0:57 Introduction
So, my name is Matthew Griffin. I'm the founder of the 311 Institute. So we're a global futures and deep futures think tank looking up to about 50 years out. So when we're actually trying to look at different futures, we look at Horizon 1 which are the next five years. When you actually have a look at the future, increasingly the financial services industry and what it represents is a key battleground to win. And as we sit here today and as I'll go through towards the end of this presentation, you can see the ground shifting beneath your feet.
2:46 A VUCA Max world
Now from a global perspective really there's only one thing that's actually changing and as a community we should think that we could get our head around one thing that's changing unless of course that one thing happens to be everything. Now in the 1970s, the United States military coined this phrase VUCA to represent their point of view of the future. What they thought the future was going to look like, volatile, uncertain, [complex] and ambiguous. But about two three years ago we added this MAX acronym to it. And the reason for this is today a lot of the changes that we see are massive. They are global in nature and they are system level. So if we have a look at this from an economic perspective as we see the two superpowers fighting it out. So namely China and the Americans. We see the Americans using sanctions and blacklisting more as an economic tool or to force or weapon depending how we want to phrase it.
Now from a global level this means that we are seeing the payment systems split. We're seeing the global reserve currency under threat. We are fundamentally seeing an economic decoupling which is going along BRICS too as well as China, Asia, India and so on so forth. We also see massive technological change. The US and China at the moment are in a fight for the future of artificial intelligence. The Chinese basically with relatively limited compute resources now have some of the best foundational AI models and increasingly their AI models are open weight and open source. The Americans basically are going in closed and proprietary sort of AI models: Anthropic, Google, OpenAI.
Now at a global level, the American AI companies will argue that a closed ecosystem brings in trust, reliability, security. But there are challenges with that. The reason why the Chinese are going an open source route is because the Chinese fundamentally want AI applications to be built on a Chinese technology stack. And there's risks with that as well. We see the future accelerating typically. Do you think that we're going to be moving faster in 10 years' time as a society and as an industry or do you think we're going to be moving slower? Because generally we're going to be moving faster. And I'll show you why we're already moving too fast. And then the changes that we see are exponential.
Now what we mean by this is typically we talk in terms of exponential technology. Artificial intelligence compute is doubling every 3.4 months. When we think about computer chips that's an exponential technology but disruption is also increasingly exponential. And increasingly you can all be disrupted within a day. And I shall show you how.
6:43 Seeing the future, and strategic drift
So with the vast majority of organisations that I work with, I think most of them only see 30% of what's happening today. So if you take technology for an example, everyone talks about AI. We saw generative artificial intelligence emerging in 2016. These technologies, deep fakes, synthetic content, and the impact on fraud, as well as generative artificial intelligence suddenly was on every board agenda. So, we can see the future today, but only if you're looking in the right places. Now, if you only see 30% of what's going on, you're missing 70%. And if you're missing 70% of what's going on, you get strategic drift as an organisation. It's more likely that your vision, your strategy, the markets that you care about, how you execute is all going to be changing and this affects your results.
Now from a leadership perspective, the boards that I work with typically depending on the sector they will say the CEO will typically say we look at four areas. We look at what's happening in our industry. So if you're in the financial services industry, you're probably looking at the financial services industry, right? How many of you are actually looking at other adjacent industries? What they're doing, how they're coming into your market. Then we look at markets which are economics, economies, consumer behaviours and habits. And then we look at tech. And increasingly we put technology by itself because if you know how to leverage technology, you can gain an unfair advantage.
But if we take your technology away, you're basically dead in the water. And then they care about senior leadership development. Now, one of the biggest problems that organisations have when it comes to seeing the future and trying to be future ready, maybe not future proof, is complacency and inaction. It's all very well being the company that you were 10 years ago, but as we heard, with so many startups coming up, with so much capital flooding the market, if you're an organisation that stays static, firstly, you're going to get hit with fraud. Secondly, your customers might start moving away. An example of this is Lloyds Bank in the UK. This year with a new CEO, they've decided to become a fintech.
And unless I was wrong, I think fintech started to emerge in 2005, if not earlier. So, 20 years late, but they're moving. But if you think that you're okay where you are today, for every one of you, I can show you a new competitor that's going to try to eat your breakfast. Another problem that I see in leadership teams is honesty. Now what I mean is I don't mean that leadership teams are dishonest and there are things that the organisation just doesn't talk about. Now in a market that's changing very very quickly if there are things that we can't talk about within a company you're not going to know about the problem. Therefore, you're just going to get hit in the face by that problem.
If you want to run an effective business and if you really want to stand the best chance of being the best organisation that you can be, people have to be able to come to you and say, "I've seen this problem. I've seen this problem. And I know you might not like these and it might be an uncomfortable conversation, but I thought you should know." The more issues, problems, and things that you can see, the greater your point of view. And you can discuss these whether you care about them, whether you don't care about them, and what you do about them. So, we all have people within the organisation who just don't want to tell it how it is. And that's a culture thing.
12:39 Volatility: shocks and options
Now when we have a look at volatility, there are lots of different volatile shocks that we're seeing at the moment. We're seeing an inflation shock. Now we believe that over the next two to three years global inflation might be sticky. The price of oil and gas influences around 80% of the global economy and we think inflation could start sticking at about 4% in the United States. That means higher interest rates unless of course you see a trend called fiscal dominance. We see lots of market volatility. So we see huge numbers of tariffs which affect everyone. And we see sanctions. And these sanctions and blacklists are increasing. So organisations have to move out and cut their links with Russia. But on the sanctions front, if we have a look at China, China's oil refineries recently had a message from the CCP where the CCP said, "If the Americans are trying to sanction you, just ignore it." So, we live in a highly volatile world. You know, we see that whether that's conflicts, whether it's new technologies, whether it's changing consumer behaviours.
And so in a highly volatile world where the uncertain future is kind of the main future, you need to build out options. So this is a mixture of resilience, diversification because if something happens over here, you are protected.
14:42 Uncertainty: AI, regulation and dynamic strategy
Now when we think about uncertain, artificial intelligence is right up there. Now, the AI models that you're using today, like ChatGPT, have an IQ of 155. They have 4,000 times more data than any human, and they learn 500 million times faster than any of us. By 2030, we estimate artificial general intelligence in whatever form it comes out in, could have an IQ of over a thousand. But before we get from here to 2030, we've got agentic artificial intelligence, quantum artificial intelligence around 2029, which will change how you calculate risk, loans, how you assess fraud, KYC, AML, money laundering, that sort of stuff. But we also have neuro-symbolic artificial intelligence which are AIs that think like humans. This is where we combine neuroscience from humans with reasoning models.
In addition to that we have photonic GPUs and systems coming through. When we have a look at the cost of AI at the moment, the vast majority of organisations that are using artificial intelligence are using the best most expensive foundational AI models for the most basic queries. So when you think about the impact of AI as you start embracing agentic artificial intelligence in the labs, we're way beyond that already and these will come through faster and faster. Consumer behaviours are changing and the regulation landscape is always changing. And when we have a look at regulation, we're regulating everything from digital, virtual, physical assets, everything from cross-border.
Increasingly regulation is taking up around 10% of organisations' revenues. So compliance is one of the biggest cost centres for the vast majority of people in the industry. But in order to reduce uncertainty, you have to try to see more. The more things that you can see, the more things you can discuss. And the more things you can discuss, the more things you can prepare for. If you can't see it coming, it hits you in the face. Now, from a foresight perspective, as things get faster, if we're moving faster today than we were 10 years ago, and we're moving faster in the future, then your five-year strategy is really a three-year strategy. Your three-year strategy is a one-year strategy.
So, increasingly, you need a dynamic strategy. How many five-year plans that you all created had agentic AI in there? You probably put it in what a year ago, year and a half ago. How many of the plans must you've got all the other things in? So the more that you can see, the more you can discuss, debate and prepare for.
18:23 Complexity: geopolitics and systems thinking
Now when we have a look at complexity, we see geopolitical fragmentation. Essentially, we have the great powers including Europe that are trying to outmanoeuvre one another. And in Europe for example, basically we are trying to build out our own payment systems. The reason being according to Macron and Merz etc. because increasingly Europe does not necessarily trust its old allies as much as it used to. And so this affects you when we have a look at GRC. Increasingly we are developing new GRC systems that are either automated and/or increasingly autonomous. And then when we think about technology, it's not just about AI. It's about how different technologies combine and converge.
For example, your smartphone is not one technology. It's a combination of hundreds. And so increasingly when we have a look at technology convergence, yes, we have Web3 and blockchain, we've got metaverse, we have [unclear], a variety of different flavours of AI, quantum computing, GPUs, all sorts. And so to try to master complexity you need to try to think in terms of systems because this thing is increasingly connected to this thing. Now systems thinking is difficult but you have to think how do all of these things that we are looking at connect with one another. How does what's happening in the world of artificial intelligence connect to what's happening in geopolitics?
The answer there is we see an acceleration in AI development because China wants the best model, America wants the best model and then we have Europe somewhere in the middle.
20:42 Ambiguity: private money and the truth
And then when we think about ambiguity, the competitive landscape, one of my clients is Walmart. Walmart processes $716 billion of revenue a year. They don't like paying Visa or Mastercard for the transaction fees. And so between Walmart and Amazon, who control $1.5 trillion dollar of consumer spend, they're creating their own stablecoin networks. So Walmart is essentially creating its own version, in Christine Lagarde's words, of private money. When we think about consumer behaviours, again, when we think about how consumer habits are changing, I've got a couple of examples for you there. Some that you might know, some that you might not know. And when we think about ambiguity, I suggest you need to think in terms of long-term futures, typically 5 to 10 years. But this is how you deal with ambiguity.
You find the truth. If someone tells you something or if you have a hunch, don't accept it. Try to find the truth. Because in an increasingly ambiguous world where your companies and where your sector is reliant on the truth so that you can steer the companies properly, see it firsthand, experience it. Increasingly in a world basically where we can create content en masse using artificial intelligence as we see with deep fakes and etc etc. We are now in a period of truth decay and then you have to be ambitious and you have to experiment. The vast majority of banks and fintechs will tell you that they are ambitious, but they're not compared to some of the companies that are coming up.
And then you can't just look at what's happening today and what's happening tomorrow this week. You've got to be continuously vigilant and you've always got to be trying to reinvent yourself before you get out-invented. Then you've got to repeat it all. This is a continuous process. It only stops when the future stops.
23:34 The speed of change
Now when we think about the speed of change, the speed of change itself is increasingly your problem. And it's not a problem really to solve. So let me explain. You've got consumer behaviours, markets, regulations, and all sorts of different things that are affecting your markets. And consumer behaviour can change in a couple of days. It can change in a couple of months or it can take years. And at the moment we are going through a huge consumer behaviour shift that you are all probably doing and I'll tell you that later. You've got governments. Governments like the European Union used to take years to do anything. The EU AI Act for example. But now what happens on Friday is suddenly changed on a Monday. If you think about data sovereignty in Europe for example, we're now starting to think about deregulating artificial intelligence while at the same time we have new tariffs, new sanctions, new blacklists coming through. So governments are whipsawing quicker than ever before.
Physical infrastructure typically a cycle for physical infrastructure is decades. And then as we see all of this change, we all have these kinds of people. As we see all the change happening around us, you've got employees that say, "I've seen the change. I'm going to change today." And then you've got other employees that say, "I've seen everything changing. I'm not changing." And that's about 8% of the workforce, we reckon. And then you've got the rest basically, they're somewhere in the middle. So you've got people who are changing at different speeds according to what's happening outside. But then to trump them all, the speed of digital disruption is now 69 milliseconds.
The reason why it is 69 milliseconds is because when a new artificial intelligence model is produced like GPT-5, Gemini 3.5, Grok 4.5, Anthropic Claude Opus 4. When the large foundation labs like for example Moonshot AI in China when they release their new AI model with whatever it does its new capabilities its new features its new performance it's available to 4.5 billion people who are connected in 69 milliseconds that's how fast it travels so if you think about GPT-6 when GPT-6 eventually comes out from OpenAI it's available whatever it does it's available in 69 milliseconds and then we have other challenges. These artificial intelligence models are only ever going to get more powerful.
So when we have a look at cyber, increasingly the cost of me attacking you as a bank or financial services company is dropping to zero because increasingly I can build new cyber weapons in five minutes with no skills, no resources, no expertise, no money and very little time. Cyber will become one of your biggest issues. It already is a big issue. So from a speed perspective, you have to learn how to manage each layer at the right speed. If you're looking at physical infrastructure and employees, you manage that at one speed. But if you're looking at technology and you're trying to manage your technology stack in months, years or decades, you're going to be behind.
30:24 Disruption and the moat
And then when we actually think about disruption, this fintech and accelerators love this. Disruption hurts if you're not the one doing the disrupting. And it's true. If you're the one doing the disrupting, you love it. If you're being disrupted, you don't. So, Amazon, Amazon is being disrupted by agentic commerce. So, you go to OpenAI, Perplexity, Google, and increasingly by you search for your products in there. So, Amazon's e-commerce systems are being disintermediated. Amazon took Perplexity to court to get them to stop them to do it. So Amazon's defence is we're going to take you to court to stop you doing that. But Amazon was also started being disrupted by super apps like TikTok. Temu went from zero to $50 billion in revenue in 3 years.
So we disrupt Amazon with Mastercard and Visa and Discover. Increasingly they're being disrupted by agentic AI to AI payment systems which is why in Visa's case they say we don't really know who wins the pay the future of payments so we're just backing all of them but increasingly when you have a look at companies like Mastercard they're buying cyber security companies and Microsoft. Microsoft's moat was billions of lines of code. So if I can increasingly use artificial intelligence to clone software like Workday, Salesforce, ServiceNow, all the different sort of core banking systems, IBM's COBOL systems for ATMs, what's your moat? But then how do you leverage it?
So whether or not you are disrupted as an organisation comes down to these four things. Your people. So can you see new things emerging? Are they threats? Are they opportunities? If they're either, what are you going to do about it? Do you chat about it? Because if you don't have the right people, then you're either not going to see something or like IBM, you're going to see something and put it under the bed and ignore it. And in IBM's case, they lost half of their market cap and half of their company. In Accenture's case, they saw things coming. They moved. Within the space of 10 years, they tripled their revenues from $20 billion to $65 billion. The only difference really in that scenario was the people. Then vision, strategy, you see things changing. So what's your vision? What's your strategy? What's your company purpose?
And then none of these are really worth anything if you don't know how to execute. If you have the best business plan, the best products, but you can't execute, your competition will always win.
34:19 The intelligence transformation
Now, from a financial services perspective, increasingly you are not undergoing a digital transformation. We have digital native companies and now we have the rise of AI native companies. They are not the same. As you start embedding more AI within your companies, AIs that take information in, analyse it, that you then give the power to do something with that act, make a decision. You're undergoing an intelligence transformation that is a fundamentally different set of partners, projects, programmes, workflows, operating models. You also move from automated to autonomous operations. And increasingly you can buy access to human level intelligence from the foundational labs for whatever 10 bucks a month.
And outside of the token costs, you can scale human intelligence now for almost nothing. So think about all the resource that you need in your organisation. None of you have enough people. You all have hiring ads out. What do you do with this? [Then there are] agentic core banking systems that are intent-based. So you simply say what you want to happen and the AI systems will say, I understand what you mean. I know what needs to be done. I know the systems I need to coordinate with and off I go. Now, from an organisational perspective, most companies only see AI as doing two things. You'll see AI as a technology that you use to automate people and processes.
It's fine. You need to be a lean operating company. Or you use artificial intelligence to amplify your staff, which is another way of saying help them improve their productivity. Right? 99% of companies use AI to automate or improve employee productivity. You are using AI to improve the company that you already have. And you're bolting artificial intelligence onto a digital business model to create a Frankenstein. The best companies increasingly companies like Google and so on and so forth as well as BNY Mellon have a third expand. None of you are short of opportunity. In fact, you have too much opportunity. You just don't have enough time, money, people, or resources to have a 100% market share.
You haven't got it. So, this is where artificial intelligence comes in because it helps you expand your revenues, your profits, your markets, and your market share. Because AI under the covers is this AI is an accelerant. So for example, it will help you do market analysis faster, develop new products faster, do competitive intelligence faster, do governance faster, manage fraud and authentication faster. So it's an accelerant, but it also helps you build capability and capacity. Now think about it. If you had access to a human level artificial intelligence and I said to you, next year I want you to grow your business topline by 20%. You can create autonomous bots, autonomous agents, all sorts.
39:02 The business of banking
And then finally basically when we actually have a look at the business of banking itself you have banks that use AI cloud services microservices and everything else. But a bank is still a bank. A bank takes your money loans your money out to someone else and then takes interest on the loan. The actual bank business model to consumers is we take your money and we keep it safe and protected against cyber nuclear threats, regulation and all sorts. Fintechs, most fintechs are banks just that are digital native. Fantastic app, fast onboarding. So when you think about the Medici business model of we take your money and we loan it out, that's a 14th century Italian business model that every bank still rides on and every fintech sort of still rides on. Then you've got decentralised finance, so crypto, decentralised share buying, decentralised asset management, etc.
The companies that are the most ambitious today, there's one in particular. When they came out, they said, "We're a super app." And the London market and investors said, "No, you're not. You're a fintech." They said, "No, we're not. We're a super app." And everyone said, "No, you're not. You're a fintech." Revolut are in the financial services market, but they want to IPO when they hit 200 billion. Revolut is increasingly an organisation that is a fintech that is building a super app that bridges traditional banking with decentralised finance that we can then all tap into. They are the fastest growing bank on the planet. In the UK, traditional banks didn't think that Revolut would amount to much.
And they're not alone. Then you've got banks embedding their services into everything like fridges. You can do banking on your fridge with Samsung. Your fridge increasingly is a robo customer. It runs out of milk. It buys milk on your behalf at a discount. It takes the savings, so it saves you a euro perhaps. It then asks you, "Would you like to put this euro into a fund?" So like a robo wealth advisor. And you say, "Yeah." And so this fridge now invests the amount of money it's saved into a fund and you get richer. You can embed banking in everything. So you've got embedded banking and then you've got companies basically that say we'll own the ecosystem. We'll own all the companies.
So they own the whole lot. But this is what you are all doing and you don't maybe know it.
42:44 Disintermediation: when AI becomes the bank
Increasingly, you are all being disintermediated. Artificial intelligence is gathering 165,000 data points on all of you. It knows you better. Think about how you used to use Google search and how you use Gemini or how you actually use ChatGPT. 165,000 data points. A year ago, Sam Altman said AI will learn your entire life, which is handy. So, it knows everything about you. In the US last year, 54% of people were using ChatGPT for investment advice and financial advice. And you probably are as well. 80% of high net worth individuals are doing the same. What are you all doing? You are all changing your behaviours. Now, a year ago, Sam Altman hired 150 people from Wall Street to create better financial models and so on and so forth.
And a year ago, I said, "Watch out for this." How many of you have logged into ChatGPT recently and ChatGPT has said, "You can now connect your bank account to us." ChatGPT can now manage your financial life. Now, at what point does ChatGPT get connected to your digital wallet or your HSBC bank account? And here's the rub. Increasingly, I say to ChatGPT, take £10 out of my HSBC account. I want it to yield 10%. ChatGPT is an intent-based system. And now it goes, I know what you mean. I will figure out how I make 10% returns for you. I will choose where to invest it. And who with I don't know where it's gone. ChatGPT as well as others are increasingly disintermediating all of you. And if they really lean into it, OpenAI becomes the world's biggest bank and newest fintech.
And that is it from me. So thank you very much for listening and keep vigilant. Thank you.
Transcript from the recording, edited only for filler and caption errors. Audience questions are summarised; the host and other speakers are left out.
