
Trading Tomorrow - Navigating Trends in Capital Markets
Welcome to the fascinating world of 'Trading Tomorrow - Navigating Trends in Capital Markets,' where finance, cutting-edge technology, and foresight intersect. In each episode, we embark on a journey to unravel the latest trends propelling the finance industry into the future. Join us as we dissect how technological advancements and market trends unite, shaping the strategies that businesses, investors, and financial experts rely on.
From the inner workings of AI and ML to the transformative power of blockchain technology, our host, James Jockle of Numerix, will guide you through captivating conversations with visionaries who are not only observing the future but actively shaping it.
Trading Tomorrow - Navigating Trends in Capital Markets
AI Co-Pilots in Wealth Management
In this episode of Trading Tomorrow – Navigating Trends in Capital Markets, host Jim Jockle speaks with Elemi Atigolo, Co-founder and Managing Partner of Consult Venture Partners, also known as ConsultVP, about the emergence of AI co-pilots in financial services.
Atigolo outlines how AI co-pilots differ from general generative AI tools and describes where they are being integrated into advisory workflows, including meeting intelligence, research, and client reporting.
The conversation explores both opportunities and risks: how co-pilots may increase advisor productivity and client capacity, the implications for firms that delay adoption, and the potential for agentic AI to take on multi-step tasks.
Welcome to Trading Tomorrow Navigating Trends in Capital Markets the podcast where we deep dive into technologies reshaping the world of capital markets. I'm your host, jim Jockle, a veteran of the finance industry with a passion for the complexities of financial technologies and market trends. In each episode, we'll explore the cutting-edge trends, tools and strategies driving today's financial landscapes and paving the way for the future. With the finance industry at a pivotal point, influenced by groundbreaking innovations, it's more crucial than ever to understand how these technological advancements interact with market dynamics. In wealth management and financial services, a new class of tools is rapidly moving from buzzword to business reality AI co-pilots. Powered by generative AI, these intelligent assistants are being integrated directly into advisory workflows, helping advisors prepare for client meetings, generate portfolio insights in real time and even automate follow-up notes and actions. Firms are already deploying AI co-pilots that boost advisor productivity while offering clients more personalized experiences. At the same time, regulators and clients alike are emphasizing the importance of trust, compliance and the human touch.
Speaker 1:To help us unpack what AI co-pilots really mean for finance, we're joined by Elami Atagolo, co-founder and managing partner of Consult Venture Partners, also known as ConsultVP, a generative AI consultancy for financial services. Elami has nearly two decades of experience across AI, wealth management and business transformation. He's led the creation of Finley AI, one of the first conversational financial AI assistants, collaborated with the UK's FCA on compliant AI pilots and is recognized as a thought leader, featured by the Financial Times, the Economist and the BBC. With his rare blend of financial advisory experience at HSBC and St James Place and deep technical expertise as an AI developer, elamy brings a unique perspective on how AI can transform finance responsibly. Elamy, first and foremost, thank you so much for joining us today.
Speaker 2:Thank you. Thank you for having me. It's a pleasure.
Speaker 1:So let's start with the big picture. How do you define the AI co-pilot for finance and what sets it apart from the broader wave of generative AI tools that we hear about?
Speaker 2:Yes. So when we look at an AI co-pilot for finance, how I would actually define it is a AI assistant that has domain specialization or expertise specifically in finance, that acts to support or assist an advisor or financial professional, almost like a pilot or co-pilot in a plane. You have, say, the human that is in control, in command, and the AI looks to handle navigation, analytics, some of the heavy lifting. But the key part when we look at an AI co-pilot for finance, in comparison to, say, a generic AI tool, for example, is that an AI co-pilot for finance should be generally embedded or integrated within some of a firm's workflow, whether that might be their CRM, their data lake, for example. It could be embedded within some of their research data, portfolio information, even client data, obviously with consent, and it allows the human or the individual and the co-pilot to interact so that AI can access the information and provide the resources and results to the advisor, meaning the advisor can be much more efficient, much more productive, save a lot of time and hopefully, you know, increase their overall client book.
Speaker 1:So we're seeing major firms roll out AI co-pilots for wealth managers. It seems like you know if you're in the industry, there's an AI project going on, but you know from. You're in the industry, there's an AI project going on, but you know from your perspective what are the most exciting use cases that you see emerging right now.
Speaker 2:And, yeah, you know, one of the things that's really great to see that we are seeing this uptake Now. I remember some years ago speaking to wealth management firms and so forth about the use of AI. You know it took a bit of time. Some of the use cases that I'm seeing at the moment it generally surrounding areas such as meeting intelligence, so the ability for advisors, for natural professionals, to access information in real time, so meaning that within certain meetings they can transcribe, they can create reports, especially once they have, obviously, client consent, and really that saves them a lot of time. It also helps to reduce some of the risks that are involved so that when they're in meetings, they can ensure that the information is accurate.
Speaker 2:The other areas that we've seen is with respect to research, so the ability to research information, whether that might be analyzing you know, massive amounts of earning calls 10Ks, 10qs, for example being able to research information. As it happens, this usually takes hours and hours, so being able to decipher and synthesize massive amounts of large data and pull out insights that an advisor can use is also another key area. Probably the other area that we've started to see is with respect to client reporting, so the ability not just to create summaries but really to create full end-to-end workflows from the start to the finish. You know, those are some of the exciting areas that we've seen, and really some of the firms have taken things forward. Probably some of the key things that we look at is it's not just obviously about the features, it's also about the underlining and fundamentals, which are very important, and so when we're looking at these use cases, we have to ensure that to get the most long-term ROI is that you look at the fundamentals in play, and that's to do with transparency, consent and so forth.
Speaker 1:I want to ask you a couple of questions about Finley AI, but before I get into it, why don't you tell us what Finlay AI is?
Speaker 2:Yes, so Finlay AI is a financial AI agent as an API. So it's an agentic agent that allows the ability for firms also fintechs and financial firms to integrate a domain-specific AI agent for finance. So it allows them to build on top, allows them to automate various different tasks and it allows them to not just build our financial co-pilots but to also place that into applications, to place that into other tools and services they may use, whether internally or externally. And you know we spent quite a few years working with regulators and we built out a system outside of just an oval co-pilot. So it means that we understand what's happening at the far edge when it comes to finance.
Speaker 1:And what would you say? Through that experience of building and working with the regulators, you know, what did you learn about how to build a co-pilot that's both innovative as well as compliant?
Speaker 2:how to build a co-pilot that's both innovative as well as compliant. So when we look at, you know when we started building. You know we started back in 2019 and we started to use. We started with NLP, so natural language processing, and back then you could really provide the intents, you could provide the outputs, you could really direct the rule-based and AI at a time to ensure things were compliant. We started working with the regulators back in 2019. But then what we saw is when the generative AI came out we got access to it back in about 2021. And we noticed that it was fantastic in areas of being able to generate information, which was great, but because it wasn't deterministic, it was actually predictive. It meant that when you're trying to structure and control output and control information which you need generally for finance, that can be very difficult. So we had to spend a lot of time trying to fix some of the actual issues. We went from building a chatbot at that time to developing an agentic AI agent and what we found is, with building AI agents, there are actually a number of different layers that you need to build an AI agent, and so being able to work with the regulators from the very beginning it meant for us. We took a compliance firstfirst approach and a privacy-first approach, as opposed to compliance being on the latter end. What we also saw is, through the review and the work that we did with the financial regulators, and building out the AI agents and going through the different layers. So that's everything from your infrastructure layer, so your cloud layer, your foundation AI model layer, your agentic layer, your governance layer, security layer.
Speaker 2:We also discovered a layer that really hadn't been spoken about and this was actually what we call AI consent layer, and that was an AI that we discovered through the process. We discovered it more last year and we started advising our own clients, so our consultant clients. We started advising them about the implications where, if a firm doesn't get adequate consent so not just consent from clients but also consent from other B2B providers, so the information they may use, it starts to cause potential implication liabilities. And you know, we recently, you know, released some of that, those findings, publicly in a FT report that I had penned about the implications of AI consent. And why that's so important and why we learned that through the process is we've realized that as a firm and even as a technologist, you know you want to bring out innovation, you want to ensure that you know you have fantastic features and so forth, and although you may look at compliance often, you can miss the implications that arise only through the use of AI.
Speaker 2:So these are new unknowns, and one of the unknowns that we discovered was this new layer that is required where individuals, firms, companies actually need to obtain explicit AI consent.
Speaker 2:And when we disclosed some of this information recently in an article, we'd stated that this is probably one of the biggest areas that are not commonly being noticed by financial firms and also by even technologists. We said this could lead to, essentially, lawsuits and so forth, and we put this information out back in June and nonetheless, not more than two months after, we've seen class action lawsuit being taken against a very, very well-known in this case, ai transcription tool that's used heavily by financial firms, and in that case, it's a class action lawsuit in the state of California because it's alleged that the firm didn't get adequate AI consent. Now, although that's not in a financial situation, it's really more of a red flag and a wake-up call from the financial industry. So what we saw and what we learned is that compliance and also innovation doesn't have to be against each other. They can work hand-in-hand, and when you look at those areas, it can help to uncover risks that may pose themselves later on down the line. So those are some of the things that we discovered.
Speaker 1:Well, when it comes to wealth management, I think the key word we all think of is trust Trust with your advisor, trust with the institution. How are you making sure that these tools are trusted companions for the advisors and, ultimately, for the clients as well, and not just some sort of black?
Speaker 2:box. You know and that's really key Generally what we look at when we look at the key parts. Firstly, it's really about the industry being quite transparent about what's happening. So transparent about the use of actual AI, and so if you say to a client that I'm recording the meeting, you're actually noting to the client that I'm actually using an AI note taker, I'm actually transcribing your information, the information will be passed off to an AI system and you're able to get that opt-in at that point in time.
Speaker 2:The other side is being really transparent about different workflows that are happening. So if, for example, you're using AI inside a client portfolio, you want to be transparent about that position. You also want to be and understand what the workflows actually are so that you have a clearer picture, to have explainability, to understand what happens from the point of view of when we hit a button and the AI goes out and we get a response back in, what actually happens in those workflows. So what's really key is ensuring that you have transparency, you have the explainability, but you also have the knowledge, you have the education to understand what these processes are, because if you or an advisor or financial professional has a greater understanding of the use of AI. It makes it a lot easier to explain.
Speaker 2:You know we wouldn't expect you know, a financial professional dealing with a client portfolio not to understand how investments work, and that's really the same way that we see when we look at the use of AI and that helps to create broader trust and it also helps to have greater reassurance from the consumers, clients and so forth. That, as a firm, you understand and you look to work with what we call transparent and responsible AI. You take that as a first basis and you're quite public about that process. You're quite public about what you're doing to not just have the latest technology, but what you're doing to safeguard clients, to safeguard the processes that are actually in play, and you help to outline that and explain that as much as possible in plain English of course, in plain English.
Speaker 1:So let's talk about adoption a little bit here. As I look and talk, even across my own team or have conversations with individuals in the public, the use of AI tools and co-pilots varies greatly, from some individuals who are not using it all to some people who think it's hey, a really cool search engine, to people who are building businesses and a one-person shop who's taking over the universe. So, from your perspective as a wealth manager, as an AI developer, what are you seeing as the biggest opportunities for advisors who really embrace co-pilots?
Speaker 2:It's a really great opportunity for advisors that do embrace co-pilots, that embrace an AI. You know we're at a point now that the overall clients, their understanding and requirement for technology is increasing. You know, outside of finance, you know they're accessing tools via their smartphones, whether that might be your Alexa or your Google Gemini, for example, so they're interacting with AI and so, because of that, that is really pushing the need for professionals in the financial industry to really adopt the technology. Some of the overall benefits that we see is greater productivity, the ability to do things a lot faster, but also more accurate as well by being able to gather more information. It also means that they can distill and create a greater service. The ability also to provide more information in more of a real-time scenario. You know we're in a time of speed on demand, so the ability to wow your clients with being able to get back to them a lot quicker, you know is very, very vital and very important. You do have a spectrum, you know is very, very vital and very important. You do have a spectrum. You do, as you mentioned.
Speaker 2:You know some in the industry haven't yet got to that point where they see the full value, and that is really to do with an education adoption across the board. The more that they see this happening in the overall industry, the more they'll understand this change and this is all a change cycle. You know it will be almost unheard of at this point to not use email or to not use Excel, for example, but there was a time that those tools were also new, and that's really with the same of the use of AI. Once you understand what that use case can be for your particular firm to really get the ROI for you, you can then focus on that position to start to roll out. Well, how does this look for our clients? It's also important to involve clients that this is the change that's actually happening. This is what the service is going to do for you, and it also gives the opportunity to handle more clients because of the efficiencies that can be raised through the use of AI in that regard.
Speaker 1:And what do you think the risks are for those who resist? They don't want to use it. I've been doing this for 30 years. I know the markets. I'm better, I know my customers. What do you say to those curmudgeons?
Speaker 2:Yeah, so in the short term they may not see much change in the short term term and you know they may not see much change in the short term but really in the that, this, the that medium term, they'll start to see a shift and that shift will be with the competition.
Speaker 2:The shift will be with those firms and those advising firms that have actually taken up the mantle to to utilize ai within some of their practices, inside of their, their services.
Speaker 2:You know we've seen many servers, many reports, even investigators that we've done where you see really the true ROI by utilizing these tools. So the risk for these firms is that they get left behind, where you know this is all a client service position. Whether you're dealing with, you know front-facing in terms of your own client portfolio or you're dealing with the actual portfolios themselves, doing the investment side. You know being more efficient, being more productive, being able to do more with your time as obviously being the most valuable you know asset that you have being able to do more. You will start to see those other firms that are behind the curve. It will start to show, it will start to be more apparent and then what happens over time is that's where you start to see those, those gaps and then, as technology outside of finance develops and people's interaction with technology outside of finance develops, the usage develops and it then means that a firm can be seen as not being modern.
Speaker 2:It would probably be unheard of at this stage that you know, your, your largest wealth management or you know asset firms, for, in some regards, don't have a website that will be almost unheard of at this time. We'll get to a point where it will be seen. If you're not using AI, then what are you doing? How are you ensuring that you're keeping up with the trends? How are you ensuring that, for your clients and their portfolio, you're doing the best that you can in utilizing the best tools, not just from the investment side, but also in terms of the overall side for the business and for the organization, what you do in that regard, and then that can impact on your client portfolios. It can impact on their ability to really get larger clients in that space where you're missing out on those opportunities and those gains. Some of that output that you receive in investment is about your ability to utilize tools in the best way possible.
Speaker 1:Well, you know it's interesting. You talk about, you know the opt-in and we talk about, you know, individuals resisting. But I recently just went to the doctor and the doctor was like I have to record this visit. Do you mind if I use my phone? And you're like well, it's everywhere. So, to sound like a bad Star Trek line, resistance is futile. So we've talked about efficiency. We've talked about productivity. We've talked about efficiency. We've talked about productivity. How do you see AI co-pilots opening the door to maybe new business models or revenue opportunities for wealth management?
Speaker 2:Yeah, so when we look at new business and new revenue opportunities, you know one of the things is often it's efficiency that is catching the headlines when you look at the use of AI or co-pilots. But really some of the other parts is the ability to generate new revenue, the ability that you can reach out to more clients and handle more clients, but at high standard. That was almost next to impossible before, but with what we're seeing now and the abilities with a lot of these tools, you can actually do that. So you can actually go from maybe dealing with you know 100, 150 to say 200, even 300 clients at a time, but still provide them with the highest level of service. What that also means is you also have the ability to and you know open and expand to potentially other segments.
Speaker 2:So at points typically when we look at, obviously wealth management is dealing with you know high net worth, ultra high net worth and individuals, but there is still a lot of opportunity in those other segments.
Speaker 2:You know the mass affluent and even in some cases that the mass market now to try to handle those types of clients and and provide a good service one.
Speaker 2:It's very, very, you know, very, very difficult, almost next to impossible with the use of ai, use of ai, specific types of co-pilots that can be handled, and with ai, so that can mean that that can open up a whole new segment of opportunity that wasn't actually available and before. The other side that it can also do is when you start to utilize things like you know, ai insights through the co-pilots You've now got the ability to do, you know, benchmarking, so you could say you're actually a lot of clients, like you typically use estate planning that can open up a new spectrum, a new scheme which you may not have been able to focus on before, you may not have been able to deal with that type of information and you may not have actually been able to conduct that type of analysis, especially over a large pool of clientele. This can be done and so those are some of the opportunities where not only you can increase your client size, but you can also increase and open up new avenues of opportunity in other areas. So you can go upstream, scale up.
Speaker 1:You can also go downstream as well to a different segment pool as with any new technology, there's tons of conversations of AI is coming for my job, so what are your thoughts on the potential of AI in terms of replacing advisors, rather than having a human reach down into the mass markets? Maybe it's what is? What is your take on this human AI partnership model?
Speaker 2:Yeah. So my take is that it's, you know, human plus AI, as opposed to human versus AI. You know, I do feel that, you know, having the combination of humans and AI is the strongest fit, and the ability to work with an AI whether that's a co-pilot, whether that's an agent, for example, to handle different tasks, and the ability for the advisor to focus on what they do best, that is really key. Now, having said that, as AI does develop and improves, you start to see the capabilities of a genetic AI come through, the capabilities of, you know, a genetic ai come through. What I would expect to see is that the relationship between human and ai may change. You know, as clients start to get used to the use of ai, not just with inside a financial context, but just in general, you may find that ai may handle more of the um. You know, uh, quantitative um aspects. You know some of the planning, for example, some of the numbers and so on, and it may just mean that the advisors and professionals move up the value chain and they start working more in the empathy, the coaching and the things of that nature, the being able to deal with some of the complications when you look at certain types of situations planning and so forth which, where we are at the moment, you know AI is not in that position to handle those environments. I do think that what is key is that those advisors that utilize and know how to use AI appropriately will be the ones that will be replacing those advisory firms that do not know how to use AI or are not making use of AI. That will be apparent and that's probably one of the key things that I would say.
Speaker 2:So, from where we are right now, the AI human plus AI is really key. Making use and understand how AI works is also very key. You know, what we've seen in our consulting work is you know many firms that want to implement AI or have implemented AI. They are not always seeing some of the ROI because their team lacks the understanding, the training and education to make use of it, and that's really also quite key. Although you may have AI if you don't know how to fully use it, that's the implication. You know. To fully use it. That's the implication, you know. I always say it's almost like you know someone giving you the keys to a lamborghini but you don't know how to drive.
Speaker 1:You know you're not really using the power there are some great videos of of people getting into these, uh, sports cars and just like crashing them like that, like it's. It's not a. You know, your normal driver's license isn't covering the Lamborghini. So, as for firms that are thinking about adopting AI co-pilots, you know where should they start their journey. Well, you know what's the first step in terms of getting this correct.
Speaker 2:And yeah. So the first step is to understand and analyze what your actual priorities are. You know, one of the risks that where firms tend to stumble is that they see AI as the golden chalice and this will solve every single problem I have. Now, although that might be the case, you don't want to try to boil the ocean straight away. You want to take your time, identify within side of your firm, even within side of a workflow, where do you believe that ai will have the greatest um roi and you focus on on that point first. Yep, that's the first part.
Speaker 2:The second part is to get your data. You know ai, especially the former ai that we're in at the moment. Data is very, very key and having and clean data structured and data that's not necessarily in just silos, but organized data, is very important. Now, that is difficult to have just because the nature of financial services you know, we're around for a very long time and often you have siloed data. But our ideal scenario is to start to focus in having a particular data lake, because if you have the information, you've got a particular use case, you can start to work in that particular use case and that specific aspect of data.
Speaker 2:The other part is ensuring that you have the right training in play. So, before you integrate, implement, consider AI. It's really about, well, in terms of the teams, in terms of the individuals that will be using the AI, what's their understanding? What's their knowledge? What does that look like? What training do we need to implement? What training do we need to do before, after, during the process, to get the most out of this tool? Because, if not, what can happen? It can be an expensive exploration where you don't see the ROI, and there's been obviously different studies about the use of AI and whether or not you get the greatest amount of ROI. But a lot of that is to do with some of those key fundamentals have not been done.
Speaker 2:And, lastly, it's about the governance. So ensure that there are the right policies in play. So, look at what policies do you have? What policies need to be written? If we're looking at vendors, what do we need to ask? What questions do we need to know? What do we need to know about these structures in terms of where the data goes? What happens if we are using our own data that we have, that we've got licensed to, if that data goes out of our environment? What does that mean? Could that be a compliance issue? Could that be an IP issue? For example, is that a breach of contract if that gets sent to an AI model? These are all the things that you really need to think about to ensure that you have a very positive and really great outcome for your firm or so.
Speaker 1:Well, unfortunately we've made it to the final question of the podcast. We call it the trend drop. It's like a desert island question, and if you can only watch or track one trend in AI co-pilots over the next couple of years, what would that be?
Speaker 2:Okay, if there's one trend, I would say that it would be agentic AI.
Speaker 2:When you look at AI co-pilots, so at the moment we've seen co-pilots and we've been discussing co-pilots that really act based on instructions by a particular human or so forth, but what I would be tracking is the use of agentic AI where that, instead of it being just a, you know, an individual response by a human, we're seeing more that, you know, an individual may set a particular task structure and that task may require many different iterations and many different processes before you get to the end, but that one request may handle many different things, and that would be the area that I would be focusing on and that one trend, because that will one unlock massive amounts of productivity, efficiency and revenue opportunities through that that space, and it will really turn um and create a much more cohesive approach between the human and AI position, because it really means that you know professionals, financial professionals, advisors can really focus on certain specific areas and also tailor certain new skills in certain sectors, and I really allow the AI to work on various different tasks that can help to push productivity through the roof.
Speaker 2:So that would be the area, if I was on a desert island that I would want to focus on.
Speaker 1:Well, hopefully then the agents could build you a boat and get you off. I want to thank you so much for your time, your insights on the co-pilot topic. Thank you so much for your time, your insights on the co-pilot topic. It's amazing how everything is just continuing to develop so fast and perhaps we can have you back again and you can give us an update, because I'm sure the world will be different months from now.
Speaker 2:Thank you. Thank you for having me. It's been a real pleasure, thank you.
Speaker 1:Thanks so much for listening to today's episode and if you're enjoying trading tomorrow, navigating trends and capital markets, be sure to like, subscribe and share and we'll see you next time you.