AI, Regulation, and Trust: Using Data Science to Protect Consumers at Scale — Edmund Towers, Head of Advanced Analytics & Data Science Units at the FCA

0:00Edmund Towers:These are really high-impact personal issues, right? And it can be easy to then abstract this up to sort of, you know, AML, right? And there's three-letter acronym, but really it's absolutely fundamental that we use technology in ways that can help fight some of these really serious issues.

0:19Dr. Raoul-Gabriel Urma:Welcome to Data & AI Mastery, the podcast where we bring you cutting-edge insights, practical advice, and inspiring stories from the leaders shaping the future of data and AI across the globe. I'm your host, Raoul Gabriel-Urma, founder of Cambridge Spark, the leader in transformational data and AI upskilling, career development, and progression. In each episode, I will be diving into real-world case studies of companies harnessing the power of AI to drive innovation, reduce costs, and create new business opportunities. So whether you are an aspiring data scientist, AI engineer, or seasoned executive, this show is designed to give you the tools and knowledge to stay ahead in a world where data is transforming every aspect of business.

1:04Dr. Raoul-Gabriel Urma:Stay ahead, stay inspired, stay masterful. Welcome to Data and AI Mastery.

1:12Edmund Towers:hello good to see you ed how are you i'm very good thanks raul yeah looking forward to this conversation so just to kick yourself so you're the head of advanced analytics and data science

1:28Dr. Raoul-Gabriel Urma:units at the fca a very exciting role so tell us a bit what does a what does a day look like

1:36Edmund Towers:well well you know my my daughter always asks me that question she's like why what are you doing today yeah and the easiest thing is for me to say is i'm spending a lot of time in meetings but but that doesn't really uh bring it to life so i mean you know my role has kind of two elements to it at the moment the first part is about helping use data science analytics techniques right across the FCA to help us be a more effective regulator. And that really means making sure that consumers using financial services products are protected, making sure that markets are clean, making sure that the UK is a great place to own and run a competitive business.

2:20Edmund Towers:And, you know, I get to do a whole range of different things, from projects that help identify financial scams and protect consumers from those scams, to looking at how we can test and assure that firms have the right sanctions controls in place so that their money thrown through the UK financial system is clean and free from that to creating new synthetic data sets to help detect money laundering and all of that in the financial crisis space to then looking at things such as what's happening in the credit markets, right? Do consumers have access to the right financial credit products, credit cards, mortgages?

3:00Edmund Towers:How do different consumers move through their own financial journey? What can we understand from that? Where can we see differences between firms? Where can we spot issues that might be of concern? So one side is all kind of looking at that and really how we deliver these existing FCA strategic priorities. The other side is about the kind of what the future and the what's next, not just for ourselves, but for financial services. So it's everything AI. I'm sure we're going to spend a lot of time today talking about that. But that's a combination of how do we use AI internally, but also what's our regulatory approach to the use of AI in financial services?

3:41Edmund Towers:And how do we support firms in their own goals around AI use?

3:45Dr. Raoul-Gabriel Urma:Great. Well, incredibly intellectually stimulating. Lots of really interesting use cases to look at and very impactful for the industry.

3:59Dr. Raoul-Gabriel Urma:So I'd love to kick us off maybe with the second part first, which is around, you know, AI in the external world in the industry. Like, how do you think about the role of the FCA in the world of AI? own.

4:13Edmund Towers:So financial services contributes a massive amount to the UK economy, about 10 % of GDP, which is roughly between like£200 and£300 billion a year. It employs over a million people. And technology and AI is absolutely at the heart of financial services. Whether you're talking about a high street bank or a big multinational wholesale firm, many of whom consider themselves technology firms first these days, the UK is already kind of a world leader in the use of technology in finance or what's often called fintech. Whether that be a small startup firm who's looking to attract investment into the UK and into their business, or whether it's a big incumbent incumbent firm like an institutional firm using technology to deliver better products and better services to their consumers so that's the kind of context right and then ai is the like technology innovation of the next decade at least we are already seeing the massive potential of that and and there's a huge opportunity there for the uk to lead the way in the use of uh safe and responsible AI in financial services.

5:34Edmund Towers:So as a regulator, that's what it all comes back to. We have to create the right regulatory environment that supports that, that helps firms innovate, that helps firms deliver new products that can better serve those consumers. And we want those firms to build their AI here in the UK, right, and then to export it to the rest of the world. You know, the flip of that, going back to the second part of my job title, is that if we want to create that environment for firms and for financial services, then we should absolutely be embracing it ourselves, using that technology, you know, to make the most use of our resources, making sure that our people spend their time and their expertise on the things that really matter.

6:20Edmund Towers:you know the fca's goal itself is is not just to be a leading regulator but our goal should be to be a leader amongst financial services as well that's great and uh there isn't it's a lot i

6:33Dr. Raoul-Gabriel Urma:really believe in uh you got to practice what you preach uh so to to encourage yeah you got to use it yourself so i'm 100 in with you and i do agree with the opportunity available in front of us in the industry. So I might ask you an unpopular question on purpose, but when we think about regulation and regulatory, there tends to be a view that it's slowing down innovation, it's putting roadblockers and so on. Obviously, that's not the case, right? So I'd love for you to talk us through what's the balance between accelerating innovation, yet, like you said, doing it responsibly. How do you think about this kind of trade-off?

7:15Edmund Towers:Yeah, I mean, it's a great question. And actually, you know, glad you asked that because I think, well, the most important, certain types of regulation can be, like, let's start with that. Certain types of regulation can cause problems for innovation. I think the really important thing is that in the UK, we have this approach for financial services regulation, which is outcome and principle focused, right? So we have these regulations like, for example, the new consumer duty, which says firms must give consumers the information they need to make decisions or that the product and service must represent fair value to the consumer.

7:56Edmund Towers:So we focus on that high level outcome, but not on how that's delivered. Right. So firms can innovate in how they deliver those goals. You say the goal is clear, but how you get to that goal, there's a lot of freedom to innovate.

8:14Dr. Raoul-Gabriel Urma:Great. So can I take you into maybe the more external facing work that the FCA conducts? So maybe can you walk us through what are sort of the key strategic themes when it comes to AI that, you know, the FCA is really involved in?

8:30Edmund Towers:That is why we've launched this, I think, world-leading, honestly, initiative called AI Live Testing. It's a voluntary service, right? First of all, we are not going around and insisting that every firm wanting to deploy AI has to go through this process. It's also very much focused on the really significant use cases. So back to the points we're talking about, these kind of reimagining use cases, where the real impact is, you don't really need to come and have a conversation with the SA if you're planning to roll out you know co-pilot across the organization but if you're planning to change how you're interacting with your consumers right and you know how that then maybe you might want to you know you don't have to come to it but you might want a little bit of additional engagement with us and so what we do is you know we take we opened it up to firms like at the end of last year we had over 50 applicants we selected eight from that and there are a range of firms around so it's from the you know the smaller kind of fintech firms through to some of the bigger institutions so firms like NatWest, Lloyds, Monzo, Santander like you know also taking part in this and we take a use case that they have where they are not just doing a proof of concept because I think there's you know lots of people can do the proof of concepts but it's about how we deploy those things properly you know and safely and responsibly so we take a use case we think about what are the potential regulations and the potential risks around that.

9:58Edmund Towers:And then we work through some of the technical details. So this is like not just regulator talking to compliance official in the firm, it's like technical teams in the firms engaging with my team and my technical team. And in fact, we're working with another partner organization called Advi, who are kind of specialists in AI testing to kind of then go through, right, well, how do you, you know, what's the system you build around the model what are the what are the guardrails what are the you know the prompt engineering and prompt management approaches what are the you know what data is going into the into the model from your side what's the the rag pattern that you you know how do you kind of and then where is the sort of like human oversight and where's the monitoring and the tracking you know that potentially goes alongside this and and mentioning lots of things there not to say that all of those things that you know are required in every single situation right but you start to kind of build a richer understanding of this and that something is ready to use.

10:55Edmund Towers:And then we let firms test in real life, right, with an element of reassurance that if something was to go slightly wrong, they have to put back that problem, but we're not going to hit them with a big kind of fine or enforcement action, right? So it's designed to build that maturity and then confidence towards deployment so that these solutions can start to have real kind of beneficial effects in industry and move from poc into you know production great yeah sounds like a fantastic

11:29Dr. Raoul-Gabriel Urma:initiative to de-risk uh and accelerate like a safe adoption but also sounds like it's a great way to build collaboration across different partners in industry no doubt there's a lot of lessons that can be shared to the wider community so i think that's super cool i hope you're enjoying today's conversation if you're finding the insights useful please do take Take a moment to subscribe to the Data and AI Mastery podcast and leave us a review on Apple Podcasts, Spotify or YouTube. Every new follow helps us reach more people and shed incredible work being done by today's Data and AI leader. All right, let's go back to the episode.

12:06Dr. Raoul-Gabriel Urma:Before I call it, I was kind of reading up on some of the strategic teams that, you know, I guess are current, right? So financial crime, consumer protection, becoming a smarter regulator. Can you talk us through why these three and why these are kind of the priorities?

12:21Edmund Towers:So there are technically four priorities for the FCA. They are our new strategy set for the next five years. Financial crime, helping consumers, smart regulator and innovation and growth. And they apply to pretty much everything we do, not just to kind of AI. But, you know, we've talked about a few of those themes already. So innovation growth is about how we help firms safely adopt AI, how we help UK firms be a leader in that global market, right? And ensure that we are kind of able to adopt here and kind of hopefully export to the rest of the world. So that's kind of one element. The financial crime, I guess, is some of those use cases we talked about.

13:09Edmund Towers:I mean, most firms are using some form of analytics. Pretty much every firm is using analytics or ML or in some situations, you know, on some of these problems like fraud, scams, money laundering detection. So, you know, we are also doing that internally ourselves. We do our own work proactively trying to identify fraudulent financial promotions using web scraping and then take action against those via the ISP. but at the same time we're also doing this work where you know we know that uh there are sort of more complex ai driven approaches to money laundering that potentially can reduce the improved detection rates and reduce false detection rates you know that also cost firms on it and these but these are real problems i mean you know i think it sometimes you when you talk in the abstracts well like these things sound a bit dry especially for people who aren't there, right?

14:04Edmund Towers:But money laundering is about, you know, that's the money that finances terrorism, it's the money that finances, you know, human trafficking, that finances the global drug trade. These are serious, like, these are really high impact personal issues, right? And it can be easy to then abstract this up to sort of, you know, AML, right? And there's three letter acronym, but really, it's absolutely fundamental that we use technology in ways that can help fight some of these really serious issues it's the same for scams i mean you know you i think all of us either know someone personally who's been subject to some kind of scam or i've read about it in the in the newspapers right this is these are these are real impacting issues so the more that we can do to help identify these kind of things take action on it fraud you know as well like push payment fraud it's really it's really fundamental and technology has a massive role to play in that The industry is already using it.

15:00Edmund Towers:We are using it, but we're kind of committed to kind of pushing and working with firms to help them move further, which is why, you know, one of the things we've done is generate this synthetic transaction data set using data from one of the major firms in the UK and then working very closely with the Turing Institute, actually, and to develop this new data set that we're looking to make available to firms who want to innovate with this in the new year. um so sorry i i got stuck talking on financial companies i think it's such an important issue but then you know the helping consumers point is you know we all as consumers want to make the most use of our money financial services like is a really important way for people to achieve their life goals right so whether that is you know buying uh buying your first house or saving for retirement or being able to afford that home improvement all that time so you know having access to financial services that can help you achieve those life goals and do it in a way that it's safe and trustworthy and usable is really actually fundamental to how most of us you know achieve our life goals so using data analysis in that area to to understand different consumers how they move through these journeys what products they have access to are there any gaps are there ways in which consumers could be better utilizing those kind of things other ways and for you know are there any areas we want firms to be kind of improving how they operate?

16:26Edmund Towers:Like these are really, again, really, really important kind of topics.

16:30Dr. Raoul-Gabriel Urma:Hey, that's hugely exciting. And it's great that, you know, AI offers such an opportunity to scale the impact of those initiatives and make it more sophisticated and do good for society. I mean, at the end of the day, protecting consumers is like vital. So it's great to hear that can do even more. So I'm excited about it. And I guess on this note, you know, if we kind of like fast forward a couple of years into, you know, I appreciate the industry is moving fast. But what do you see are some of the biggest opportunity that AI presents for consumer outcomes, like in particular?

17:07Edmund Towers:I think one of them is helping consumers get the right product or service. So, I mean, you know, we talked earlier about the kind of agentic opportunity. We are not quite there yet, right? But that world where I can have a digital twin agent of edge, right, basically of myself or you, you, Rahul, you know, next time you want to take out an investment product, right, or you want to move your business bank account or whatever, you know, you can ask, you know, an agent to do that on your behalf to go off and do it like research all of the different offers, products and services, you know, ask you.

17:51Edmund Towers:like, well, OK, well, what are you most looking for? Like, you know, if you're getting car insurance, do you care about whether there's windscreen protection? Are you going to drive it like abroad? Do you like park it on, you know, but and then remembers all of that detail. Right. So so it goes on and finds that gives you some offers. Maybe you kind of look at the options and you say, yeah, go and purchase that for me. And it goes and does it right. And then a year later, when it comes to that time and you've got to you don't have to remember oh yeah you know it's the 15th of december it's time for me to renew my current you know your agent is like oh well by the way it's renewal time i've looked i've found these things just want to check that these are still your requirements this looks like the best deal right you know i mean that kind of world across multiple different types of products i think is really going to be valuable for people and you know that conversation around at its heart that's a really complex research problem right it's like i have to go i have to look at these things even if i'm using a comparison website i still have to click into lots of documents make sure i understand the specific policy terms and conditions etc etc but if i've got someone who there is kind of prompting me and sort of helping me guide me through the process reducing some of all that manual legwork then i think it's uh super valuable oh such a great

19:10Dr. Raoul-Gabriel Urma:to answer it and uh i'm fully botting on this uh digital twin ed version of digital twin

19:16Edmund Towers:well you're going to have your digital well as well so you know you've got a piece of the everyone's going to end up with that i think is my is my you know long-term dangerous to make long-term prediction but i think that's possibly the way we're going to end up i don't know my

19:29Dr. Raoul-Gabriel Urma:mom would feel about it but i can see the uh you know it's always good to be able to to simulate decision actions and so on and it you made me think that if that's the case then it's likely there'll be an increased importance on concepts like memory for agents and data storage and data infrastructure so it might be a full loop whereby we need to think harder about the data we store how we utilize it to to make you know the the personal personalized advice and guidance even more effective hey Ed can I take you to a quick fire round of questions you can yeah go on you know everybody is is wondering how to stay you know up to date as a leader right like wherever whichever role you have whichever business functions you're in you know there's a lot of progress and there's new releases every week so what do you do yourself at to like stay up to date and what sort of guidance would you give others?

20:27Edmund Towers:Well, you know, I mean, it's the usual kind of things of like podcasts like this. But actually, one of the cool things, we have this group that we just sort of accidentally formed in the MTA. It's called, like, it's just a Teams chat and it was called Moonshots originally because we were trying to think of big ideas. And then we just got, you know, there's a group of people on there who are always like sharing, oh, I've just seen this new article, this new release, like the race list. And it's such a nice informal way of people to share. So I think that kind of thing of like having a few people that you are operating in the same area with, you just kind of share like those articles on a sort of like, you know, Slack channel type of format or whatever.

21:06Edmund Towers:Now is really is really cool, I think, because it always makes sure the things are just like with a slight tuning to your specific situation as well. Great.

21:16Dr. Raoul-Gabriel Urma:Love that. And a couple of more personal questions now. First one. What's your favorite programming language? oh oh python absolutely nice phew we can uh we can we can stay friends it's as good as good as word uh what was your favorite subject at school oh i don't i mean you know i i don't have

21:36Edmund Towers:an answer that i always like being a bit of a polymath if i'm honest like so you know when i was at school i did a whole mixture of like arts and different science uh like subjects from doing like physics and biology as long alongside our politics and history at the same time so it's like you know i don't i'm fascinated by interdisciplinary stuff right there i think the the point between subjects and where you bring those different disciplines together is what

22:04Dr. Raoul-Gabriel Urma:really interests me so humanity you know yeah sounds great a bit of everything but like the

22:11Edmund Towers:intersection that the how do you like yeah the play between like physics and history or politics So, you know, that kind of point is really what is interesting, I think, like bringing those disciplines together.