Azeem Azhar sets out his vision for 2022 and shares the trends he thinks will change our world this year. Some changes are hurtling down the pike, like the growth of Web 3.0 and the metaverse, while others are rumbling along more slowly: the continued dominance of Big Tech, and the tireless march of artificial intelligence.
Azeem’s discussion touches on:
- Why we should differentiate between “tech crypto” and “money crypto.”
- Why the metaverse is so much more than Meta.
- Why the “climate tech” tag might become obsolete.
@Azeem
@exponentialview
Further resources
Exponential View Newsletter
Is Three the Magic Number: My Take on Web3 (Exponential View Newsletter, 2021)
Learning From 2021: Azeem’s Takeaways (Exponential View Podcast, 2021)
AZEEM AZHAR: Welcome to the Exponential View podcast. I’m your host, Azeem Azhar, and let’s get the year started. First of all, Happy New Year, happy 2022, and I hope the next 12 months are spectacular for you. Now, in this special edition of the podcast, I am going to talk about six interesting themes that I think are going to shape not just the coming 12 months, but beyond. Now, many of you who read my newsletter will know that every couple of years, I put out some predictions with a slightly longer duration. I think predicting what will happen in the next 12 months is a hard exercise. It’s akin to betting on which horse is going to win the race. But what we can do is say useful things of some of the underlying and fundamental themes. Now, my themes for 2030 still hold together. You can find those themes if you go back to the newsletter archive at exponentialview.co. I also identify some key mega trends in my book, Exponential, or if you’re in the U.S. and Canada, it’s called The Exponential Age.
Now, out of all of this, I have identified six things that I think will become important over the next few years and we will be able to notice during 2022. Number one, I think we’re just going to hear a lot more from technologies that a few years ago seemed incredibly science fiction and futuristic. I’ll just pick up on a few of these. Quantum computing is an area that has been waiting in the wings for a while. There was a surge in funding over the last two or three years in quantum computing startups. And we are really making progress, whether it is smaller companies like IonQ, or Rigetti, or PsiQuantum, or whether it’s the more store walk businesses like IBM, or even Honeywell, or Google Teams who are building quantum computers have been at it for the last eight or nine years. And as I’ve spoken to them over the last year or so, I’m getting signals of more and more confidence that larger and larger, more practical quantum computers will start to shift. And I think in 2022, we will see some useful milestones.
But the other thing I think that we are starting to benefit from is work within the efficiency of the algorithms and the software that works with these early quantum computers. And the combination of the computers getting a little bit better, the algorithms getting more efficient has peaked the interest of industry and those quantum computing teams that are sitting in the financial services business, the pharmaceutical businesses, the material sciences businesses. I think we’ll start to show some early results in this intermediate period of quantum computing development. We’re going to be talking about quantum computers a lot over the next six or seven years, but 2022 I think will be another year where there will be, forgive the pun, a quantum leap and the amount of discussion we have about this topic. The other area of futuristic technology I think is fascinating is in fusion, the ability to harness power from smushing together lighter atoms and on a similar timeframe to what’s happened in quantum computing.
There’s been a surge in investor interest infusion based on the fact that there have been real milestones. And there are increasingly plans that look like they embed more engineering risk than science risk. In 2022, I don’t think we will naturally see operating fusion reactors connecting to the grid. I’d be very surprised if we do. But I think what we will see will be increased funding and larger scale prototypes and some meaningful engineering developments on that path to getting fusion reactors connected to the grid within the next five to seven years. But this theme of futuristic technology is not just about quantum computing infusion as if that wasn’t enough, we’re starting to see a real growth in the investment and development of innovations out in space. What started over the last few years with the low hanging fruit as it were launching earth observation satellites that were much, much lighter weight and smaller to the CubeSat standard, now looking into more sophisticated types of projects from energy generation to micro manufacturing in space.
And as a sector, it’s drawing quite a lot of interest. Quantum computing, fusion, space technologies, and it doesn’t stop there. I’ll leave one that I was a bit skeptical about a few years ago that has surprised me in the pace of development and interest, and that is electric planes. We are starting to see a whole range of new companies tackle the idea of aviation powered by batteries. Some it’s very, very new form factors, I think about the taxi-sized, Milium, autonomous electric planes through to companies making much larger scale passenger jet-size, electric planes, more trials, maybe some semi-commercial trials, but definitely more investment and more company formation and growth in that sector as well. And it’s quite exciting to look forward and say, “Well where we expect these breakthroughs to happen, where we expect the news to shift, where we expect the interest to move to is in what I can only call futuristic technology.” And it just seems so much deeper than the last mobile app.
Number two is about climate tech, where it’s come from, most importantly, where it’s going. Climate tech has been booming. I did some work with PWC over the last couple of years to track the sector in the thousands of companies in the sector between 2020 and 2021, there was a 210% growth in investment in private companies looking to tackle the issues around climate change, whether it’s direct decarbonization to producing alternative industrial and business processes, to the ones that we currently use processes that perhaps might be zero carbon, whether it is cement or meats or foods or materials. Climate tech now represents about 14% of all VC investments. And it’s an area that I’ve been tracking really since about 2017 when I noticed founders talking about climate change. But here’s my thinking, my thinking is that 14% is going to rise and rise and rise. And at some point, the idea of climate tech will disappear. And it’ll disappear in the same way that the idea of the dotcom disappeared. A couple of years ago, I chatted to Andrew Beebe, an investor at Obvious Ventures in Silicon Valley.
And like me, Andrew was present during the dotcom bubble. And Andrew was the one who flowed to this point, which was it was a moment in time where every company got an excess valuation by saying it was a dotcom company. And then dotcom became everything that you did and now, no one talks about dotcom companies. And Andrew’s argument is that climate change is going to become so important. It’s going to become important to customers, whether those customers are businesses or consumers, and it’s going to become important to governments. And so policy, taxation, regulation will demand companies have climate and net zero at their core. And so as a result, every company, certainly every startup will need to become, look like and articulate how it is a climate tech firm. The idea there is not of course that this loses its importance, it becomes more and more important. But rather it actually redefines and redesigns what it is to be a company over the coming years. Number three is the continued acceleration of AI, artificial intelligence as a technology.
Now, we’re about a decade in to the current AI wave that started back in 2011 with breakthroughs in building certain types of deep learning systems. And 2021 last year was really tremendous in terms of how research frontiers move forward. But I think we’re just starting and we’re just starting I think for three key reasons. The first reason is that the large companies who have dominated a lot of the development of AI have figured out that there are models that work and that work if you scale them and get them bigger and bigger. They did this with deep neural networks, they’re going to start to do this now with a set of technologies called transformers which were first developed in 2017. And transformers started as machine learning models that were really good at dealing with human text, natural language, with all other shortcomings. But it turns out that transformers as they get more complex, so they get trained with more data, they can generalize into other areas. They can move from natural language into maths and protein sequencing and images and video prediction.
And we’re only scratching the surface right now from an industrial perspective. And what tends to happen within these industrial contexts is that when we figure out an approach that works, we’ve explored the terrain and we figure out this particular approach works, industry [inaudible 00:09:30] lots of resources into it. That’s certainly what industry did when it came to deep neural networks and their ability to deal with human spoken speech and images and so on. And I think the same thing will happen as we pick off this new domain that transformers allow for. It ends up being a bit of a slam dunk investment and so companies will make that investment really, really extensively. But there’s a second aspect to what happens from the industrial perspective, which is outside of the really, really large companies. And that is that the rest of business where most of the world’s GDP lives, where most of the world’s employees sit, where most of the products and services we buy come from, are finally starting to cotton onto how to use AI systems within their operations.
So these aren’t exponential age firms, these are companies of the last 50 or 100 years. They are industrial age firms. And it has taken them time to figure out how to make use of this technology. And we’ve seen that in previous technology waves that the new entrants figure out what to do rather quickly and the incumbents take a bit longer because they have lots more change, much more legacy in terms of their investments and their managerial thinking to contend with. But we are a decade in to the AI wave, that means that there are enough people who might be mid-level or senior level in incumbent companies who understand or appreciate the technology. These companies would’ve run experiments and they will be getting more confident with how to use them. So outside of the [inaudible 00:11:09], the biggest American technology companies and the cadre of smaller dynamic exponential firms that are doing really well, I think we will also see very, very broad based investment in these types of technologies across other sectors of industry.
But then the third area is where research is going. And in a number of the conversations I’ve had both on the podcast and private discussions I’ve had with researchers in the field, there is also a recognition that the approaches that we are currently doubling down on industrial context, these deep neural networks, or these transformer models may not be the only approaches that we need that we might have to take, have other scientific breakthroughs or theoretical breakthroughs in order to get AI systems up to whatever the next level looks like. So for all the power of these transformers and where they might take us to the next couple of years, it’s by no means clear that that theoretical approach as is then being built out gets us all the way to the potential of what AI might be able to offer. So I would still expect there to be significant important research developments, new theories, new implementations that help raise the energy level, help raise our expectations of what it is to build more generalized, safer, more accountable, more explainable, more powerful AI systems.
Number four is big tech will stay big in 2022 and beyond. Now, I write in my book about superstar companies that in the exponential age, we are able to develop companies that get much, much bigger than firms that we’ve seen traditionally in the past. And with that comes the extra power that they leverage. The reason I think that they continue to stay big is because frankly, even as trillion dollar or $2 trillion businesses, they’re still growing at 20, 30% per annum. Their franchises in some parts look a bit tapped out, but by and large don’t look tapped out, look like they’re going to continue to grow. And they have a ton of momentum. My sense would be that they’d continue to stay large. What could slow them down? What could stop them in the next two or three years? Well you won’t get regulatory action in a couple of years that’s significant enough. And even if you did, would it really be deep enough to tackle the size of these companies? Imagine in this nuclear scenario that a powerful regulator determines that a company needs to be broken up.
Maybe they might argue that one needs to get broken up, but there are four or five huge technology companies, maybe six. It’s inconceivable that that rule would apply to all of them. And it’s partly why my recommendations are not so much a specific argument that these companies should be broken up, rather that they should be reviewed that their license to operate should be analyzed. And we should find the parts of their operations that look like they are essential to us as citizens and on our requirement to live in a modern digital world. And those parts of operations for big tech companies should come under greater scrutiny. Whereas other parts, they can run like aggressive, fast growing businesses. So there’s a dynamism and there are different qualities to these companies and they need to be treated differently. My sense is that big tech will continue to stay big. That isn’t to say we won’t see new trillion dollar companies enter the list of the world’s largest companies. Tesla basically joined those ranks. And you could imagine there being many other really significant companies.
But I suspect that big tech, as we know it today, will continue to stay big. Number five, crypto will continue to divide opinions, but I think we’ll start to see some real progress in what we might think of as tech crypto. So there’s an investor called Kyle Samani who I’ve spoken with in the past who set a really interesting analysis. He’s a crypto investor. And Kyle’s argument was that there were two different types of people in the world of Web 3.0 cryptocurrencies, blockchains, money crypto and tech crypto. And I thought it was a really useful insight because crypto is one of the most divisive subjects in technology that I’ve come across in 25 years. And I think part of the reason is that the money crypto stuff is really the one that captures the headlines, right? It’s all about the asset prices. It’s all about the value locked in. It’s all about the price of Bitcoin and it’s volatility. And I think what it does is it deflects us from what’s really happening within tech crypto.
So tech crypto is that bit of crypto where really smart developers have gone off and taken this really nascent technology that can do some things quite well and has a lot of limitations like any other technology. The things that crypto does well really relate to the ability at a software level to have systems that can be trusted without trusting any of the participants within them. And this is a bit of crypto that I find particularly interesting, because it talks to the applications that can be built that are actually useful. Now, with any new developer platform, it takes a while for there to be useful applications because you have to get the platform mature enough, and then the components within that mature enough and people have to figure out how to write to it. You have to recruit and train developers. And all of that is quite far away from the reality of, “Well I just want to use my computer to listen to some music,” and it takes a while to get there. But we have had a while here.
And so we are now starting to see interesting applications being built on this platform. And one I talk about quite often is Helium. Helium is a crypto-based bottom up, global, low power data network for IOT devices, for things like GPS packages and sensor modules in your farms and fields. And normally, it would be really expensive to build a global data network. What Helium does is it uses a blockchain mechanism to incentivize individual wifi hotspots to connect to the Helium network and provide service to these IOT devices. And Helium I think for me is one of the examples of the thing that you can build in this tech crypto land that makes it distinct and different to what we could build before. So in 2022, what I would expect to see in not just ’22, in the years to come is more development of these core fundamentals, these technologies, these services that are built on the underlying principles of what blockchains allow you to do that are hard to do with other technologies. Now, I don’t think this is going to stop the divisiveness that exists around the discussion of Web 3.0.
But I think it is something to take a look at, right? To peer below the hullabaloo and the smoke and the noise, and to say what’s really going on? And is there a developer ecosystem that comprises developers who are building interesting applications with users who are getting some utility from them and some economics that runs between those two? Because those are the underpinnings of a healthy ecosystem. And if the ecosystem has its own health and ability to self-sustain and then to grow, out of that might finally come some useful applications. So number six, the growing importance of the metaverse. Now, let me start with a confession, the metaverse, which wraps up some ideas around virtual reality and 3D worlds, is not something that I have really got to grips with historically. I mean I’ve been looking at virtual reality since about 1995. I tracked VRML, the virtual reality markup language, which was an attempt to build some standard for defining 3D worlds at the time that HTML was being developed and maturing.
And I played around with those early headsets then. In 2006, 2007 when Second Life was emerging and massively multiplayer online games like EverQuest were becoming popular, I ran teams that looked into those areas and I put the first Reuters journalists in Second Life back more than 15 years ago. A decade after that when VR headsets from HTC and Samsung and Oculus were starting to be sold to consumers, again, I had a small team that looked into that, looked at product roadmaps and technology developments and content creation tools and publishers and what could be done. And in 2021, I got my hands on an Oculus Quest 2 and played with it for a while, particularly when I had an injured knee and couldn’t run. So I used the VR boxing game to keep fit for a few weeks during my recovery process. And yet, throughout all of this, I personally didn’t feel a pull towards metaverse experiences. However, data speaks louder than personal prejudices and intuition. And one of the things that has certainly emerged and looks like it’s going to continue to emerge is really, really deep and broad interest in the metaverse.
Now, it’s not just about Facebook. It is also about Disney. It is also about enterprises, the Accentures of this world. It is about platforms like Unity. Unity makes the leading games platform for third party developers with mobile applications. And they’re great with 3D physics engines and being able to develop these experiences. Unity has a platform that lends itself to creating tools for metaverse development. So a lot of companies, a lot of founders are looking at the metaverse as the next computing platform. Now, they may get it wrong. But again, as I said, data has to speak louder than intuition at this point. And there is some quite interesting data. In 2021, Oculus sold more headsets than Microsoft sold Xboxes. In fact, the number of Oculus headsets sold was about 8.1 million. The Xbox sold about eight million. Now, that doesn’t make the Oculus the biggest gaming platform by any means. You’ve got the PlayStation and you’ve got Nintendo as well, but it’s an important milestone and the growth rate is pretty unimpeachable.
On top of that, the amount of time that people spend on those devices is also growing. So I think that there is some underlying momentum between consumer preferences, between the technologies getting better, between companies staking their claim that says there is going to be development in the metaverse. There’s also that intersection between the NFT space, non-fungible tokens, which ultimately are at some point these digital products that are perhaps best experienced within an augmented reality, extended reality or virtual reality experience than looking at them on a webpage. And NFTs of course bring with them a bunch of economics. I mean we can disagree with how that pricing works, but we can’t avoid the fact that there are economics attached. And once you start to get that intersection between developers of different types, experimenting in different ways, users demonstrating voluntarily that they are interested in this, they’re spending more time in there and there being an economic connection between those, you start to get the hallmarks of an ecosystem that might make sense.
So I didn’t really talk about AR or VR or the metaverse in previous years, March, I’ve done a little bit of analysis. But I think that this is a theme that has emerged and will continue to develop over the coming years. So that’s it for my trends of 2022, and I’m really glad to have shared them with you. Of course, stay tuned because the podcast is going to be a year’s worth of challenging, fascinating, surprising and interesting delight. And we’ve got so much coming up. And I would really love for you to hear those ideas. And please do take a moment to tell your friends about the podcast, forward this onto them, give us a five star review on your podcast platform of choice, that would make us all very happy for the start of the year. A Happy New Year to all of you. This podcast is produced by Mischa Frankl-Duval, Frederic Casella and Marija Gavrilov. Bojan Sabioncello is the sound editor, and it’s a production of Eπi Plus One, Ltd.