1091. Insights: Can AI make us financially healthier?

Hello and welcome to FinTech Insider Insights, I am Ross Gallagher, the head of consulting here at 11FS. Today we're talking about financial wellness and whether we need to rethink what that actually means. For decades improving financial wellness has largely meant improving financial literacy, teach people to budget, teach people to save, teach people to invest and teach people about pensions and debt. But knowing what you should do with your money doesn't necessarily make those decisions any easier. Now AI offers a very different proposition. Instead of simply giving us more information, AI could increasingly understand our circumstances, recommend what we should do next and potentially take some of those decisions out of our hands entirely. So is the future of financial wellness really about making everybody better at managing money? Or could it be about reducing how much managing they have to do in the first place? And if AI starts making decisions for us, does that make us financially healthier or just create a completely new set of problems? To discuss this, I'm joined by Lucy Weiman, co-founder of wellness. Lucy, welcome back to the show, it's lovely to see you again. Maybe you would just reintroduce yourself through our listeners and tell us a little bit more about wellness. Sure, thanks so much for having me on again. As you mentioned, I'm Lucy, one of the co-founders of wellness. We are on a mission to make regulated financial advice accessible and affordable for mid-market women. So L is our proprietary AI, she helps women understand their gaps with their financial score, she helps them build confidence and connects them with female advisors when the time is right. So this conversation feels very on-brand for us because we are living inside the line between AI guidance and regulated advice kind of every day. So I'm excited to dive into talking about it. Yeah, likewise. I mean, it's such an important topic. So it's great to have you on and sharing your sort of firsthand experience and perspectives with this Lucy. So thank you so much. Next up, we have another one. Welcome back to the pod for Aditi Subaru financial services for Snowflake. Aditi, it's always great to have you on the show. How are things going both yourself and with Snowflake? All well, thank you so much, Ross. Always great to be back on the show. Things have been very exciting, as you can imagine, with lots moving in the world of data and AI at Snowflake overall. I've been spending a lot of time with banks and financial services customers in actually helping them make the most of AI from an outcomes perspective. And the focus has very much been on how do you actually get meaningful outcomes? How do you explain to the AI the context of the underlying data that they're working on? And that can have significant impact on the benefits that that AI can help financial services firms and also their customers achieve. So it's been a really interesting few weeks. Yeah, I'll bet. And I love you framing it in the context of outcomes, because I think there's just so much potential for improving those outcomes for those end users, as you so right, you say, and I think we're going to get into a lot of that as we move through the show. But that is our awesome panel. I think we've got a really interesting topic. So I think we should dive right in. We mentioned it at the top of the show in the intro. So I suppose starting with financial literacy seems like a reasonable place to start. I think this is maybe the sort of the little mismatch that we've seen traditionally. I think we've largely treated financial wellness as maybe an education problem. We sort of give people more knowledge and better tools than they'll make better decisions. But I don't know that that's that's really played out sort of how we expected. I don't know if that relationship has necessarily been that straightforward. I think we'll unpack that as we sort of get into get into the show itself. But Lucy, I think it'll be an interesting place to start. Maybe just take a little temperature check. What do you what do you what do you think being financially well actually means? Yeah, it's a good question. I think it often or certainly historically would have been reduced to just, you know, have I got enough money for the things that I need or want each month. But that's a pretty narrow slice of kind of bigger picture financial wellness. I mean, for us at wellness, we have this financial score and it works across six pillars. So, you know, budgeting's part of that, but it's also protection, your mortgage, pensions, investing and legacy. And we're really trying to help people see their gaps across this full picture. And we're definitely still seeing lots of gaps. I mean, half the women we speak to have never checked their pension, nearly two thirds have got no will and half have got no critical illness. But I think most interestingly, the average confidence score we see is 5.9 out of 10. So, there's still, you know, a gap there in terms of just general confidence around financial decision making. And when we score, we're looking at a mix of practical and knowledge base. So, we're thinking about education and, you know, actual gaps. So, we might ask, you know, have you got enough income protection to cover six months about pay, but also do actually understand the difference between a cash isre and a stocks and shares isre. So, I think we kind of see literacy, I don't see it as kind of financial literacy, then confidence, then outcomes. I kind of see them all being very linked. And I think, you know, knowledge is kind of the unlock. Rather than it being a separate thing, we still need it, definitely, you know, without it, you don't know what you don't know. And I think if you don't understand your vulnerability is, then you don't even know what needs fixing. And you're certainly not going to take action on doing anything about it, but equally, knowledge alone doesn't move the needle. I don't think, you know, you need these three things together. You've got to understand what's wrong. You've got to feel confident enough to act. And then you need a system that actually helps you make the right choice, rather than just showing you the options, I think. I think that's spot on. And I think you're exactly right. I mean, you know, the sort of financial wellness and maybe financial literacy or the knowledge that we talked about, they're not necessarily one in the same, but they also don't exist completely in isolation. And I think that's such an important point. A DTM really interested in your perspective as well around what does it actually mean to be financially well? So I was thinking about this, especially when I knew what the title of the show was, but also more broadly, obviously, and the way I see it, I would almost bucket it into three pillars. So financial wellness as a whole to me is effectively a combination of knowledge, access and agency or choice. So the first thing exactly, like Lucy said, you need to know what you need, what is good for you and what you should be doing. And that knowledge or that literacy is effectively the first unlock. The second piece though is access. If I know I should have a pension, but I need to jump through multiple different hoops and obstacles to even open a pension account or like do what it takes to transfer money from one account to the other, I might know what I need to do, but not do it still. So I think having that access and whether that is under the bucket of financial inclusion for the underbank to whether that is just making the elderly and the vulnerable, like more prone to use digital banking than they are otherwise, it spans the whole gamut. But access is a big part of financial wellness. And then the third piece around that is the agency, which kind of ties into the confidence. I know what I need to do. I can do it, but am I brave enough to press the button? Do I have the confidence to close one account and open the other? Do I have the ability to say, well, this is a great SEIS scheme. And therefore, I want to maybe invest in it. All of that needs to come together to then give you both a feeling. And I think this is where the the interesting thing is, right? That it is one thing on paper. It is a whole other thing emotionally to feel and believe that financial wellness. There are so many people, especially every older generation in my family who are very well on paper financially, but they still feel scared. And I think that link is also something we often tend to miss in the broader definition of financial wellness. There's also something very interesting here. And this kind of reminds me of how how my kids were taught to keep washing their hands in COVID. They were explained why they should do it. They had all of these songs. They had like multiple different drills and practices of washing their hands. However, the moment somebody is not looking over their shoulder, they would stop doing it. And it's very interesting that there are a number of different surveys around financial literacy, more specifically, would say that even if you know what you should be doing, you are very, very unlikely to actually keep doing it unless forced into it. So I think that bridge is also something that we need to be very aware of when thinking of literacy as it connects to wellness. At least at that point, the DT made about confidence is so. important, isn't it? Because it sort of goes back, I suppose, to some of the statistics that you mentioned around where there are gaps in people's financial lives, and there were some really staggering statistics in the Mills report, the FCA's recent report, sort of setting out their perspective around the impact that AI is going to have on financial services. The advice gap, so only 9% of consumers currently use traditional advice, just about 30% hold life or income protection. You've got really low numbers around accounts switching. You've got around 900,000 customers that are currently unbanked. You've currently got about 300 billion pounds worth of savings in the UK sitting in low interest accounts. And I think maybe some of those statistics just go some way to reinforcing a DT's point about that, that lack of confidence in people not necessarily feeling comfortable or backing themselves to take some of their financial decisions even though they might know in their heart that it would benefit them in the long term. Yeah, definitely. I think, you know, you've really hit the now on the head there. I think consumers do want education. They still need education, but they also want more than just guidance. You know, that's why I think human advisors do still have a very key place in, you know, financial services, at least, you know, AI can unlock a lot, but without that confidence, people don't actually make decisions. They feel sort of stuck in this overwhelm of like, what should I do next? And it's also why we have to be careful because some people are going to just, you know, General LLM's for validation on the financial decisions. And, you know, honestly, they shouldn't be. And I think it, you know, even with all these tools available, money still often fills just too difficult or you're going to get to it later. It's, you know, unless it's really pressing as in I'm running out of money, I think it's hard to keep that urgency and to, especially if it's outside of people's comfort zones for them to have that confidence, like you say. And, you know, one of the things that we get kind of push back on a lot is, you know, well, why would someone use wellness, couldn't they just talk to Claude or ask chat at GPT, but I always kind of say the same thing, which is, well, they could, but will they, you know, like, are they, are they actually, do they know the right questions to ask, do they have the impetus to ask them, even if they ask them, and they get an answer, do they have the confidence to actually act on them without input from someone that they feel can genuinely give them advice and tell them that it's the right decision to make. So I think, you know, the goal is really to like find that tipping point of what is going to make something, you know, what's going to make someone want to solve the problem right there and then rather than it being filed away or rather than them losing their, their confidence again. And I don't mean, you know, the super fun actually savvy, I'm kind of talking about it every day, consumer. And I think, you know, I love that example of your children washing their hands because I was thinking about my own examples and just thinking about some of the content that we do, for instance, you know, we run a webinar on the pension gap. People on the call, they're so engaged, they're like, oh my god, I've got to sort this out. Yes, you know, they totally buy into it. But the second they leave the call, life just kind of gets in the way again. And then I think, you know, there's a gap between like knowing what you should do and actually doing it. So I see a real role for AI to play in helping nudge people to make decisions as well. And that's something that we've done with L is that she can you know, notify you if something changes in the market or if you have a life event can help, you know, show you what that might mean you need to do. And I think, you know, yes, it should be, it's also the information and it can help with that education piece. But like, how do we find a way for it to keep showing up at the right moment so that people's good intentions, don't just evaporate, you know, the minute that they leave a webinar or the minute someone stops watching them wash their hands, you know, how do we kind of keep that impetus and actually make it lead to action? I think that's, that's a very distinct kind of phenomenon that we're seeing popping up in the AI world also in general. So I spend a lot of my time helping banks kind of automate business processes or use AI to like make better decisions like me spoke about before. What you said, Lucy, is validated completely. And I had once read about this really interesting study. I should, I should have looked it up actually. But back in 2014 or 2015, they ran a study in the US, which was actually kind of a meta survey. So that study itself studied more than 200 different studies that came before it in the world of financial literacy specifically. And they wanted to determine how much of a change in outcomes or overall financial statistics did literacy and financial knowledge or being on courses, et cetera, make in participants. Guess how much the impact was? I don't know, I should guess because I also looked at this survey, but it was practically nothing. So I think that's kind of the point here. And the reason I deferred back to the whole automation piece with AI is that when it comes back to Lucy's point of having that human advisor, somebody who's going to push them to actually make things happen, somebody who's going to give them that to your students and saying, I've got your back. Like AI can never tell you that it has your back. So this whole perspective, which, and I've mentioned this before, even on this show, like I love which NatWest has a tagline around it, which goes around moments that matter. So it is those high impact decisions, which actually need that human intervention or a person or somebody whom you can rely on, whether that is buying a new house, whether that is, like you're having a baby and you want to figure out how your life insurance is set up for that, et cetera. Those moments that matter is somewhere where that advisory needs to come in and push you to actually make that decision and do the writing and make that action happen. The interesting thing, however, is that if you look at the market across available apps and AI tools and even robot advisors, et cetera, to some extent, a lot of those are very short term, very tactical kind of decisions and interactions. Should I top up this amount and round it off and you can invest in charity or you can put this into a savings account? Yes, you can. Or do you want to forego this coffee and then like invest in your pension with it? Yeah, maybe. However, that there is a big gap between these short term things which we rely on AI to do or help us to versus the really long term high stakes, high value decisions that are critical to ensure financial wellness. And I think even in the AI world, there is still a big gap there. It is the big ticket items. It is the long term thinking that needs that buffering/support. And they're not really seeing enough there in the space of AI versus the tactical, small ticket behavioral items where there is a lot of AI available. So I think there is still that gap in terms of making actions happen for wellness. Do you think that we're starting to see though sort of movement in that direction? And I suppose there's a couple of things. Again, if I refer back to the Mills report that sort of maybe suggests that to me. So I mentioned about the advice gaps. You've only got around about 9% of consumers in the UK that use traditional advice. But then you've got one in five UK adults, so about what I was saying, 20%, open to AI making decisions for them. You've got around 26%, trusting general purpose, Lucy, tools such as chat GPT, Claude Gemini, etc., for financial advice. So I suppose I was starting to see a genesis and maybe the difference between that 9% seeking out sort of traditional human advice and the 26% or so that are comfortable doing it through AI. Does that suggest that we're sort of moving now much more in that direction? I would still like to kind of double click a bit on the nuance there. So most of these statistics which we tend to read and surveys, especially of people interacting with or using AI. Most people are using it as a glorified Google or perhaps as money supermarket on steroids. So if you actually look through those queries and the prompts that are put in, most of them tend to be explain this to me, i.e. what is this product, what is this offering, what does that mean or do a comparison. Should I set up a current account with this bank versus that bank and what are their interest rates and what will that mean for X amount of money or should I draw out an insurance policy from this place versus that place and what does that mean? Very few people are actually going to chat GPT and saying this is my situation, this is what I do, this is how much money I earn, I want to plan for retirement which is 30 years from now, give me a financial plan. And again, even those people who do do it are highly unlikely to actually give the AI all the information that it needs to be able to give that advice in a holistic manner. And it is often very fragile. are straight because these surveys tend to almost mislead people who are building products or who are building AI tools because they tend to think that the reception and acceptance is much higher than what it actually should be and therefore the problems that they are trying to solve tend to be the wrong problems. But we almost need to figure out, yes, there is that receptivity, yes, there is that interaction which in itself is very encouraging because to Lucy's point, if that first unlock of literacy and understanding and knowledge about financial services, offerings and product and risks is taken care of by the AI, then the institutions, the regulators and the advisors are already starting off from a level which is much higher than what it would have been otherwise and then they can spend their time and effort in genuinely advising people for sustainable, more impactful outcomes. So I think that is my bug bear on AI-related statistics of financial usage and interaction. I think that's brilliant though because we all get swept up in the hype cycles right and we all get a little bit carried away when we read some of those things but actually maybe just grounding ourselves in that little bit of nuance and taking a more realist view on where we're at right now is no bad thing. What do you, Lucy, do you sort of agree with that in terms of where we're at? I do agree and I feel very happy hearing that because that's kind of the hypothesis that we're building our business around is that actually particularly for women, AI adoption is still a lot so then it is for men and I think there is still a trust piece like exactly like you say, we're very happy using it for research but I still think there's a big trust gap around taking a decision, making a really big high-states decision from an answer. I think if we think about what is happening, which you've kind of alluded to already, it's really great for how that building, I think we see it, what you say in these small things that we're doing already, so for instance for me I use plums, rainy day pot or whatever that automates something that I'm already quite comfortable with as a concept but I think if you're, you know, if you're happy putting money manually into a cashiser you're probably happy with AI taking on the automation of that but would you be as comfortable with automation moving your money into something less familiar, you know, an area that you don't understand as much or that you've not touched before and so that kind of goes back to this whole knowledge plus confidence plus trust piece which is, you know, if you're happy to do something manually and you understand the concept behind it, you're probably happy to automate it and hand it over but, you know, if you're not at that point, I think it's a different conversation and I also worry a little bit about the kind of convenience of these tools at our fingertips, like stripping away understanding and I kind of even see that in myself, you know, using Claude for tasks that maybe I don't fully fully understand and then if it gets things wrong that I don't know enough about the topic to spot it and I think that's a real kind of worry with people using it, you know, one is great that it can educate and inform but you've got to have that grounding because I think if you're coming at it without that sort of base layer and trying to get it to, you know, make comparisons for your, make decisions for you where you don't fully have, you know, a comprehension on what it's doing, that is a big risk. So it's, you know, it's fine right up to the edge of your own understanding, I would say, but past that point, it's not really saving your time. I think it's kind of, you're just outsourcing the risk or, you know, you're not really, you don't have that same ownership. So I think, yeah, there's definitely lots of new ones around how we're actually using it and what that means practically. So the way I, I often think about this is, and you say you might, you might resonate with this, right? Like in the GTM world or in the business building world, we have strategy and we have execution. And I think where we are at is kind of counterintuitive because though we use AI a lot for the research, for the analysis, for the comparisons, for the explanations and so on, what we are letting it do, and this is true across most of the consumer space, it's also interestingly true across the enterprise space to a large extent. What we are allowing AI to do, and in my opinion, thankfully so, is that small time execution. Like for example, top up the saving for me, or run, build your own agent who will run a search and optimize your current account across where the best offers are or the best reads are or the best cashbacks are and so on. All of that is fine. But where it comes to that long term strategy, there are very, very few tools or banks who are letting that be automated by AI. So it's almost like, yes, help me take the decision, but I still want to be the person pressing the button to make that happen. And even at a larger institutional stash enterprise scale, this is being very much incorporated in as part of the workflow, as part of the standard interaction. You might have an AI telling a relationship manager that here is the recommendation you should give this customer, but whether or not you actually go and redirect their money based on that recommendation is still something that the customer and the relationship manager will need to actively initiate. So I think that's something which is still happening in the AI world and the other thing which this sort of boils down to, especially when you start thinking of that decisioning is a lot of the decisions we take, especially with respect to finances and kind of high stakes finances are very multi-pronged. I might have all of my bank statements in front of me. I might have all of my like investments in returns and fund prospectuses and all of that in front of me, but the decision that I take will not only be influenced by that data, but it will be influenced by that context layer of saying, well, actually, I don't want to set up my own company into years time and I know I will go on maternity leave six months hence and I am really worried that maybe I don't know, maybe my house is going to have a leak which I'm not going to be able to fix. None of that data at the moment is consolidated in one place for any kind of technology or any system to be able to act and recommend in a manner that those decisions will be a whole decision. It's almost like the whole person. They're not going to be able to do that because infrastructurally and architecturally that single view of customer does not exist. And I think that's the other huge obstacle in AI actually automating any of this because it just doesn't have that information to automate it. So I think that's the other blocker for AI to be able to do more and more in terms of not just taking those decisions but more importantly taking those actions. Yeah, because as you say, I think is such a large proportion of that full picture is quite personal, isn't it? An even individual appetites to risk and all of that sort of stuff can be difficult. Difficult to gauge and people can say one thing about, oh, this roughly this comfort level with risk but then actually in practice, it's quite different and all of that sort of stuff. And I think this has given us a really useful sort of grounding in terms of where we're out today in terms of how people are using AI around managing their personal finances and actually that probably a lot of what people are using it for at the moment is relatively low stakes. You know, automating some roundups or moving money over into a pot. And I think that's super, super useful. I think we're going to take a very, very quick break now and then when we come back, we'll jump into, you know, I suppose what we think the direction of travel is and the future of sort of where this builds from here. So please don't go anywhere and we'll be back with you very shortly. Welcome back to the show. Now in the first half, we asked whether we've put too much emphasis on educating consumers to make better financial decisions themselves. In the second half of the show, we want to take that idea one step further, you know, what happens when AI doesn't just educate or guide us but starts deciding for us. And I'm going to refer back to the the Mills report. I think the sort of opening quote in the foreword was that AI offers a once-in-a-generation chance to close the information asymmetries and frictions that have long left people making core financial decisions. I think in the previous half of the show it was a bit of a reality check maybe around where we're actually at today but I'll ask you both maybe a detailed start with you. Do you guys sort of agree with that as the sort of the promise, the potential of where AI can start to move us towards and I guess what needs to happen together is from where we are now to start to realise some of that potential. I'm kind of dawn on this one, Ross. So the way I see it is from a technical capability perspective, yes absolutely. We will get to a point where the models are so intelligent and they can correlate so many different things and they can extrapolate so intelligently that all of this discussion that you had in the first half around knowledge and information and interpretation etc will be covered. The big reservation that I have this this though is that the problem lies not in the AI capabilities or the AI literacy but it starts way before that in terms of again the underlying data, the underlying information and what actually feeds into that AI but the second big chunk here which I don't think we've touched upon since you see mentioned it right in the beginning is the intention and the trust. So for example I'm sitting here in India in my parents house at the moment and I have some savings in India which my dad bless him has been managing for the last 20 years or so. I don't even know what they are, I don't know what he's doing with it but what he has given me is complete trust that whatever he's doing he's doing with the right intentions and he's doing it with full attention like he will do what it takes to find out what he needs to find out and do whatever is required to make sure that the outcomes from that other right outcomes from me. For any individual to get to that state with an AI model we're a really really long way off from that and I don't even know if that's ever going to happen because at the end of the day these are all large language models run by third party corporations who have their own sets of legal checks and balances and nuances and privacy considerations and all of those other things. So I think that intention and attention gap which contributes to that trust is not something that AI will easily be able to overcome. Please see what do you think? I mean I like that separation a duty that you set out around the difference between the sophistication and the capabilities of the models but then the trust in almost like the human factor that almost being the sort of check the checks and balances on how far this actually how far this actually gets and how quickly. Yeah I agree I think you know it doesn't surprise me that people say that they're ready for this because in lots of ways you know who doesn't want to get rid of some of this like burden that's living in your head of financial decisions you've got to sort out that you don't have the time to do or who doesn't want to be making loads of money from some investment that an AI could do without you having to spend hours and hours of research and worry but I think the question for me is more like our products actually meeting that appetite are they able to I guess that's kind of what you're saying and also are they being like designed responsibly in terms of how they governed and and I think you know when we think about it so we are in the process of becoming an appointed representative of a principle firm so that you know they will have oversight of our regulation and governance and how that works traditionally is you know spot checks on how an advisor to this advice looking at outcomes you know but what does that look like from an AI point of view I mean we have designed RAI to only deliver guidance she doesn't stray into advice when a question you know touches advice we pass over to a human advisor but if what we're talking about is us moving in that direction where AI eventually would take on more of the actual advice how is that governed and you know and because that will be so important to people's trust I mean you've kind of I mean obviously your dad the level of trust is beyond what you'd ever feel towards probably an external party but I think people like the fact that financial advice bank is regulated that you've got that kind of layout of even if you don't understand what that looks like practical what it means it's a regulated governed industry where people are you know have to adhere to certain standards and I think we are in this completely unknown territory with you know AI becoming more and more prevalent in people's financial decision making and you know like we said we're maybe not completely there yet but what does that governance look like and then how is that peace handled I think is really really a very important question you know at the moment that you know the direction of travels heading that way a little bit so the FCA launch targeted support back in April which let's firms make group level suggestions without running a full advice process and simplified advice is the next stage which I think is currently still out for consultation but that is aimed at cutting this admin burden so that advisors can serve you know simpler customer needs much faster and I think that's the sort of way that we're moving in terms of AI actually being able to execute things and I think that aligns with what we said of it being simple decisions make me maybe that is the first phase so nice is my event and the first phase is this lower state stuff but yet how we actually tackle the bigger you know at this point I still see that being very much a humans role and I hope that AI is just going to provide the automation for all the you know all the sort of lower level admin pieces that makes an advisor's capacity so limited so that actually we do improve people's financial outcomes and wellness but it's not necessarily because they suddenly all use AI to make all these decisions for them but they're actually a lot of this prep bit that we talked about before is being done by AI and then a human has more time to have a relationship with the customer and fully understand their picture and help them make those more complex decisions with the the detail of what their life actually looks like right now and what their goals are and their values are and you know as you mentioned there's so many extra bits that are beyond just the numbers that it can read from your paperwork I think to be a little flippant about it that in itself is a huge uptick right like I mean if you go back to the cliche of time as money anything which saves people the time of either doing the research or doing those tedious manual mundane kind of admin stuff it's great it's useful but I think the problem which you highlighted Lucy it's I almost think it's a broader AI and financial services problem as opposed to a pure advice only problem and in a way there are two aspects of this one is that this is the one area where I feel really really glad to be living in the UK or in Europe because our regulatory take on safety of AI while it might be a bit frustrating for early state startups and people building in that space I think it is definitely along the right lines for safeguarding of people on the street and the people that that AI is likely to have an impact on so everything around consumer duty or the UI I act etc at least it is putting some guardrails in place so that people don't have the adverse effects impact them without their consent for what it's worth but the second piece of that is more the problem around accountability and liability at the very basic level if you ask charity pd a question it gives you a wrong answer and then you go and make an investment based on charity pd's answer who's responsible is it you is a charity pd we don't have an answer to that that's a much wider scale problem of where does the responsibility and accountability lie the good thing is there are some frameworks in finance where this is already starting to be solved so for example if we think of the payments frameworks and the fact that American Express has come in and said that we will honor a transaction which was done by an agent but authorized by a human being even though it was a wrong one like people are trying starting to figure out frameworks and fallbacks that can set in place when AI is actively involved in the action around a financial transaction and I think where it comes to advice again that is two parts it's whether or not what was told to be done was the right thing and then secondly was it done the right way so maybe at the point that the decision was taken or the advice was given by an AI that was the right investment but three months down the line something else happened how do you resolve that and how do you solve for that so I think all of those questions are still very much open questions but we are starting to see early shoots of analogies and parallels in other part of financial services world which will hopefully then fall over into financial advisory and investments and so on so yeah time to watch the space with more attention I think and because that's obviously one really important aspect of the regulation is around that accountability and that recourse as you say but then I suppose another important aspect is that you know you sort of mentioned around the consumer protections from a sort of fraud and cybersecurity risk as well and you know AI really has the potential to accelerate some of those risks I mean I think it makes potentially makes attacks faster and more scalable and more persuasive and you know I think it's regulation is going to be a really important determining factor around and how far we get with the rollout and adoption of AI and financial services as it always is and but it's important not necessarily just to see that as a headwind you know it is a really important role to play in actually protecting consumers as we sort of move through this process. Oh my goodness the field of fraud monitoring and overall financial crime monitoring has just just blastered off after AI. And it is very much a principle of set a thief to catch a thief. Just like you have AI enabling fraudsters to be able to carry out much faster, much more frequent, much more intense attacks, you then have AI tools. And there's so many of them. Even at Snowflake, we work with our customers to kind of safeguard the data and apply the models in such a way that that monitoring can take place runtime across humongous quantities of data. I think that is definitely something that AI is helping in massively. But I think the bigger problem here comes in where it is more a question of figuring out what the right thing to do is. Again, to the previous discussion that we were having in terms of that accountability and recourse, how do you even know whether what you've done is right? And you might have no malicious intent, you might have no desire to actually do something wrong. But more out of lack of knowledge and information, than out of ulterior motives, something ended up going wrong. In my opinion, that at the moment is a much larger risk, especially across automated AI advisory. And the reason I say that is we do a lot of work with banks who are kind of trying to get a 360 view of a customer or to personalize their products and offerings for customers precisely to lead to that financial wellness. The big challenge with that is anytime a bank says they have a 360 view of the customer, it is basically 360 degree view of a customer as far as their lending goes. Or 360 degree view of a customer as far as their pension score. McKenzie has said that on an average, a bank has at least 10 ledger systems, which by definition means that any information you have on your customer and any kind of data that you hold on your clients, which you are then basing your recommendations and your advice on is fragmented. And that siloing of data leads then to limited visibility obviously, but then also to advise being generated on a very limited view. And it's almost like when we think of wellness, it is very, very similar to physical wellness and medical wellness, right? The fact that my hypertension tablet is actually causing my eyes to dry up, unless you are a really, really informed and accomplished and experienced doctor, you are not going to say that the reasons your eyes are so eyes maybe you need different hypertension tablet because these are two completely separate practices and specialities, much like what is happening in the banking world with customer data. And therefore the recommendations that are being made, the advice which is being given, even if you are running AI till the cows come home, unless you're pulling it all together, your AI is not giving the right advice. So I think that is a bigger risk at the moment than genuinely fraudulent slash malicious attacks using AI. - I completely agree. And it's classic, isn't it? Because we can sit here and we can sort of fantasize about the future of AI for hours, but the reality is what's going to hold certainly sort of legacy institution back. So the same issues that have held them back with digital and various other transformation programs down through the years, which is the siloing of all of their information. And I mean, that only gets amplified in an AI context because the information or the recommendations are supposed in an agentic context, the actions that come out of that, they're only going to be as good as the input in terms of the data itself. And I'm so sad to say this. I think we are rapidly running out of time. Lucy, one thing that I did want to just reflect on before we finish is the role that AI can play in a sort of financial inclusion context. And I think we sort of started off saying that, you know, potentially it has a role to play in plugging some of the gaps that we've seen across financial services. But is there also potentially a risk that it almost exacerbates the problem? You know, I think it's like people who have a reasonable degree of financial literacy and financial understanding that gets sort of compounded and accelerated other people get left behind, people can access maybe better models other people can. Is there a risk there as well? Yes, I think to both, I mean, that's why it's a difficult one because in lots of ways, and this is exactly what we're trying to do at wellness is use AI to make financial advice more accessible and more inclusive. And, you know, traditionally, advises typically require, you know, 250K in investable assets before you can even get through the door and they often charge a 2000 pound off from fee just to get started. And the reason that they set the bar there is capacity which we touched on already. So, you know, it doesn't make commercial sense for them to serve this mid market. And AI is what can change that equation. So how exciting is that that we can, you know, increase capacity get more people access to, you know, financial advice where they do have this whole picture of what your life is and they can give you really nuanced advice. And also, I think we did talk about this a bit but, you know, get people more advice ready because I think that is the other problem is that most people like what, you know, financial advice isn't for me or that, you know, investing isn't for me. And actually, you know, if you've got a mortgage or a pension, you are investing. So I think there's, you know, it's actually for most people, but the barrier is often the affordability and the fact that it doesn't feel accessible. So I think it's, you know, really exciting in that sense that it can, you know, democratize access to a degree. But, you know, you're absolutely right that if we're looking at, you know, people with lower financial literacy, you know, who aren't using these tools already, I think actually they're sort of falling further, you know, the gap is widening even more. And the risk of kind of poor decision making or, you know, yeah, just sort of lack of understanding of if it's, you know, where do they even go? What do you know, AIs and for me is actually a real true fact for lots of people. So I think it does leave some people more vulnerable to poor outcomes. So there's this weird kind of, you know, disconnect, I think, between opening it up for some and actually maybe widening the gap for others. And, you know, I mean, when we think about it, we want to keep the guidance lay free forever up so that, you know, as many people as possible can get access to good information that it, you know, and hopefully that helps in some way. But I think, you know, you've also got to, you've got to have access to that access. If that makes sense. So, I mean, and that's a much, much, you know, bigger problem than we can probably solve or discuss right now. But, and that's kind of how I see it. - I really love that point though about being advice ready. And that's absolutely spot on. I think investment is like such a good example. I mean, we've spent decades in this country, sort of, you know, perfecting all of the small print and all of the very cautious language around investment and the value of your investment. I go down as well as up and it feels like in most people psyche now they sit alongside sort of like crypto and gambling and that sort of stuff. And obviously that's not necessarily the case. So, there's also a mindset switch that needs to happen there as well. And I think that's been such a recurring theme throughout this conversation is like, well, could AI do this or enable this or unlock this? Well, yes, but we've also got to look at these things in the round and consider all of the other sort of external factors. And, Aditi, I'm going to give you the final word. And we talked at the top of the show, you talked about sort of outcomes and AI in the context of the outcomes that it can deliver both for institutions and also those end users. Are you bullish or are you confident that as an industry, we can harness all of those different things and sort of bring them together to deliver better outcomes over time for all of these different parties? - I very much am. There is no doubt about the fact that I am bullish and I say that for two reasons. One is organizations and more importantly, financial services institutions have a duty to play the role here of enabling that access to access, as Lucy said. So, if I am an individual and I do not have the ability or the motivation to ask that question directly to cloud myself, my bank should somehow find a way of doing it. If I am an individual who does not even have a bank or have access to a bank, as an industry, it is our duty to make sure that those who are outside of the traditional systems are somehow pulled in. And AI is very much that vehicle, which is going to reduce cost to serve and enable more people to come within the realm of financial services, whether that is traditional, whether that is challenger, whether that is through different modes of delivery. So, the way I see it is, I'm definitely bullish. AI is definitely the vehicle. And there are definitely signs that, financial services institutions are changing their thinking towards banking because AI is enabling them to do so. So exactly to lose these points, thresholds are being lowered, openness is increasing, alternative ways of using data, taking decisions, tailoring outcomes for better consumer results is definitely happening across the industry. And that makes me, that makes me really, really happy to see it. And it's almost in a way, we just have that slight marginal lead on the right intentions versus the really difficult obstacles. And that's what makes me happy. Honestly, I could not think of a better way to sort of wrap up this conversation that was awesome, it was like a call to arms. And I'm here for that all day long. But yeah, so that does wrap up today's discussion. And thank you so much to our guest for what was such a fascinating conversation. Before we go, would you guys just tell us a little bit more about where we can find out more about you guys, find out a little bit more about your companies. Lucy, let's start with you. Yes. And so you can find out more about us at wellness.ai, which is W-E-L-L-E-M-E-S-X. And you can find me on LinkedIn. Awesome. Thank you, Lucy. Aditi, how about you? So, please go to Snowflake.com to find out more about what Snowflake is doing in the world of data and AI. And there's lots of different case studies and customer stories also, which might serve as some inspiration in the space. And please find me on LinkedIn. I'd love to chat more. Excellent. All right. And you can find me as ever over on LinkedIn. And you can find out a little bit more about 11FS at 11FS.com. And thank you very much for listening. If you like what you've heard, please do follow our podcast. And as always, if you want to join the conversation, find us on social media. Just search for 11FS or FinTech Insider or email podcasts at 11FS.com. Thank you very much again. And goodbye.