Michael: Hello, hello, hello, hello. And a very warm welcome to Technology Now. A weekly show from Hewlett Packard Enterprise, where we take what's happening in the world around us and explore how it's changing the way organizations are using technology. We're your hosts Michael Bird...

Aubrey: And Aubrey Lovell. And in this episode we're exploring the rise in interest rates and what's behind the scenes that can teach us all about forward planning, strategy and AI. We'll be looking at the tools banks are using to predict the future and act accordingly when it comes to setting interest rates. We'll also be examining how the world of finance is increasingly relying on machine learning to augment human expertise. And of course we'll be looking at the books that are changing the way you see the world. So if you're the kind of person who needs to know why, what's going on in the world matters to your organization, this podcast is definitely for you. And if you're enjoying it, subscribe on your podcast app of choice so you don't miss out. Here we go.

Michael: Okay. Well, interest rates have been climbing around much of the world for the last year as banks try to curb soaring inflation driven by high food and energy prices. And in March, the US Federal Reserve raised the target range to between 4.5% and 4.75%, a 15-year-high. And pushing borrowing costs to a new high since 2007.

Aubrey: And it's not just the US though. In the UK the Bank of England has also been regularly raising rates at the fastest level since 1989. It's a great year. Do you know why, Michael?

Michael: I don't know why.

Aubrey: Side note, I was born that year.

Michael: Great year. That's a great year then.

Aubrey: So you can blame it on me, but that's okay.

Michael: Yeah.

Aubrey: So anyways, it was the fastest level since 1989 to combat inflation, which is top 10%. And similar decisions have been made in many other parts of the world. The decision to increase interest rates to try and curb consumer borrowing and spending is, rightly, made by humans. But increasingly, AI is playing a part in developing the predictive models economists rely on to make those decisions. And that opens up some interesting questions about the use of black box AI in the economy. So why should we care about rising interest rates?

Michael: Well, this week we're joined by Adrian Lovell, who is no relation to our own Aubrey, I don't think. And he's a chief technologist for the financial services industry at Hewlett Packard Enterprise. Adrian works with some of the largest organizations in the world to build understanding around, among other things, AI and machine learning in the FinTech sector. Adrian, thanks so much for joining us.

Adrian Lovell: Absolutely, my pleasure.

Aubrey: So Adrian, when it comes to setting interest rates both on a customer and national level, how much technological support are economists currently getting?

Adrian Lovell: It's very varied. Some economists rely heavily on automated systems driven by AI to give predictions. Others prefer to use their gut feel, their history, their more traditional methods. But I think the majority tend to use a blend of everything. Most don't use one source, but they take multiple sources and make their predictions from that.

Michael: And so how does modeling and economy work with AI? Is it quite a prevalent thing that's happening nowadays? Because obviously they weren't able to do that 10, 15, 20 years ago. Presumably there are hundreds and thousands of variables. It feels like it's basically an impossible task.

Adrian Lovell: It's very complex. And I've not seen... Out in the world that's not set. It's not happening. But I've not seen an entire economy modeled in AI to the depth that an economist would be looking for to make future predictions. However, you do regularly see AI used to predict subsections like retail sales with it for food products and things. So you take bits of economy and model those and you take all those multiple different modeled things and put them together to get your final picture.

Michael: Do you think it's something that is going to increasingly happen more and more? Do you think there's going to be more AI modeling on a much, much larger scale?

Adrian Lovell: Oh yes, a 100%. I think when it comes to this sort of thing, the more different models, the more sources of information you've got, the closer you're going to get to a more realistic prediction. Because ultimately what we're trying to do is we're trying to predict the future. So the more different versions of predictions you can get, the more likelihood that is when you mash them all together and get a human being says, "That's what I believe." They're more informed and are more likely to be accurate. Same as weather forecasting, for example. The more different ways you think about the weather, the more likely it is your weather forecast will be accurate. The same is true of an economy.

Aubrey: Speaking of accuracy, how accurate are solutions at the moment? So for example, can an economist plug in a set of variables and kind of let that system go?

Adrian Lovell: It's very, very variable. When the world is not doing crazy things and everything is nice and calm and not a lot's going on, then your models are going to be reasonably accurate. When something completely untoward happens in a geopolitical environment like one country invading another or something of that nature, the models are completely thrown out of kilter. Because that's not a scenario that's typically being built in. So everything goes left field and that's where the human intuition comes back into play. So it's always a mix of human and machine, not one or the other.

Michael: And what are the challenges right now regarding AI in finance, are they like regulatory or technical challenges?

Adrian Lovell: All of the above.

Michael: Okay.

Adrian Lovell: So it's a common topic with regulators at the moment as to, where should we trust AI to use or should we trust AI? And where is it appropriate to allow AI to do some things versus other things. What we're seeing is a focus on the outcome of a given process rather than a regulation for a given technology. But I think thinking is evolving around that and there's going to become areas where, is it acceptable for private client data to go through some of these massive models that may be inherently biased? Or all sorts of other reasons why they're not ideal. But today it's predominantly used again as a human advice tool rather than the end state for a large proportion of uses in financial service. Not all, but the majority.

Aubrey: So Adrian, why should other organizations be watching the way AI is used in the financial sector? What can the rest of us learn from that?

Adrian Lovell: I think in financial services, we've been fairly early adopters of some of very advanced AI technologies in a lab scenario. But we've been quite cautious of where and when we then put those technologies into production. That's becoming a lot more frequent now. It's a lot rarer to see things happening in the lab. And I think that's what people should start looking at, is the way the financial services industry has moved AI from this lab project thing to just another part of the development ecosystem of normal application and business systems' development. So it's almost become just another tool in the toolbox. A very powerful and important tool, but just another part of day-to-day business. And I think that's where other industries can look to financial services and really learn about how to get value themselves from it.

Michael: I'm going to ask a quick bonus question. [inaudible] The Bank of England recently published a report at the start of October, which basically said in short, "Yay AI." But they did warn about the possible need for more regulation. Why is that?

Adrian Lovell: I think there's a number of concerns from the Bank of England and other regulators around the globe around the potential misuse of advanced technologies. And that's not limited to AI. But as you make more advanced technologies easier to consume, you have to be cautious that you don't let the technology do things that you're not expecting it to do. An example of that would be if you give a piece of technology, a business process, full automated control over somebody's life. Something that is, especially in financial services, a large part of what we do in financial services impacts somebody's life. If you give that technology full control over that and it doesn't behave as expected, you can have devastating consequences on somebody's life or business. So to prevent that ever even being a remote possibility, you apply regulation to ensure that the technology can be used responsibly rather than irresponsibly.

Michael: Yeah. Brilliant. Okay, thank you, Adrian. That was really fascinating. And we'll drop some useful links about some of the stuff that we've been discussing in the show notes. And we will come back to you in a moment Adrian, so don't go anywhere.

Aubrey: Next up, it's down to you, our audience. We open the floor for you to give us your recommendations on books which have changed the way you look at the world, life and business in the last 12 months. They could be technology-based, have changed the way you work. Or they could have just made you look at the world in a totally different way.

Michael: And if you want to share your recommendations, there's a link in the podcast description. Just record a voice note on your phone and send it over.

Rutwik Shah: I'm Rutwik Shah. I'm currently the director of data and clinical product management at a health startup called Kaliber Labs. And one of the books that I came across recently is a book called Zoobiquity by Dr. Barbara Natterson-Horowitz, who is a cardiologist at the UCLA Medical Center. And the book touches upon the various similarities and differences that exist between human and animal health. And then specifically trying to find patterns or mechanisms that can be translated from the animal domain which could positively impact human health. So for example, with cancers that we so commonly see in human beings are almost non-existent in larger creatures such as elephants and whales. Actually turns out they have protective mechanisms at the cellular and genetic level, which are now identified by researchers and could be translated and implemented in humans to preemptively prevent the emergence of cancer. And then this is again, a very interesting book that I would recommend to the listeners of this podcast if they have an interest in the intersection of human and animal health.

Michael: Thank you. Thank you. And once again, do keep your book reviews coming in via the link in the description. So Adrian, have you read anything recently that you can recommend to us at all?

Adrian Lovell: I've got to be quite annoying for the podcast here in that most of my reading is around the industry. Financial services, AI. I tend to read in specialist journals and magazines and things rather than books. So I'm not going to recommend a book to you today, but rather I'm going to... I know. So you might have to cut on this or ask me to come back in a couple of days time when I can go and speed read something to [inaudible]. Because I read rubbish books because that's what I enjoy reading. But what I tend to read when I want to immerse myself in my professional life is either academic papers on a topic. Or more accurately I tend to read summaries of academic papers and where that academia has been applied. And I tend to find publications such as The Economist, a fantastic start. So that's tends to be where I gravitate to rather than rather long books on the topic. So that's unfortunately not quite what you're hoping to touch.

Michael: Still a good answer. Still a good answer. Still a good answer.

Aubrey: All right. It's time for questions from the audience. You've been sending in your questions to Adrian Lovell on the topics of AI and FinTech. And we've got a couple lined up for you now.

Michael: Yeah. So James from Hong Kong asks whether you can see a future where AI assistance will be mandatory as part of due diligence in financial policy setting.

Adrian Lovell: Hi, James. I think it's not out of the realms of possibility. I think it's quite unlikely that a specific technology would become mandatory. And the reason I say that is technology moves so fast and regulation often moves very, very slowly to define a given technology in regulation. The regulation suddenly becomes out to date and people can't use the even more advanced technology when it comes along, because the regulation says they have to use this old thing. So I think it's unlikely. I think what's more likely is that regulators may require something gets run through a regulatory controlled system, which happens to be powered by AI. As opposed to AI itself being regulated.

Aubrey: So we have another one for you, Adrian. Hailey from Dartford has asked almost the opposite question. She wants to know how you think FinTech institutions and particularly banks will cope with the black box nature of AI if and when it makes a suggestion or assumption which turns out to be incorrect?

Adrian Lovell: That's a great question, Hailey. And that's really a very, very common question and challenge that we're facing right now. The regulators are very concerned about all black box technologies in use across business processes, across multiple different industries, not just financial services. And they're looking back to core business processes outside of technology, which is what do you do to make sure your black box is operating as expected? And if you don't want to use it anymore or if it breaks, what are you going to do instead? So the regulatory approach so far, or at least the direction of travel, seems to be step away from the technology and look at the business processes around it to make sure that you can swap out that black box or turn it off or keep an eye on it. Rather than having to understand what's in it. They would rather you understood what's in it, but they recognize that's not always going to be the case.

Michael: Cool. Thank you Adrian. And again, we'll drop a couple of links in the podcast description for more on these topics. Right then, we are getting towards the end of the show, which means it is time for, this week in history.

Aubrey: This week in history.

Michael: Which is a look at monumental events in the world of business and technology which has changed our lives. What do we have, Aubrey?

Aubrey: The clue from last week was up, up and away. And we actually have two exciting events to talk about. The first man in space and the launch of the first space shuttle. First off, on April 12th, 1961, Yuri Gagarin became the first man to orbit the Earth aboard the Soviet spacecraft, Vostok 1. He orbited the Earth for an hour and 48 minutes before reentering the atmosphere. Then on April 12th, 1981, NASA launched STS one, sending the space shuttle Columbia on its maiden voyage. The mission was a test bed for reusable spacecraft. It was actually slated for launch two days earlier. So the whole same day as Gagarin thing was just a happy coincidence. And all 135 shuttle missions would launch before the program was closed in 2011. And side note, Michael, I actually grew up in Florida, so I actually could see a lot of these launches, which was absolutely fascinating.

Michael: I mean, I've always wanted to see just a spacecraft launch. But man, a shuttle must have been so cool. What was it like?

Aubrey: It's pretty cool. I mean, it's hard to describe in words, but I will tell you that one of the most incredible experiences is the night launches. Because you're actually turning the night into day. It's so powerful, the whole sky goes bright and for a few seconds...

Michael: Wow.

Aubrey: It's like 2:00 PM in the afternoon, but it's really 11:00 PM right? So,

Michael: Oh my goodness.

Aubrey: It's pretty cool.

Michael: Gosh, that would look so cool.

Aubrey: And next week, the clue is Bill Gates is feeling blue screen. Know what it is, keep it to yourself.

Michael: And that brings us to the end of Technology Now for this week, next week we'll be discussing alternative roots into tech and STEM with high-performance computing specialist at Alces Flight Limited. And wait for it, classist, Kristin Maritz. So here we go. [foreign language]. I have no idea if I'm saying that right, but apparently it's Latin for see you then, according to Google. Aubrey, any good at Latin?

Aubrey: Not at all, but it sounded beautiful to me. Good job.

Michael: Thank you. Thank you.

Aubrey: Until then, thank you to our guest, Adrian Lovell. Thank you so much, Adrian. And thank you all so much for listening. Technology Now is hosted by myself, Aubrey Lovell, and Michael Bird. And this episode was produced by Sam Datta-Paulin and Zoe Anderson. With production support from Harry Morton, Alicia Kempson, Allison Paisley, Alex Podmore, and Ed Everston. Technology Now is a Lower Street Production for Hewlett Packard Enterprise. We'll see you next week.