Episode Transcript
Emily: Hello and welcome to FTSE Russell Convenes. I'm Emily Prince and I'm excited to talk to our keynote speaker, Clara Durodié, about AI. Welcome, Clara.
Clara: Thank you for having me.
Emily: You're very welcome, it's great to have you here, Clara. I have to say, I'm truly in admiration of your career and what you've done and the choices you've made actually, at a very early stage and frankly, very pioneering. That choice that you made early on from going from, frankly, a kind of comfortable, well-established career in asset management to pursue Python. What motivated that and what clarity of thought really drove you into that choice and how much has it shaped your outlook on AI today?
Clara, Well, the conversation I started having around AI must have been around 2011 and it was a function of the level of complexity and the size of compliance teams and risk teams. We, in the front office, were just looking at our colleagues in compliance teams. Their numbers were growing faster, almost exponentially every week, and I was thinking to myself that this can't carry on like this. It's just not sustainable for a business.
So I went to my board having done a little bit of research. I have always been, having said that, interested in technology, going back to the mid-90s. It was just one of those interests one has. So I went to my board and I said, "How about we do something about this technology that's coming through, artificial intelligence?" The question was, "Is anybody else doing it?" And I said, "Well, maybe that's the point, that no one else is doing it. So it gives us an edge, gives us the ability to stay ahead." The room went cold. We were going to do what everybody else does, and that's what we did.
Three years later, I resigned and left it all behind. I started learning Python in 2015, founded Cognitive Finance Group in 2016, and the rest is history. I was a contrarian investor, trained by one of the best in the field in London, and I remain a contrarian thinker. That's where we find value in our industry and, as it turns out, in my career too.
Emily: Yeah, very good choice. It strikes me, Clara, that what you're describing is so important and actually what differentiates the winners within AI. It's not necessarily that specific skill or specific experience, but actually it feels, from what you're describing, much more about attitude and, frankly, the courage that goes behind it.
Clara: It’s the attitude. It’s the conviction. Like any successful fund manager or investor, we know that if you have conviction in your ideas, typically it works out. So it's the courage of having your own ideas and staying firm behind them. Clearly, after you've done enough work to have a sound analysis of everything, you've got to continue to support them.
You and I spoke about it, and I think it's one of the most important values of leaders in our industry going forward: the ability to stay humble. Humble, yet courageous. They are seemingly two qualities at tension, but they're not actually; they complement each other. We are in a space where the leaders who will make the right decisions can see and know what they don't know, and they have the curiosity and the courage to go and find out. Those who think that they know everything, it's clearly not the case. Critical thinking will be a highly valued skill amongst the leaders of our generation.
Emily: I certainly agree, but it feels like a true responsibility that we're setting out for ourselves. One of the things that strikes me as you were talking is that it feels sometimes we become very short-termist when we're talking about AI, perhaps not helped by the level of marketing and hype that's around it. How do you make sure that we're really lifting our heads up and understanding what AI really means for us in the longer term? Even over the next three years. How do we gain that stability, persistence and true ambition that we can actually execute behind?
Clara: I would argue that AI is a long-term investment. It's a long-term story. Anybody who treats it as a short-term opportunity is more interested in running after what looks good rather than what is actually good for the business. I think they most likely won't be the winners. The winners will be those who take the slow approach, pay attention to what needs to be done, and are intentional about their final outcomes, business outcomes clearly. They have to have the vision of where they are going and where they are taking the business, as opposed to where technology takes them.
So the winners will be those who have this long-term view and are able to withstand a lot of pressure and a lot of hype. I would argue that if they explain eloquently why they're doing it, you'd be surprised that investors and shareholders would actually value that more.
Emily: Yeah. I think there are so many important points in there. Actually, when we think about all of this, it reminds you of the responsibility that we have. Linking together those points around humility is kind of striking in terms of the responsibility that we are all leaning into, but also the technical understanding of the role these systems are playing within our organisations, or can play in our organisations. How do we make sure that we're re-establishing responsibilities, assuming that we're not delegating responsibilities to model owners? What does it look like in terms of parenting? I think you've coined the term “parenting agents”. What does that mean to you, and how do you practically get the right framework in place so that you can uphold that duty of enablement while also maintaining clear control?
Clara: There are two important points that you've just raised. One is ensuring that the values that we hold as individuals, our backbone, are reflected in how we design these systems. Just like we raise children, our children will reflect the values that they were brought up with. I was inspired by looking at my son when he was little and seeing behaviour traits in him that were actually mine, for better or worse. So that's one important point: the values that we have are reflected in the systems that we build and design.
The second point is the oversight function and the need to have humans involved. Human in the loop. It's part of governance. It's a governance tool going forward. Some people say we should have humans involved because they will carry the responsibility. I would argue a slightly different point, which is that humans in the loop will enable not only better governance and oversight, but will also ensure that the systems are slowed down to a point where a human is able to execute quality oversight over them.
Clearly some people will say, "Yeah, well, we don't want that. We just want the system to run, to be the fastest and the quickest." I'd say the slow-hand strategy ideas, which we coined and promoted to our clients and which they have adopted, are all about slowing down what needs to be slowed down and being intentional about what you want to achieve.
Running just because everybody else does it, and scaling and rushing to achieve something, that's not the point. You are missing the point if you do that. Any shareholder that puts that level of pressure on companies to keep rushing, keep rushing... actually, we've seen that in quite a few cases. Apple was one example about 18 months ago. They were criticised for being the slowest and actually, if you look at the quality of the outputs and the products they have brought to the market 18 months later, you'd see why and appreciate that.
Emily: One of the things I think is really unique about your role, Clara, is that you're meeting with CEOs and executive teams across the whole industry and also in other industries. I know that you've just returned from a major energy conference. I'd love to know, from your perspective, what are the questions people are not asking? What are the CIOs and boards not asking enough about?
I would put them into three main categories. The first one, and one of the key questions because therein lies the competitive advantage companies are building today for the future, is: how are we rearchitecting business models, operational models, profitability models and revenue models so that we build a sharper competitive edge and the arbitrage stays with us? So how do we retain arbitrage whilst deploying tools that everybody has access to and incurring the associated token costs?
Exactly. So that's the first question. The second question is that we need to understand first that energy security is absolutely key. I keep saying to people that the AI story is an energy story with some software on top. That's what it is. The moment you flip the narrative that way, everybody understands that we are into something. We are building a digital construct depending on an architecture we don't really understand.
So the first point is energy security. The second point is communication security, and by that I mean undersea cables. The UK is 99% dependent on undersea cables for its entire digital construct, which is quite something. As one of our experts specialised in data centres mentioned to me about a week ago, in addition to that there are the 500 data centres that are being built, and that is an additional undersea cable story which we don't have. I think boards need to have these kinds of conversations and touch on these points.
The final question that is not being asked, or not being monitored properly where there is a lot of room for improvement, is: how do we ensure that we are investing in this as a long-term story, as a long-term business reinvention, as opposed to a short-term initiative?
This tension between these two objectives can create a lot of confusion. But good leaders, strong leaders, can tell the difference and they stay committed to the long-term story. Again, to deliver on a long-term story, you need intentional deployment.
Emily: That dynamic in terms of really understanding the supply chain and arguably much more concentrated dependencies is interesting. As a community, we talked a lot about what it meant to have one model that we were all dependent on, and there was a lot of focus around that, but not necessarily the other side of what this means from an energy perspective, which is a much more unfamiliar topic for many in financial services and arguably one of the most important conversations for us to be having now.
I actually want to take us on a slightly different note. One of the conversations that you and I were having was around the role of psychology. In the realm of things that are perhaps not commonplace conversations, why is it that you feel so strongly about the role of psychology in financial services now?
Clara: After I left the industry in 2015, I was very keen to do a PhD. I went to my old school and said that I would like to do a PhD at the intersection of neuroscience, artificial intelligence and investment management.
Emily: You really are a pioneer.
Clara: And the answer was, "You want too many supervisors. We've never done that before." But neuroscience has always been an area that opened up possibilities for people like us who are practitioners and involved in designing technology for the future. Neuroscience teachings and research, when overlaid on product design and business strategy, enable us to have a different perspective. I would argue a more nuanced and more valuable perspective.
Psychology has its own role, and it's so important to have that. Equally important, and perhaps in my view where there is a lot of value, is bringing a neuroscientist onto a product design team in financial services. I know that sounds completely off the charts, but I think neuroscientists are definitely our new best friends. Alongside energy executives, they should be part of a bigger conversation that we have. These are skills and areas of knowledge that we need to bring into our boards and into our investments. I would very much like to, and hope that, the industry will open up a little bit more.
Emily: When we think about the older systems that we have in financial services that are often deterministic, we often have a defined outcome and it's relatively easy to measure whether we did a good job. When we look at these systems, it’s actually about the evaluation frameworks behind them. It is that link between trust and how people associate trust. It feels like an incredibly neurological set of decisions. It's not just the right outcome, but whether it was presented in the right way that garnered the right level of trust and responsibility that we needed to ensure.
Clara: My conviction is that people have an emotional relationship with money, whether it's institutional money or personal money. It is an emotional relationship conditioned by years of life experiences.
Emily: And bias.
Clara: And bias. That's why we need experts in understanding us better, so that...
Emily: We can get that representation.
Clara: Exactly. The technology we are creating, the digital workforce that we are creating, and that will populate our industry and become intrinsic to everything we do, will have the qualities and systems to make decisions, judge and seek the truth.
Emily: That's quite extraordinary really. You've talked about cognitive capital, and I'd love to hear, as you're thinking more deeply about that, what it represents in the next few years for our industry.
Clara: I would say that cognitive capital will become an entry point in every single valuation. It will be a conversation that shareholders and investors will have. It's the compounding quality of the data, the systems that we bring in and the digital workforce that we bring in. If they have it, they're valuable. If they don't have it, then we need to go back to the drawing board and see how we can improve them and in what type and shape.
Emily: Fantastic. Clara, it's been an absolute pleasure speaking with you today. Thank you so much for making the time.
Thank you for joining us at FTSE Russell Convenes.
