Saurabh Gupta — President, HFS Research[00:21]
Hi everyone, welcome to HFS Unfiltered. Today, we’ll be talking about AI transformation and insurance. What a topic. I’ve got two gentlemen to really describe it and have a lot of war wounds and good experiences, I hope, to describe that journey. Before we get started, Prashant and Ian, welcome to the show and thanks for joining. Why don’t you quickly introduce yourself and what you do, but also tell me when you hear that AI in insurance, what’s the one word that comes to your mind as you’re describing yourself? And maybe Prashant will go with you first.
Prashant Hinge — Chief Information and Transformation Officer, MSIG USA[01:12]
Sure. Thank you. Thank you for having me here. It’s a great pleasure. So I’m Prashant Hinge. I am the Chief Information and Transformation Officer for MSIG USA. We are a specialty insurer here in North America connected to our parent company that resides in Japan, MS&AD. My role here at MSIG USA is to drive transformation, leveraging technology, data, and people. So I’m responsible for the technology, infrastructure, data, and also the operation side of the house. So I’m really looking forward to this conversation. Oh, yes. And one buzzword that comes to mind to me is it’s a buzzword. That’s what comes to mind when I hear that.
Saurabh Gupta — President, HFS Research[02:08]
Fantastic. I beat that now.
Ian Maher — Research Fellow & Senior Advisor, HFS Research[02:13]
Yeah, I’m sure I can beat the buzzword, but we’ll try. I’ve worked on both sides of insurance, P&C, Life & Annuities, some Lloyd’s syndicate work in the days of renting sourcing, and a lot of operations on the business and technology side in the States. And then more recently, working with a broad range of carriers and intermediaries and insurances, they face some of the issues that we have today. I’ve been lucky enough to know Prashant for many years. And you’re right, we have lots of war wounds and experiences that all research does, I think, just to give us a confidence to address change. You know, that might be the word that I would go through. You’ve got a very, the oldest sort of commercial business industry in the world in insurance. And you have the most, you know, recent emerging technology that has got, you know, so much hype from above. And in working with the likes of Prashant, you know, our role is to manage expectations, is to understand what success looks like. And so we don’t spend bad dollars in the wrong areas. And so I think the way would be excitement, I think there’s a huge opportunity, but with that opportunity and expectation comes realism. So I’m going to give you a couple of words, Saurabh, as to where we are with insurance and AI.
Saurabh Gupta — President, HFS Research[03:43]
So you gave me three, change, excitement, and somewhat of hope. So look, I think, Prashant, I need to ask you, why did you call it buzzword? Is that because there is a gap between ambitions and reality or something else? I have to peel that onion a little bit.
Prashant Hinge — Chief Information and Transformation Officer, MSIG USA[04:09]
There are. So number one, if you go back 20 years, every three to four years, we have replaced AI with some other term in insurance. It was robotics. It was digital. It was big data. It was service-oriented. I could go on and on and on, right? So that’s the reason why I think it’s a… We’ve been doing this for the last 20 years. The main reason why I think it’s that is because, number one, AI is a tool. It’s a capability that is extremely powerful. And what it drives is new ways of working. Change in an organization. You could do more with less. And where that comes from is from real people and a solid business strategy. And now AI becomes a tool for you to achieve those goals. So to me, this is another very powerful technology, no doubt. But until you apply it to the right problems and apply it at scale, it still remains a buzzword because you don’t want to just do AI. You want to solve a business problem using AI. So to me, until it’s clear that it’s a business problem that I’m solving for, that’s going to be meaningful to the top or the bottom line, to me, it’s just a buzzword.
Saurabh Gupta — President, HFS Research[05:36]
That’s very well said, Prashant. And Ian, from your vantage point, what do you think is the gap between what’s holding us back, right? From making, instead of being a buzzword to making it reality, what’s holding enterprises back?
Ian Maher — Research Fellow & Senior Advisor, HFS Research[05:55]
I think, sorry, I was able to look across all industries and then think about insurance. What’s different about this? I’ll describe it as technology. It is that the incubation period that the providers have had to understand how this change to a service delivery model can take place is the shortest that I think we’ve ever seen. And you have providers that in many ways are still learning actually how to integrate AI into their service delivery offerings. So you or Prashant would turn to the market and seek experience from the outside and bring it into an MSIG. In many ways, what I see is that digital transformation leaders, chief information officers, they’re having to actually learn alongside their trusted partner. And if you don’t have trust in that partnership, they’re all very isolated. And often we see sort of a fail rate far higher when you don’t have that partner to actually go forward with. So I think that that’s the environment that enterprises, carriers, intermediaries, MGAs, etc. are all facing is what do I turn to to actually understand what’s the right use case to Prashant’s point I need to solve a problem not just see this as a gimmick and think that there’s something to be gained from a gimmicky nature of this because if you go down that path Prashant I think you would agree you will end up actually making things so much worse for the enterprise the employee experience the customer experience the brand reputation of failure and those are the worst things in the world that we, you know, we want to see so I think that is really what’s shaping this period of where we are in terms of AI.
Saurabh Gupta — President, HFS Research[07:56]
Prashant, do you agree with Ian? No, I agree.
Prashant Hinge — Chief Information and Transformation Officer, MSIG USA[08:01]
And the proof is in the pudding, right? So when you look at service providers or any contracts, the good rule of thumb is what is, I think you use the term, reality versus what does your mind and body say, right? Aspiration. So between reality and aspiration, the simple question we can ask is, as a buyer, if there is an engagement with any of the partners, etc., how much of lift we are getting through that? Productivity could be other ways to capture the value on leveraging AI, compared to what’s being said in the marketing [unknown] to say, we have new AI practice, X number of people with AI, X amount of revenue. I get all of that. But if the existing incumbent customers are not seeing that value, it just becomes a marketing slogan, right? So I think this is where there is still that chasm we need to cross and create a win-win-win for everybody involved, both, not just for new dollars that are coming in from a service provider perspective, but even for existing customers. And that’s how I think service providers can create that long-term stickiness. Otherwise, you lose trust because you’re speaking from both sides of your mouth.
Saurabh Gupta — President, HFS Research[09:29]
Yeah, and Prashant, I think the key word there is value. And the question that I wanted to ask you what is value is it just productivity is it doing like doing things faster and cheaper or do you think it’s fundamentally going to reshape how insurance does underwriting or claims or you know fundamental operations how do you describe the value of AI is are we sort of going down the rabbit hole of, you know, let’s keep doing things faster and cheaper, or I think you mentioned earlier that it’s a new way of doing things. So, in your words, what does value really mean when you talk about AI?
Prashant Hinge — Chief Information and Transformation Officer, MSIG USA[10:13]
Right. So a couple of key aspects here. Number one, value contribution to the P&L, right? That’s one of the important things on how we measure. What we measure is a top-line impact, bottom-line impact, and are we able to differentiate from the pack or create a differentiated environment as a carrier for our customers, right? I think that’s the most important aspect of it. It’s creating differentiation using technology, and you can measure that in different forms. And the number two piece of it, which is to me, it’s on a spectrum. Yes, with any tool deployment, which AI is a very powerful tool again, there is going to be immediate impact to your hygiene metrics, right? Your speed, your quality, and other basic KPIs that we have operationally measured over many, many decades. But the question really becomes is thereafter, now, what is, how do we create the right level of amount of guardrails that is going to use the framework that we have built? And now you can absolutely do more with less, not linearly, exponentially. So on one hand, it’s about creating differentiation in the marketplace so the customers feel that we are delighting them every day. And on the other hand, it’s not just doing 10, 20, 30, 40 percent more anymore. It’s about doing it exponentially better because the technology is that powerful. So those are the two sort of rule of thumbs that I’ll look at this equation and qualify some of the use cases or investments that we do.
Saurabh Gupta — President, HFS Research[12:11]
And that’s very, very well articulated, Prashant. But Ian, that itself is a dilemma, right? Because on one hand, you have this need for speed and, you know, we need to innovate as of yesterday. We need to digitally transform. You need to sort of improve on the bottom line. But then, especially in insurance, it’s a highly regulated market, compliance, regulatory issues. You know, AI creates perhaps even more risks. How do you balance all these things? The need for speed, cost, innovation, compliance. What’s the trick to balance all of these things?
Ian Maher — Research Fellow & Senior Advisor, HFS Research[12:54]
A lot of leadership in the carrier is exactly weighing up all of those things that you’ve just said, Saurabh. But we’ve got to remember, particularly the space that Prashant and MSIG play in the commercial and the specialty space, that that market itself is growing fastest in the nominated segment of the insurance marketplace where the carriers can actually start to write their rules about the product and the coverage areas separate to what the state may have required from a filing and a standardization point of view.
So that is the largest segment in the commercial lines that is growing some other different words gap appetite in that business already, and I think for the carriers I think it comes back to recognizing that ultimately there are touch points with the policyholder all the way through the insurance life cycle. Those points of value creation haven’t changed, so the claim experience remains the number one experience as to why a commercial client of MSIG will want to stay with MSIG is at that moment the truth as far as the claim goes. So for the carrier, I think you have to start to create in parallel with your standard offerings open discussion with your largest clients, whether it’s direct commercial businesses or MGAs or wholesale brokers or whatever it might be, and discuss the willingness to introduce AI into the service offering with that segment directly. Explain this is the risk that we might face, but this is the reward. Really, you might get a competitive edge in front of your retail brokers, where you may be looking to establish your specialty in a niche marketplace.
If we introduce this underwriting, a set of techniques that embody the ability of AI to be more accurate or to improve my underwriting, then there are significant returns for us, but we’ll need to both take that risk. We need to be both part of that planning session. So a specialty player has to look, and I think segment, and have those dialogues as they do. The leadership and the relationship stack between the carrier and the commercial client is absolutely the point of truth and those discussions need to be taking place at the CEO level, the likes of Prashant being part of those discussions, because the impact of what Prashant builds within the carrier set. It was truly flowing through not just the carrier but to the distribution side as well, and this is where the influence of people like Prashant and lots of the best leaders in this industry have got the foresight to recognize that the impact is not just within my own organization. It’s going to be experienced upstream all through to the sales and distribution side, and so I think, in summary, I think it’s been open and transparent to choose the clients that you’re willing to create proof of concept to share the risk, because the rewards to that commercial client can be tremendous in terms of creating a differential against their competition.
Saurabh Gupta — President, HFS Research[16:12]
So Prashant, what Ian is describing is not technology or process or operations. He’s describing it as a leadership issue, if I’m getting you right, Ian. Do you agree? And then how are you as a leader personally? What are you doing personally to sort of move the needle on this track.
Prashant Hinge — Chief Information and Transformation Officer, MSIG USA[16:39]
Right. So the teams are aligned is the most important thing. And when I say teams, I agree with you. It’s less about technology. It’s more about other factors, starting with leadership. Number one, CEO level down. The CEO and his direct reports, and aligning on what our short-term, long-term goals, and targets are, and being very clear on how we are going to achieve that—the operational tactical methods, and that AI is going to be a driving force to drive productivity. And then over time, exponential value for us as we get the base foundation together.
So to me, that is more important than what an LLM can do on an ACORD form. If you don’t have that, what ends up happening is you create point solutions in the middle of a life cycle of a process of insurance. And while it may work, it may work really well, but you’re surrounded by all the legacy, the silos, the tech debt that you have in other areas. So even though you are doing really well as a whole enterprise, you are not creating, but you are destroying value, right? So to me, leadership, that’s why and making sure there is alignment across the board is the most important factor.
Saurabh Gupta — President, HFS Research[18:09]
Now, I couldn’t agree more, Prashant. If you look at the data, practically everybody has some sort of GenAI or an agentic AI pilot or a POC running in every nook and corner of their organization. But there are only single-digit numbers of these initiatives which actually scale up into a production-grade environment where the ambition is meeting reality, no matter how you… And we’ve done lots and lots of studies on this, Prashant, and no matter how you slice and dice this data, the answer is always 5% to 7%.
And I think when you peel the onion, it’s exactly what you just said. It’s not because the technology doesn’t work. It’s organizational silos. It’s tech debt. We don’t have the right data. People are not scared, or people are scared of this, and all of those things. So, I think the one other topic that I wanted to ask you, Prashant, is that you very eloquently talked about the top-down approach on how you need to align with the leadership and then sort of create the goals and create the how to achieve those goals. What about the bottom-up? Because what we are seeing in our research is that a majority of the employees are actually scared of AI. So we might be using AI on our, sitting on our couches on Sundays. But then when it comes to Monday corporate life, we are a little bit scared because, you know, the narrative is AI is going to take away jobs and all of that. And if the employees are not in it, then how do you scale this up, right? It’s ultimately everything is related to people. What’s your perspective on this whole talent equation? Because I think that’s an unsolved question.
Prashant Hinge — Chief Information and Transformation Officer, MSIG USA[19:58]
Absolutely. So it’s really about hearts and minds when you think about it ground up. And technology has, as it’s evolved, matured, it’s just going to do more and more work that was done by humans before. And to me, the future is still more hybrid, integrated with human in the loop versus just technology or AI running by itself. I don’t see that at least happening anytime soon. So the best way that it worked is, I don’t think this is any secret, is by engaging the people at the right time, is by communicating what the today and the tomorrow is going to look like, what roles do we want them to play.
And also, we are all getting well-versed on AI on our own in our personal lives. We want them to bring that version of them to work every day, right? So we have a concept we call as an employee triathlete, where there are three things that are going to be required for us to flourish in future. Number one is your business context. No matter which part of the company you sit in, you’ve got to understand the business goals, the business strategy, processes across the value cycle, right? Not everybody needs to know everything, but you’ve got to know some processes in and out. That’s a business context. Number two is AI skills. And AI skills are maturing every day. It’s really about prompting, flow engineering, those type of activities, which you can practice in your personal life and bring those to work. Or there are other tools at work that are going to help you do that in a safe, secure manner, right? Anyone that says that, anybody that says that I can’t use AI for my daily job, they need to look at it more closely because I’m sure there are aspects of jobs that can help you do better. And the last but not the least is your functional expertise, whatever your roles are, be it in technology or in other places of the organization. That’s your functional expertise.
All of these three combined, that really becomes a massive catalyst for change in the organization. We are no longer in this era where we need 40 people teams to implement new solutions. We are going into a very small—three to five—people oriented team that is going to solve for outsized problems, right? And that’s why you need this triathlete part of that group helping solve for that. And these are not just people that you can hire anywhere, right? These are people who are willing, entrepreneurial, able to learn and able to understand and see what their blind spots are and how can they fill those with help of the leadership and keep moving forward.
Saurabh Gupta — President, HFS Research[23:08]
I love the triathlete description that you have, Prashant. I’m definitely gonna steal it from you, but before I let you both go, let me just ask you, for the people who are listening to this, you know, what’s the one mistake to avoid as you think of scaling AI and actually moving the AI excitement and the buzzwordiness to reality. Ian what’s it from your perspective—what’s the one mistake to avoid?
Ian Maher — Research Fellow & Senior Advisor, HFS Research[23:55]
I think the one mistake that I’ve seen over the last 12 minutes or so, Saurabh, is actually on the enterprise side trying to force a change with the providers that is a false change. It’s a change based on [unknown]-based economic aspirations of what AI can bring. And these, like when they hear enterprises, you know, wanting to break a contract and reestablish a new set of economics based on this thing called AI, it tells me so much about where that relationship truly is.
And for me, the mistake is to approach this unilaterally as an edict, either externally or even internally, where cost takeout is being challenged for some crazy number. But this industry will only succeed collectively. And, so I would say carriers, the Prashants of this world, must be having these dialogues with the solution providers and partners today and planning for the success and the benefit realization that AI can bring in 6, 9, 12 months. This is not a new lateral step to take. This is not a master service-servant relationship. And it’s asking telling you, and it means we’ve spoken to a number of clients, how the relationship really is between the enterprise, the recipient, and the service provider. And so you better have strong relationships because you’re going to need each other. I would say that’s the starting point for success. It starts with leadership alignment, and it will cascade all the way down through the organizations. If you do not have it, you will not have the insight, the market knowledge, and the skill sets from the service providers for you in the carrier to succeed.
Saurabh Gupta — President, HFS Research[26:04]
Fantastic, Ian. Prashant, what’s your advice for one mistake to avoid?
Prashant Hinge — Chief Information and Transformation Officer, MSIG USA[26:12]
I think for me, if you want to run a marathon, you just don’t show up on the day of marathon and start running. Rain, cold, heat, no matter what, you got to practice and prepare and plan. And even when it’s comfortable and uncomfortable, even if you want to or don’t want to. Now, the business equivalent to that is, really, we’ve got to have the right foundation, right? When it comes to process, data, integration of your end-to-end systems, wherever possible, and then comes AI to actually drive the value for you. If you want to directly jump to the marathon day, in this case, it’s AI. You’re going to do one thing right but when it comes to scaling you’re going to keep, you’re not going to be able to finish that marathon. So to me it’s do the planning, do the preparation, and that’s where it’s going to take the hardest, that’s where it’s going to be the hardest thing to do but it’s necessary to be really good.
Saurabh Gupta — President, HFS Research[27:21]
Fantastic guys I think there’s so much to unpack in the 10-15 minutes that we’ve been talking but I think my takeaway is that this whole AI dreams versus AI reality is not really a technology question anymore. Prashant, as you mentioned, it’s about getting ready for that marathon. AI is a marathon, and you need to get your processes, your data, your people, your leadership in line. Otherwise, you know, it’s going to continue to be a death by a thousand POCs versus really scaling it up and bring your people along. I think, Ian, that’s very well articulated, whether it’s your customers, whether it’s your vendors, whether it’s your employees, whether it’s your leaders. You know, we’ve got to come together and collaborate on this. But fantastic conversation, Prashant and Ian. Thank you so much for taking some time out of your busy lives and sharing your perspectives with us.
Prashant Hinge — Chief Information and Transformation Officer, MSIG USA[28:27]
Thank you. It was nice to talk to you both. Thank you.
