- Solve real problems with our hands-on interface
- Progress from basic puts and calls to advanced strategies
Interactive Options Course
Posted December 11, 2025 at 9:58 am
Join host Andrew Wilkinson and guest Alex Gunz as they explore whether aging power grids and legacy infrastructure can sustain the explosive growth of AI, robotics, and future technologies. From data deluges to energy bottlenecks, they break down the challenges and opportunities shaping the next decade of innovation.
The following is a summary of a live audio recording and may contain errors in spelling or grammar. Although IBKR has edited for clarity no material changes have been made.
Welcome to today’s episode. It’s that time of the year again when analysts and portfolio managers compile their review of the year and often provide us with a glimpse into next year. My guest today has his finger on the pulse of technology, particularly future tech. So welcome back to the program, Alex Gunz, portfolio manager at—how are you, Alex?
Wonderful, Andrew. Thanks so much for having me on the show, and as you say, it’s a very pertinent time of the year where everyone is asking us how we see the world for the next 12 months ahead.
Well, you’ve just released the latest edition of the Annual Future Trends compendium. So in a sentence or two, Alex, what’s the core thesis of Volume 12 and how does it evolve last year’s view?
Well, it is great to have the opportunity to talk about this, Sandra, and I guess the way to frame any conversation on this topic is, look, we’ve been writing about investing in future trends for over a decade by now, and each compendium really represents a continuation, a follow-on from the prior one. And really at the heart of everything we do, it’s this very sort of Darwinian idea. Darwin famously said it’s not the strongest of species or the most intelligent—and we might replace that word with businesses or trends—but those that are most responsive to change that will actually succeed. So what we’re really trying to track and to monitor as effectively as possible is how the key trends that we follow have been evolving in the last year. Now, as you are aware, and as all your listeners will be aware, the topic which is undoubtedly front of mind is AI. And AI, we’ve all seen over the last 12 months, has accelerated massively. There are obviously caveats to this growth—I’m sure we’ll cover some of these later on in the course of this show—but really, I think the key takeaway that we are seeking to highlight in this document is AI is not going to leave any industry untouched. It will only widen the gap between the winners and the losers, but the path from here to the sort of metaphorical sunlit uplands promised by AI is not going to be linear, just like it has never been linear for any other general-purpose technology in the past.
So let’s start then with data growth and the opportunity for AI. The amount of data produced globally is growing exponentially, with projections showing that data creation in 2020 represents just about 1% of what’s expected by 2030, and most of it will be unstructured. How do you see AI models evolving to handle this data deluge? And what are the biggest opportunities and challenges in parsing unstructured data at scale?
There’s a lot to unpack in that question, Andrew, without a doubt. And I think really the way I would begin to answer this is go back to the very, very first piece of thematic work we ever wrote. We called that, in fact, The Data Deluge, and we argued that the amount of data produced and consumed is only going to grow exponentially. However, that data, or those data, will have zero value unless you secure them, you store them, and you analyze them. That creates a whole range of investment opportunities. And again, to reiterate, what AI has done has really been to turbocharge this dynamic—to put it on steroids. So if we go to the beginning of your question, it really does not matter how good your algorithm is unless you actually have that data in the first place. And the businesses that will be most successful, that will be able to get most out of their data, will be able to use AI to leverage those data insights most effectively, will actually begin with high-quality datasets and datasets which, even if they’re unstructured today, with the help of algorithmic support can actually make them more structured going forward. So we think it is about the quality of the data—number one—that will actually dictate how we see this whole dynamic unfolding. I would add to this that, you know, when we think about the sort of modern technology ecosystem, if you will, clearly cloud compute is one part of the equation; AI is the second part of the equation; and I think it’s a theme we haven’t covered so much in this year’s compendium, but we’ve discussed it on this program in the past—we’ve written about it previously too. But really, the third pillar, I would argue—we would argue—of modern computing is quantum. And quantum, just to remind you, to remind all of our listeners, is really this advanced model for computing, whereas rather than just having zeros and ones in a sort of traditional binary world, you can be zero and one at the same time, to put it very, very simplistically. And we would argue that quantum really represents this way of accelerating everything, and it can help actually deal with some of these unstructured data challenges.
Alex, US data centers are projected to consume more than 10% of the nation’s total power by 2030, and fast forward two weeks, that’s only four years away. What are the biggest challenges and opportunities to securing reliable energy sources for these factories of the 21st century? How do you see the energy mix evolving to meet this demand?
I’m really glad you asked that question, Andrew, because I think it’s absolutely critical, and really pretty much everyone we speak to in our network would suggest that this is really the key bottleneck. You can talk as much as you would like about AI, how it’s going to change and revolutionize the world, but if you don’t have the power to support it, then really the rest of the debate is irrelevant. And there are probably two quotes that really spring to mind that I’ve heard from chief executives recently in this context. One described the sort of land grab or power grab for data a bit like being akin to Game of Thrones—you know, everyone is trying to reach that metaphorical land, iron throne, having that secured source of data. The second, which I think is perhaps more interesting and more nuanced and really captures the essence of your question, is that the debate—and step back from politics for a second—is not an either/or debate; it is an and debate. And what that means very simply is the following: it is not a discussion in terms of how you power your data centers, how you secure access to that power to drive AI.
It’s not this lovely decision where you’d say, okay, I want to have a little bit of liquified natural gas or combined-cycle gas; I want to have a bit of nuclear; maybe I won’t worry so much about solar. You really need every single input you can possibly have. And just like any asset allocator really understands the benefits of diversification, any utility provider or EPC that is spinning up a data center at the moment will really understand this idea of having multiple input sources. We may come on to talk about energy a bit more later in this program, but I think it’s also important for you, for your listeners, to differentiate between those categorical sources of baseload power—so that’s typically going to be your gas, your nuclear; we think geothermal in the future; we’ve written about that in the course of this year—and then you’ve got the other more variable sources of power, which would include wind, would include solar, possibly hydro too.
Given that much of the US and European grid infrastructure is decades old, what innovations or investments do you believe are most critical to modernizing transmission and distribution systems? How urgent is the transformation, and what role should public and private sectors play in accelerating it?
The easiest part of your question to answer, Andrew, is the question about urgency, and the simple answer is: very urgent. The reason why it’s very urgent is as follows. Basically, most grid infrastructure in the United States, in Western Europe, in other developed nations, was last built after—or last properly upgraded, if you’ll—sorry, after the Second World War. So let’s just step back from AI for a second and think about what’s happened. We’ve had huge population growth, we’ve had huge growth in disposable income, and really it is not about asking ChatGPT or any other LLM a clever question, but it is simply every time anyone wants to watch a Netflix video, wants to order a new pair of sneakers, whatever it might be. This is sending bits of data over networks that were not fit for purpose.
Then let’s consider a couple of other dynamics. Nearly everyone would accept that global warming is a massive issue, and that has had an impact on grid infrastructure. If you have your pylons, your pipes, whatever it might be exposed to more extreme weather conditions, just common sense would dictate they’re going to erode and need repairing. The other dynamic—and again, this comes back to the point about securing data—is a lot of grid infrastructure was really not built for a cyberattack era, if you will. So in other words, just needing to put in place the protocols to make that grid infrastructure secure is becoming more and more important.
So I guess if you pull it all together, we’re talking—and this is not my figure, I think it’s from the World Economic Forum—we’re talking about a trillion dollars of investment needed globally over the next decade. That’s a massive sum. I don’t necessarily have an insight into how that’s gonna split out public versus private, but I think, again, to reframe the observation I made to your prior question, it’s not really an either/or debate; it is an and discussion. Every company, every country is going to want to diversify. There will be a logic for energy independence. There will also be a logic for decarbonization wherever possible.
With 95% of the technologies needed for full transition to water, wind, solar, and storage already commercially available, what are the biggest barriers—financial, regulatory, or social—to achieving large-scale decarbonization? And how realistic is the goal of meeting climate targets through electrification and renewables by 2030?
I think there are a lot of moving parts, Andrew, and we’ve touched on some of them already. Really, again, the debate comes back to this idea of embracing every possible energy source. That’s our opinion. Having a balanced portfolio, recognizing what can provide your baseload, what can provide your intermittent solutions. And I think really, to answer your question directly, for carbon-neutral or decarbonization solutions to become fully mainstream, it’s easy to say, look, the cost is coming down. We’ve seen a traditional decline in costs for solar, like we’ve seen, say, for semiconductor chips or the cost of cloud compute—it follows a very natural curve, and with scale you’d expect this only to fall lower.
Most accounts would suggest that if you look at an apples-to-apples comparison, if you will, the cost of solar today—or if not today, then by the end of this decade—will make it effectively the cheapest source of energy on this planet. That’s lovely. But the challenge, as you know, as your listeners know probably, is that that’s great, but the sun doesn’t always shine. It’s pouring with rain in London today as we speak. And you really need effective storage solutions.
As we said, solar is going to be one part of the equation. We think geothermal actually deserves much more attention. And really, the key issues are political. To answer your question again: political buy-in, energy independence. I think every country is aware of more unstable geopolitics, and against that background, if you can build solar farms, that’s very easy in the United States; if you can drill for geothermal, that’s very easy in the United States, especially with all of the expertise within fracking. You’ve got natural gas under the ground, you’ve got oil under the ground; even wind—perhaps no one in the Trump administration will say this publicly—but even wind will actually play a role. And I think you’re going to see very similar dynamics in most other countries.
Let’s turn to some of the AI adoption and challenges of adopting them. So AI adoption is accelerating and it’s delivering measurable ROI across industries. Yet issues like cost, privacy, trust, and cybersecurity remain significant challenges. What are the biggest hurdles that organizations face in scaling AI responsibly, and how can they balance rapid innovation with these critical concerns?
Well, one of the most interesting data points I’ll share with you, Andrew—and we’ve talked on prior shows again about cybersecurity—is that about 90% of cyber incidents, full stop, are caused as a function of human error. And against that background, in that context, I think every business, they understand on paper the case for AI. Even if they don’t fully understand the case for AI, there is a fear of missing out, there is a fear of becoming obsolete, there is a degree of sort of CEO or senior management signaling to suggest to all stakeholders that they’re not falling behind. So that’s the background.
The reality is that practically implementing AI is quite different to talking about it. You’ve obviously got to get board sign-off and internal buy-in. Many people are risk-averse. Many people want to keep their data siloed, perhaps for obvious reasons. They don’t want it leaking out onto the wider web. They don’t want models like ChatGPT or Claude or any of the others using potentially confidential internal data. And there is the danger that without proper employee training, you will see human error creep in. These data sets may be misused; you may have the cybersecurity problems that I began this particular discussion with—you may have those issues arising.
So I think, you know, the expression we often use is that everyone’s familiar with this idea of ChatGPT, but rather than GPT standing for “generative pre-trained transformer,” we need to think of AI as being a “general purpose technology.” So in that sense, it’s just like electricity, it’s just like the internet. These things are going to go on to change the world. We will use them in ways that, you know, you and I cannot conceive of—probably many younger listeners can’t even conceive of as well. But the path from today to that AI future is not going to be linear. There are going to be a lot of hiccups along the way, and risk aversion, the other factors I called out, are going to certainly play a role.
Alex, let’s talk about how robotics is transforming the workforce. Robot density has more than doubled in the past seven years, and companies like Amazon have deployed hundreds of thousands of robots. How do you see the convergence of AI and robotics reshaping industries such as manufacturing and logistics? And another important element to this is: what lessons can the US learn from leaders like South Korea and China in accelerating this adoption?
I think the simplest answer to that set of questions, Andrew, is pragmatism. And what I mean by that is the following. The reality is that we are globally getting older. That’s an inescapable demographic phenomenon. And so the number of people aged 60 in the world will, I believe, if my memory serves me correctly, double between 2015 and 2050—we’ll be at about 2 billion. So, a quarter of the world’s population over 60 by 2050. And this obviously has several implications. It means that there’ll be more and more people dropping out of the workforce. There’ll also be an increasing burden of people that need to be cared for, and robots can provide part of the solution.
One of the reasons, as you noted, why Southeast Asia has often been a lead is because those nations, in contrast, say, to Europe or to the United States, have been less reluctant to embrace immigration. So when you’ve got a declining replacement ratio—demographics means the future’s already happened—they’ve had to adopt robotics more quickly.
The second really important factor, and again, if you listen to Jensen at Nvidia, if you listen to many other people far brighter than me, then they will basically say that, look, AI today as we know it is typing a command into a computer and getting a clever answer. The next phase of AI, however, is going to be what many people call physical or embodied AI. So it’s effectively taking all of the traditional hardware that has made robots very good at sort of banging car doors together or assembling iPhones—the traditional industries where you’ve seen them very heavily deployed—and actually you are taking robots to a much greater level. The term—we discussed this on a prior program as well—is humanoid robots. So robots will appear, resemble humans much more, both in their sort of physical manifestations and their ability to carry out a range of tasks.
So I think, you know, the way to look at it is: today, a humanoid robot is going to be more expensive and less efficient than it will be at any time ever in the future. And if we believe that, just like, you know, three years ago we were on GPT-3.0, today we are on GPT-5.0, the quality—the reliability, if you will—of robots will only improve from here. And if you think about it from a very simple ROI point of view, you know, robots, unlike humans, can work 24/7. They can even recharge with a cable plugged in whilst they’re doing something. They don’t take sick days. They don’t want to go on vacation. They don’t need pensions—at least not yet. So this is really the future we’re evolving towards.
Sticking with robotics, let’s have a look at AI in healthcare. Robotic-assisted surgery is now present in about half of US hospitals, but is still used in fewer than 10% of global procedures. What do you see as being the main barriers to wider adoption, and how might falling costs and expanding use cases change the landscape of surgical care in the next decade?
We’re massive fans of robotic-assisted surgery, Andrew. We’ve been following this theme for about a decade. To provide you and your listeners with the context, these robots have been around for about 30 years by now, and it’s really not a case—just to be clear—of a robot physically wielding a scalpel and operating autonomously. I was lucky enough to be out in Sunnyvale, California not so long ago and to actually see one of these systems in action. And effectively, a surgeon will sit behind a computer terminal with a joystick or a set of controls and manipulate a robotic arm.
The argument is that obviously if you do this, your precision will be much greater. Particularly with AI, you can learn from prior procedures exactly at which angle, say, to cut into the body, how much pressure to apply, how deeply to cut, and so on. And the theory goes—and at least this has been backed up by about 40,000, over 40,000 academic white papers by now, peer-reviewed papers—is that the benefits of robotic-assisted surgery outweigh the costs.
In other words, yes, you as a hospital have a large upfront capital investment to make. These machines typically have a list price of one and a half million dollars and upwards. You’ve got to buy the consumables each time. But your operating leverage from that initial machine—the ROI on hospital space, the fact that patients typically spend less time in a hospital bed recovering because the operation has been performed more efficiently—all of that makes a very strong case for robotic-assisted surgery.
So to answer the question: what are the barriers to adoption? Look, I’m a finance person; I don’t work in a hospital. But the reality is cost is typically the main factor. As we’re aware, the US administration has been a bit more thoughtful about how they’re allocating dollars across the whole healthcare system. Many public administrations are under pressure. There is also, to an extent—and to come back to the Darwinian point with which we began this whole conversation—there is sometimes a reluctance to embrace change. But from an economic point of view, from an academic-paper health-benefit point of view, we see the path forward for robotic-assisted surgery only pointing in an upward direction.
Alex, final question. Where can listeners read more? Where can they access the compendium?
Alex Gunz
Well, as mentioned, Andrew, we’ve been writing about and investing in future trends since at least 2016. All of our white papers, all the materials about what we do, can be found on our website: www.heptagoncapital.com.
Brilliant. Thank you for being an excellent guest again. It’s always a pleasure, Alex.
Delighted to have the opportunity to discuss these topics, Andrew, and we’ll be writing several more white papers in 2026 and beyond.
Excellent. Alex is the portfolio manager at Hetic and Capital in London. Thanks for joining me, Alex, and if you enjoyed today’s episode, please subscribe wherever you download your podcasts from. Bye for now.
The document is provided for information purposes only and does not constitute investment advice or any recommendation to buy, or sell or otherwise transact in any investments. The document is not intended to be construed as investment research. The contents of this document are based upon sources of information which Heptagon Capital believes to be reliable. However, except to the extent required by applicable law or regulations, no guarantee, warranty or representation (express or implied) is given as to the accuracy or completeness of this document or its contents and, Heptagon Capital, its affiliate companies and its members, officers, employees, agents and advisors do not accept any liability or responsibility in respect of the information or any views expressed herein. Opinions expressed whether in general or in both on the performance of individual investments and in a wider economic context represent the views of the contributor at the time of preparation. Where this document provides forward-looking statements which are based on relevant reports, current opinions, expectations and projections, actual results could differ materially from those anticipated in such statements. All opinions and estimates included in the document are subject to change without notice and Heptagon Capital is under no obligation to update or revise information contained in the document. Furthermore, Heptagon Capital disclaims any liability for any loss, damage, costs or expenses (including direct, indirect, special and consequential) howsoever arising which any person may suffer or incur as a result of viewing or utilising any information included in this document.
The document is protected by copyright. The use of any trademarks and logos displayed in the document without Heptagon Capital’s prior written consent is strictly prohibited. Information in the document must not be published or redistributed without Heptagon Capital’s prior written consent.
Heptagon Capital LLP, 63 Brook Street, Mayfair, London W1K 4HS
tel +44 20 7070 1800
fax +44 20 7070 1881
email london@heptagon-capital.com
Partnership No: OC307355 Registered in England and Wales Authorised & Regulated by the Financial Conduct Authority
The analysis in this material is provided for information only and is not and should not be construed as an offer to sell or the solicitation of an offer to buy any security. To the extent that this material discusses general market activity, industry or sector trends or other broad-based economic or political conditions, it should not be construed as research or investment advice. To the extent that it includes references to specific securities, commodities, currencies, or other instruments, those references do not constitute a recommendation by IBKR to buy, sell or hold such investments. This material does not and is not intended to take into account the particular financial conditions, investment objectives or requirements of individual customers. Before acting on this material, you should consider whether it is suitable for your particular circumstances and, as necessary, seek professional advice.
The views and opinions expressed herein are those of the author and do not necessarily reflect the views of Interactive Brokers, its affiliates, or its employees.
Join The Conversation
For specific platform feedback and suggestions, please submit it directly to our team using these instructions.
If you have an account-specific question or concern, please reach out to Client Services.
We encourage you to look through our FAQs before posting. Your question may already be covered!