs to be addressed, how it needs to be addressed, figuring uh, out whether you know, current processes which you have uh, has any kind of uh, you know, uh, opportunity to improve this further and at the same time you know, reporting it uh uh uh, from the perspective, making sure that uh, you are you know, complying with those regulations. So from a rectech standpoint if you look at I think uh, uh generative AI plays a very, very significant, very smart role. Helps you uh, you know, mitigate the most important risk in this industry which is regulatory risk.

Speaker A: What are. So since you mentioned uh data. So, so since large amount of data is present your uh. So in, in case you uh know some buddies using uh gen, they have to play around with data, either theirs or some third party's data. So what are the ethical considerations uh around and uh so what? And also the security challenges that are associated with the use of GEN in fintech, uh not only around data but ah as overall.

Speaker B: So see uh, okay, uh, you know, fair point and I think you know uh, fairly uh established, you know, practice as well, right? Because look at this way, right? Uh, as a uh filtech, right? We are using the applications of generative AI, right? We are not building those models. I mean at least you know, in our case for example, at this point in time we are not building, right? We are utilizing some of the models which are already uh existing uh you know, uh, in the market, uh, uh, and when you do so you are interacting with the ecosystem, right? Now whenever you interact with any ecosystem, uh in the financial services which are outside your environment, there are certain uh, you know, basics, right, which you have to uh, take into consideration. For example data has to be you know, anonymous if not anonymous. Synonymous. Right? Uh, that's number one. Right. Number Two uh, you know, you're setting the right uh, you know uh, monitoring uh you know, aspect that nobody is you know, kind of sharing the sensitive data. Right. And hence the policies, the systems and the uh, you know, uh, subsequent uh uh uh audit, you know, plays a significant role. This second piece, the third piece is since you talk about ethical consideration since promote uh, it's you know uh, uh some states said right about let's say credit scoring and other pieces. See at the end uh, uh the basic problem with all the data is it has an inherent bias. Right? Because you just do not have the unbiased uh you know, data which essentially could tilt uh some of the pieces in a different direction altogether and uh, you know, can create an ethical or moral uh, you know, situation. Right. Um, in this case uh there are techniques, there are uh pieces which are available which you have to apply to make sure that uh, you are reducing or removing the bias. And hence uh, over time uh, uh while financial services have been this way, uh it will become more and more important is to have the cross cultural alignment and cross functional alignment. So if actually uh, it's not about pride team uh possibly figuring out something and just uh tech team building it and getting it out. Right. Even at every stage, your legal, your compliance, your uh, you know, diverse set of you know, people in the teams have to be involved to make sure that uh there's no inherent bias. Right. Which is uh, you know, coming into the picture. Uh, right. So that's a uh you know, another aspect which you will take care of it after that if you look at cyber security uh and other risk you know. Right. Uh, which remains there and given that uh you know uh, you are interacting with the ecosystem, uh you know, your data can be at risk. You have to make sure that you're using the right set of applications. Uh there are uh enough and more uh security uh pieces which are uh in place. And most importantly again uh, data has to be anonymous because you are trying to use it for the learning. You're not trying to uh, use it for anything else.

Speaker A: So Pramod, according to you what are the future trends that you can foresee for uh Genai in fintech? So how might advancements in this tech can shape the future of financial services? So basically how does uh, the future of fintech looks like uh to you? Uh uh with this uh new toolkit of gen AI.

Speaker C: Yeah, so very, very interesting. Um, so I'll just give you some quick uh reference point. Uh there are private banks say uh, and HDFC bank For example has this chatbot which is AI powered and essentially uh, what it does is uh, you are able to access and uh, have an experience 24 by 7. And um, with generative AI this can become lot more personalized and uh. So, so that's the opportunity when you're able to talk to uh a machine 24 by 7 have a personal conversation solutioning to your financial uh requirement which is uh, uh like I said uh based on your requirement or suggestive or predictive uh based on a lot of data. Similarly you have a bunch of index who are doing groundbreaking work on uh underwriting customer profiles uh with very limited credit history. So you have first to bank, uh first to credit kind of customer also coming in. And uh, uh you're able to put together information uh available uh on the financial side as well as on the non financial side uh together to take a uh credit call. So I think that's, that's a very interesting field and um, where you will have a good quality of underrating which is coming in. Underwriting in fact gets strengthened uh once you couple it with your financial information, your credit information along with your non uh financial information. Um, uh what um we also have seen use cases are for fraud detection. Uh whenever uh uh a system can pick up anything which is irregular, which is unusual in terms of uh your uh behavior in terms of transaction. Uh it can create an alert. All of us do get our AI based alert. A generative AI can um, probably uh sort of alert the customer ahead even if they see a couple of steps being taken. Uh for example somebody's trying to log in and try attempting multiple times through multiple ways. Now once the breach happen obviously you get to know and there's escalation in place. But a predictive uh agency can also alert you when they send something unusual happening. A different machine, a different location, somebody's trying to log in, you can get all that. So I think there's a very very interesting piece. Um uh some of the solid use cases which are really playing out uh already uh is on two levels. One is on the customer experience level and where we can talk about the financial uh advisory, autonomous financial advisory or we can talk about uh the personalized uh solutioning or we can talk about uh you know, trends being shared in trading and so on and so forth. And then the, the other side is really on the robustness of the process. Uh and which is where we're talking about efficient risk management. Uh we're talking about automated processing, uh we're talking about uh enhanced Customer security and uh, you know that kind of piece. So firstly I'm taking essentially the, the customer experience side of it and how it can really uh go to the next orbit so today. But uh typically customers seek is trustworthy on time uh and a good quality experience. Uh and every time there is a human involvement there is a possibility of an error. There's also a possibility of uh information uh not being run through the process. Uh and then which is where you, you always will have. If you can have a personal automated financial advisory I think that will be a preferred option because there's no questioning on the knowledge, there's no questioning on the compliance. All this is done by the process. Then the second part is uh your ability to introduce new products. Products which are whether it is based on demographics, based on location, based on consumption pattern, income pattern uh or our stated requirement. All this can be sort of. There could be a very wide range of uh, uh product solutioning there. Uh, I mean just to give you a quick reference you have mutual funds. Some of them will put only large cap, some of them will only put in small cap. Some of them will have a multi uh asset allocation kind of approach. Now what we're talking here is bring everything possible together uh and make it so tailor made for the customer. So somebody who's uh has a high risk appetite yet doesn't want to do uh, uh risk beyond five years. How is it that the solution will happen on the investment product? Solution will be uh amazing to see that for the first five years there'll be a very different product suit which automatically moves into a different kind of uh uh uh suggestion or recommendation from the organization side uh coming to the compliance part of it. Ashish has already covered. Compliance is the way I look at risk or uh strengthening the process is two ways. One is the compliance part of it, you know whatever is stated because compliance and the regulatory also is learning along with how the use cases are emerging. The other part of it what is ethical, right? There's a lot of information in the garb of taking services which is shared with the uh provider of services. Now how is it that you can use this information in a meaningful way and provide better solution to the customer and yet uh not breach any uh, uh level of trust, you're not breaching any privacy storing the data like Ashish also kind of uh suggested and I'm not talking about only in terms of system but also ethically intent wise that you're not double tailing it and you're not um maximizing it by you know just, just by sort of uh without taking consent and using it in so many other ways. And obviously compliance is a very very strong part of it and compliance is both regulatory which is in terms of how the data is stored, shared uh but more importantly how secured is the data uh throughout the existence.

Speaker B: Cool.

Speaker A: So we have I guess covered mostly the use cases that um are uh currently lying in the fintech space uh of genai. We have covered the security aspect of it. So one thing uh, I guess uh for our uh listeners. So um we, we have set uh ground that it is very important to you know collaborate with Gen uh AI. Now it is you uh know as it's a basic requirement now so but driving the adoption of Gen AI in FinTech. So as you know uh, there are still uh you know financial uh institutions that are evolving and you know adapting these genai. So what opportunities uh exist for partnerships between the startups and financial institutions and technology providers on those lines. Uh so what are your views on that?

Speaker C: I can quickly give you a reflection from my side. I think a bunch of things um, uh how a startup can innovate uh given that they're committed to innovate and they're committed to think differently. Right. Which is not being tested, not being uh followed uh by so many other organizations. So you'll have these established organization both on the tech side as well as on the banking side, uh the financial side and then you will have these startups trying to use the sandbox of innovation uh and then uh develop various use cases and innovate.

Speaker B: Right.

Speaker C: So uh, essentially a uh when you bring collaboration or a platform together wherein it can feed into each other and thrive together. Right. Uh it enhances the expertise, the set of expertise available in the entire trinity of uh the regulator, the established organization and the startups, uh the expertise and the capabilities which are available uh to develop something more meaningful which is more m groundbreaking which is more cutting edge, is significantly new which is what we're seeing. You'll find more innovation happening in a new age startup uh fintech and eventually getting adopted in the large organizations as they find fit. Private banks, the largest private bank in India today have adopted various AI based uh tools and learning and processes into their own mainstream processes which is able to, which is how they're able to sort of deliver their customer a better experience or create a robust uh infrastructure. Right. Then the second thing is uh, how would you at multiple levels innovate? Right. Both at a tech level and a process level and that can only happen when you uh, are trying it at two levels, one is at a customer level, the second is at a tech level. And uh, more often than not you do not want to risk existing customer base or existing processes by uh changing anything overnight. And which is where it comes in that you are able to do so much of R D, you are able to do so much of innovation disruption in a small sandbox kind of environment uh wherein the customer base is very small or nil and then you're able to adopt it after testing it in a larger base of customer. Third is uh, uh when you're able to test it and you have the collaboration from the regulator, you're able to run a compliant process and uh, uh uh give output. Then you're also allowing uh better adoption overall. Right. Uh, so there is uh less of uh, I would say no rail guards which are essentially blocking innovation to happen. And there's more framework which allows uh, for, for these kind of innovation to play out. Uh I think uh what we've seen in upi, what we've seen in uh the API based uh framework a uh lot of secured uh flow of information uh happening in which is also allowing a fast and uh, you know almost instant in flow of information is kind of created a very good user experience. Uh I also think as you have the industry and the regulator and startups coming together and investing their time, uh effort and energy in a singular direction, your ability to innovate is far more.

Speaker A: So uh Ashish, uh your views on this.

Speaker B: Uh yes. Uh so if you look at right. Uh for okay so even for the large banks, right Be the private sector, public sector, uh you know or any of the bank uh they are good in terms of adopting once solutions are stabilized. Right. So even for example the uh chatbots, right which is a typical use case uh spread across right. Uh I mean it doesn't get created uh you know, uh you know by the banks right. Uh typically which essentially mean there's a uh technology firm right. Ah which has to come and play a role in terms of creating those uh solutions now in between, right? Because there's a fin and there's a tech in between there's a fintech, right. I um, mean my way of looking at it now from a fintech uh you know standpoint right. Uh since uh you know you are trying to pick up one or two or three or whatever you know problems which you're trying to solve, right and you're very focused on solving it and uh uh you know trying to experiment, right. Uh because Fundamentally you have, you have to experiment, right? Uh, if you're a you know, kind of uh fintech firm, which essentially mean uh you know your ability to uh, navigate and uh figure uh out uh a solution would be much you know, faster, much better. Right? Which fundamentally has to you know uh, uh give uh advancement to the entire uh you know, financial services uh uh community which including the bank. So uh, collaboration becomes extremely important. Right? But collaboration is not stopping only at these three. There's a fourth angle which is a regulators. Right? Because whenever you test out any new uh technology there's bound to be uh, you know, uh, challenges and issues. Right? So effectively uh, and you don't know, right. Uh when you are you know trying to innovate. Because I mean if, if you know the boundary, uh then I mean I don't know how you innovate. Right? So hence regulators uh you know also have to play a significant role in terms of providing a regulatory uh a sandbox which has a uh, you know, certain degree of uh freedom which is available now, four of them coming together. Look at this. Right? Uh, regulatory uh sandbox given to uh you know, fintech which can uh, you know, experiment fast uh, which. Which are uh, you know by the nature of uh you know the existing uh existence uh you know, are agile in nature, can innovate you know, much faster. Right? Uh, collaborate with the bank because you know, banks you know have those uh licenses, right. In order to create those uh you know, unique or innovative product which will fill uh you know uh, the regulatory environment or uh requirement as well. Right? Uh, add to which is our technology layer, uh which is uh, you know about you know creating those uh uh, uh uh for the, for the, for the lack of better word, you know, hardcore you know, platforms right. Which can be uh, you know, fit in uh and you uh know uh, you know, make it work at a scale and then you start uh, you know, rolling out, you know, PC one by one. I, I'll give you one example and uh, I'll also you know pick up uh, uh one of the question which you had earlier, right uh, uh, which is about the use cases.

Speaker A: Yeah.

Speaker B: Right. I mean I've been in financial services for what 17, 18 years now and uh, one of the problem which I've seen consistently across, right uh is everyone trying to put their arms around uh personalization and uh, ah omnichannel experience. Right. Both of them have been uh, you know uh, uh, work in progress for ages and have always hit uh a certain degree of uh uh roadblock. Right? Because you have, you can play around with the limited set of data that can give you certain outcome. But if uh, you have to have omnichannel, then your digital doesn't talk in the same way your physical uh, you know, uh, talks or your physical doesn't talk in the same way as your physical talks.

Speaker A: Right?

Speaker B: Now uh, if you have to have this kind of uh, problem which you have to solve and which I strongly believe can be ah, solved now with the generative AI, given that, you can now start adding context on the data, put this into a ah, meaningful ah contextual argument which can be ah, uh, in a human uh, readable or human actionable form as well. Now if you have to implement this at a large scale because it takes a lot to move the elephant, right? That's where you know, possibly you know, fintech, right? You know, assuming generative AI, you know, entire uh, you know, platform, everything has been created by the technology, uh, you know, from the fintech firm, you know, can you know, pick the use cases, marry them, you know, uh, in a more meaningful manner, uh, experiment at a scale and build a scale which can be, you know, possibly utilized, you know, by any of the entity which is involved in this, you know, entire piece. Right? So because see at the end it's not technology which runs the business, right? It's a business, uh, which utilizes a technology to run the business, right? And uh, whenever uh, you know, you have a large, you know, volume of the customer as promoters promote also highlighted, right? Uh, it's your ability to experiment goes, you know, down. And hence this collaboration is definitely required not uh, only to uh, you know, experiment or innovate but also to you know, uh, make the entire, you uh, know, process, uh efficient for the end customers because at the end it's a end customer who has to get impacted. So yeah, so uh, you know, four of them coming to. I think the uh, magic can happen standalone. M. Uh, uh, there's a one uh, link or the other which is missing.

Speaker A: Yeah, true, true. So, yep, I think we have covered all aspects of how Genai is transforming the fintech industry. So um, thank you Ashish. Thank you Pramod for your valuable insights and participation in today's discussion. I'm sure our listeners will find plenty of takeaways from today's conversation as well. It was a great experience for me as well moderating this enriching session. So thank you again Ashish and Pramod. Um, we'll be back soon with another interesting topic. Until then, signing off, have a great day.

Speaker B: Thanks, Amandeep. Thanks, everyone. Thanks for having me.

Speaker A: Thank you.