Updated: December 2022
In this note, our goal is to zoom out and analyze the key long-term trends shaping our future. We first talk about the triangle: Intelligence — Technology — Energy, which we believe has been powering humanity's history and identify 5 trends shaping our near-term future: Energy and minerals scarcity, Climate change, Fracturing globalization, Aging population/falling fertility rates, and Enabling technologies. The last one is covered only partially, based on the focus of our research, and includes the Digital world, Space, AI, XR, and Public blockchain. We focus on identifying technology shifts that are likely to last for at least the next 10 years until reaching maturity and becoming commonplace like plastics or smartphones are today. We also discuss the framework of high/low-energy tech and its implications for the future of technology.
We try to follow Occam’s razor principle. It is easy to come up with a hundred trends. It is much harder, but more valuable to us to uncover the core few that underpin everything else. It helps us organize our thinking about the future and shape our investment approach.
This is a work in progress that we continue to update over time.
We think about human history as an interplay between energy, technology, and intelligence (it may actually be applicable to the universe’s history as well — but this is a topic for a different time).
This is how this triangle works. We use intelligence + technology to unlock access to more energy — starting with the stick, fire, agriculture, and all the way to nuclear reactors, fracking, solar panels, etc. We use energy + technology to grow intelligence — first, society can free people to do research and allocate resources for that, then we develop information technologies (writing, printing, internet) to spread knowledge and enable collaboration. Our growing intelligence allows us to develop new technologies to unlock more energy.
There are three groups of effects driven by this dynamic.
The Direct group has to do with the dynamic itself. We experience it as (hopefully) improving levels of education and growing opportunities to learn and contribute, a never-ending stream of technology innovations, and an increasing number of machines that use energy. For context on the last one — our hunter-gatherer ancestors used to live on ~3MWh of energy annually, every US resident today consumes almost 30x that amount in the form of electricity, heat, transportation, and all the energy used to grow the food and manufacture the products. Multiply the growing population by the higher levels of education and account for the increasing connectivity and access to existing knowledge (internet) and you get exponentially more intelligence at humanity’s disposal.
The Quality of life group is driven by us using our increasing intelligence, advancing technology, and growing energy supply to improve our quality of life. From food security, medicine, and basic shelter to luxury cars and lavish vacations.
The last group is the unintended consequences of the first two. It includes deteriorating ecology, climate change, falling fertility rates, and energy shortages, because of dwindling supply. It also includes weapons that can wipe us out if we’re not cautious enough.
Below we will go through each of the trends one by one. But before we do, a few additional notes:
When saying “technology” people often assume things like smartphones, the internet, airplanes, plastics, nuclear energy, or mRNA vaccine and gene therapy. But there is also social technology that deals with how we organize ourselves: social orders, laws, religions, non-profits, labor unions, etc. The term technology in the triangle above includes social technology as well.
Adversity plays an interesting role in the dynamic described above. The triangle will work its magic regardless of the presence of adversity. One can even argue that over the last few thousand years, humanity as a whole was in fact experiencing a relatively benign environment (save for a few epidemics) where our biggest threat was we ourselves. But adversity can play a very important role in protecting humanity from future threats. It’s well understood that a never-ending stretch of wars between European states enabled them to advance dramatically both in civil and defense capabilities. Benign environments can provide a significant short-term boost to the triangle’s dynamic (see the next point) but at the risk of becoming ill-prepared for future threats.
We witnessed a massive acceleration over the last 75 years driven by globalization and fossil fuels. Using energy as a metric, while global energy consumption grew 30x since 1800 more than 85% of that growth happened since 1950. Over the last 70 years energy consumption grew twice as fast as over the previous 150. This acceleration exacerbated the unintended consequences mentioned above and made humanity’s situation quite fragile on many fronts, climate change and energy especially.
The triad above is powered by an ever-growing flow of energy. A few short-term hiccups now and then are fine, as long as the general trend is up. If the energy trend reverses over a meaningful timeframe (decade) the triangle will start unwinding. The declining energy flows will force us to allocate less energy to growing/maintaining our combined intelligence, some technologies will become prohibitively expensive and will have to be abandoned, and fights over the declining energy “pie” will disrupt everything even more. Note that for this dynamic to emerge we don’t need to reach the point of exhaustion of our energy sources (like oil/gas/coal reserves that currently power 80%+ of our global energy consumption). We don’t even have to reach a point, where we traversed the peak and now moving downhill. It will suffice that we realize that we are soon to cross the peak, for us to start acting like we already have.
And now to the trends.
Energy and minerals — limited supply and growing costs of extraction
The energy sources that we use have been a major factor in the rapid rise of living standards and population growth over the past 200 years. But we live on borrowed energy.
Starting from the nineteenth century we binged on fossil fuels that powered our exponential growth from left to right.
Our energy consumption grew 30 times driven mostly by coal, oil, and natural gas. The population grew 8 times and the world GDP increased 100 times. 80%+ of the energy we use today to support our living standards comes from fossil fuels.
Fossil fuels have a finite supply. Nature generated this resource over millions of years, and we are on track to deplete it in a few centuries. While there are ample reserves of fossil fuels, it’s only a part of the equation. We need to be able to extract these resources efficiently so that the energy we spend on extraction, processing, and transportation is a small fraction of the energy we produce from them (the eROI is high).
Naturally, as we were tapping into this source of energy we started with the low-hanging fruits. As we progressed we had to drill deeper and in less hospitable environments. For a while technology advancements compensated for this, but it looks like we are past that stage now.
Fossil fuels extraction costs are increasing and we expect them to continue to increase over time.
The situation is similar with many minerals. We tapped the best sources first and now we have to spend more energy per unit of product.
One counterargument here would be that new technologies will emerge in response that will make extraction more efficient. For minerals, one such technology is recycling, which will become more cost-competitive as the extraction costs go up.
As for fossil fuels, the need to curb greenhouse gas emissions will stop us from significant investments in innovation there, thus our goal is to improve alternative sources. We eagerly look forward to advances in solar, wind, geothermal, and nuclear (fission/fusion).
The transition will not be smooth. Multiple industries will transform. There are obvious ones, such as transportation and electricity generation, storage, and transfer. However, we cannot stop there. Manufacturing, agriculture, and residential/commercial heating also rely on fossil fuels. In fact, there is hardly an industry not being powered directly or indirectly by them. That’s what happens when you take on the source of energy responsible for 80% of all consumption.
We have a lot of work to do. For context, as far as we can tell, humanity as a whole hardly ever replaced legacy sources of energy. We mostly just added new sources on top . And now we are talking about replacing not some 10%, but 80% of our energy supply, while the demand is still growing. And doing it extremely fast (in a few decades). The scale of the task is so colossal, that no matter how you slice it, it’s clear that it will require a massive concentration of resources, focused will, and global coordination.
The opportunity however is enormous. Our estimates put it at at least tenths of trillions of US dollars in new annual revenue globally by 2040.
The transition is also compelling because of the terrifying consequences of its failure. Not only because of climate change (see the next section) but also because of the civilizational collapse we will face in case of the falling energy supply.
Almost every other trend we discuss below is contingent on the energy flow staying at least flat.
Earth has been getting increasingly hotter over the last 100 years. As far as we can tell, it hasn’t been that hot for the last 2000 years. While there are several factors responsible for the earth heating up before — orbit change, axis position, solar radiation levels, neither of these factors seems to play a role today and there is a broad consensus in the scientific community that the only significant cause we can think of is the growing concentration of the greenhouse gases (GHG) in the atmosphere.
Climate change will lead to significant consequences for the lives of billions of people on the planet. The sea level will increase (by as much as several meters) and claim land, including some of the most fertile parts used for agriculture. More frequent and more severe extreme weather events will make ownership of physical assets more challenging. Water shortages will affect agriculture and water-sensitive infrastructure. Melting Arctic ice will create new opportunities for navigation in that region.
There is a lot of things being done to combat the change and a lot of opportunities for entrepreneurs.
Speaking of ecology in general, we are stressing the planetary boundaries on many fronts, like land-system change, novel entities (especially plastics), and biosphere integrity — see the diagram.
While ecology has been a concern in international negotiations for a while, very little actual progress has been made. This situation seems to start to change recently with the UN Plastic pollution resolution addressing one of the most pressing issues (the “novel entities” sector above).
Fracturing globalization and the renewed arms race
We are not disbanding globalization yet, but we are definitely scaling it back and compartmentalizing. As anyone who had roommates will readily admit, living as one big family is usually easier said than done. It requires incredible levels of trust, alignment on the shared set of values, and maturity to resolve challenges. Way more than we, mankind, currently demonstrate.
The de-globalization is likely to continue with an impact on supply chains, strategic re-shoring/friend-shoring, and intensified industrial policies via tariffs/subsidies to support cutting-edge technologies. The list currently includes chips, non-fossil energy, batteries, aerospace, and AI. It’s likely to include biotech and other frontier techs.
This trend will also act as an additional tailwind for nextgen manufacturing — robotics, additive manufacturing, and other technology improving manufacturing efficiency and reducing labor inputs.
Another consequence of the de-globalization is the renewed arms race. For a set of reasons over the last few decades, we saw a lowered interest in this type of exercise. These reasons included higher levels of perceived security, lack of resources to participate, and desire to avoid being provocative prematurely. With de-globalization proceeding, the first reason will weaken and the third will become less relevant.
We expect the new defense race to proceed alongside the following trends:
- speed — faster decision-making, faster actions, and higher speeds of everything that flies, swims or moves on the surface.
- size — making things smaller
- swarms — enabling operations in coordinated groups, with some autonomy
- support with AI — applying AI to observation, simulation, and other activities supporting decision-making, as well as operational autonomy
- cyber defense (including industrial defense)
- social defense (against manipulations of the population’s views and attitudes)
The aging population, flattening growth, declining fertility rate
Over the last 100 years we got used to the fact that the earth’s population is growing rapidly, but it’s not the case anymore. The growth has been slowing down, driven by the plunge in fertility rates going from 5 birth per woman in 1950 to 2.3 (just a bit above the level of 2.1 required to sustain the population).
In fact in countries that are home to 2/3 of the population, the fertility rate is already below 2.1 all the way to just 1.5 in some cases. Because of that, despite the growing population, the annual number of birth has been roughly flat since the 1990s. The current UN projections call for the population to reach 10.4b during the 2080s and stay flat afterward.
At the same time, thanks to the advances in healthcare, we now live longer, which leads to a demographic shift toward an aging population. The global median age has increased from 21.5 years in 1970 to over 30 years in 2019. It is estimated by the UN that by 2050 the share of people aged 65+ in Europe and North America will reach 27% from the current 19%.
Declining fertility rates lead to a declining labor force. While for some countries, it is a looming threat, for others (like China and Japan) it is already a reality. To keep (or even grow) production with the falling numbers of workers we will have to significantly invest in manufacturing automation.
More elderly people means more opportunities in elderly care and healthcare in general.
The declining number of birth is another major challenge that will have to be addressed sooner than later and is likely to become a massive project on the scale of the energy transition.
This group is the most dynamic of all. As an example, we can today say what the global labor force will look like 15 years from now because all the relevant people have already been born. However, we can’t say anything about what AI capabilities will look like a decade from now.
Still, our goal here is to identify technology shifts that are likely to last for at least the next 10 years, until reaching maturity and becoming commonplace like plastics or smartphones are today.
But first, let’s start with the big picture.
Technology advances seem to oscillate between high/low-energy intensity following the state of the energy supply. Switching to agriculture enabled massive growth in energy supply, leading to technologies for building relatively large cities, bridges, pyramids, and so on. A switch to fossil fuels produced another surge in energy, leading to technologies for trains, cars, trucks, rockets, skyscrapers, and more. These all are high-energy-intensity technologies. But as energy supply growth decelerates , technology innovations shift to the low-energy end of the spectrum. We get PCs and smartphones, internet, video calls instead of travel, etc. A smartphone delivers significantly more value than a personal car but consumes a tiny amount of energy.
We are currently in the low-energy tech period. The energy supply is still growing, but the speed is declining. This is the reason why (as asked by Peter Thiel) we’ve got 140 symbols, instead of flying cars.
It’s impossible to predict when the low-energy era will switch back to the high-energy one, as it is entirely determined by our ability to grow the energy supply. When it happens, we should be able to revisit flying cars, flying homes, space tourism for everyone, and more. There is nothing we can’t do if we have a supply of very cheap energy, like the one promised by nuclear fusion.
Until then, we have our hands full developing low-energy technologies, like artificial intelligence, virtual reality, biotech (the ultimate low-energy tech ), new materials, and anything that helps to increase the efficiency of how we do things.
Now, let’s look at some of the specific trends.
The digital world is enabled by our ability to build technology to collect, store, process, and transfer information.
Information technology helps us communicate better to get stuff done and fulfill our needs more efficiently. No surprise that it is being applied to every area of our life from commerce and finance to industry and transportation.
The growing efficiency of information technology will allow for even higher optimization of our activities over time. There is a limit though where the gains in efficiency won’t justify the additional costs.
Are we there yet? Unlikely. In the business world, North American organizations spend on average 9.1% on IT, while the global share is 6%. Getting to 9% would grow the overall market size by $2T enabling the creation of $20T of new market capitalization. Meanwhile, the NA level is not a ceiling either. The most advanced segments, like financial services, spend up to 11% of their revenues on IT. Robotization will drive that ratio up for the historically less heavy spenders, like transportation, storage, and manufacturing. We estimate that robotization is at least a $2T opportunity globally.
In the consumer space, our estimates also show a lot of growth ahead and the creation of at least $10T of new market capitalization over the next 20 years.
Digital crime, digital security, digital law enforcement, and digital defense will grow in importance.
The great news is that the digital world is a much more equal place. There are no distances in it. There are no borders. There is no skin color, age, or gender. It opens opportunities for remote work, online learning, and global reach for entrepreneurs and activists. All you need is the right equipment and an internet connection. It is critical to provide access to these things to people around the world.
Think about how much human suffering would have been saved if the mighty pharaoh Khufu (also known as Cheops) decided to build his great pyramid in Minecraft instead of the physical world.
We, human beings, have very limited physical needs — food, shelter, healthy body. The rest is non-physical — connection, entertainment, self-realization, recognition. We are also able to accept virtual things, like laws, money, awards, etc., for the real ones (as is well articulated in the book Sapience by Yuval Harari) and derive pain or pleasure from them.
Combine it with the fact that the digital world is much more efficient, as we are moving bits, not atoms, and it becomes clear why we spend an ever-growing amount of time in it. It simply enables more efficient fulfillment of our non-physical needs.
We are already quite deep in virtual reality. We entertain ourselves with online video, compete and win in online games, connect with people we care about via social networks and video calls, and study in online universities. In the workplace, many of us stare at the screen almost every minute.
The future growth of virtual reality is defined by three factors.
The first one is the physical jobs automation level. The higher it is, the fewer people will be needed there and the more time humanity will spend in the virtual world. It is estimated that we spend about 26 billion hours online every day, or about 3.2 hours a day per person. Our calculations show that this number may at least double.
The second is broadband network coverage. The digital world is only possible if you have a real-time communication network connecting people around the globe. Currently, 60% of the world’s population has access to the internet, eventually, this number will reach 100%. The bandwidth will also increase.
The third factor is the level of comfort and versatility of our virtual world immersion technology. PC was the first breakthrough, but it was clumsy and stationery. Smartphones added mobility. VR goggles are on a path to improving immersiveness. On the other end, gigantic data centers will power the back-end side. With each step, as the level of immersiveness grows, we will gradually replace activities that we would otherwise do offline with a digital substitute. It will be a more efficient approach.
Are we going to transition to virtual reality completely? Maybe eventually, however for the time being we will live in a mix of the real and virtual worlds.
Until recently we had to tell the machine how to do things, giving precise step-by-step instructions while carefully covering every exception. Now we can tell the machine what to do, and let it figure out how. The complexity of the task we can request primarily depends on the total computing power available to the machine. (The underlying algorithms are also important, but it turns out we come up with more powerful algorithms as soon as we get access to more processing power.) Recently, machines learned to come up with drafts of language and visual works, generate variations of industrial designs, guide insurance companies in adjudicating claims, and sales agents in communications with prospects. As available computing power grows machines will be able to do more and more sophisticated things.
The levels of excitement the field experienced in 2022 seem a bit excessive, though. Several challenges prevent AI from broader adoption, limiting it to deployment in closed-ended environments (lower complexity), mandatory human supervision, or both.
Nevertheless, the current state of AI opens a wide range of options for entrepreneurs to build an AI-powered business.
Public blockchain — is a protocol to store and process information leveraging infrastructure powered by a large group of independent profit-seeking actors. The results of processing are publicly accessible and practically immutable.
The processing is defined entirely by fully public and transparent algorithms. It eliminates the need for trust in any entity. Instead, you have to have confidence in the algorithms (by exhaustively studying them or relying on experts). Therefore anyone can offer a financial service on top of a public blockchain and capture market share as long as the algorithms are solid and offer better value to the market. You don’t need reputation — the cornerstone of the traditional financial system.
These qualities potentially enable the creation of a new, more nimble, and efficient financial system — operating with money and all sorts of assets. This system also allows for the introduction of new types of assets, like digital collectibles, that was impractical to set up in a traditional financial system.
While it is heartbreaking to see all the missteps of the traditional financial system being enthusiastically repeated in this new ecosystem, the hope is that these are the (very costly) growing pains that will lead to us identifying the optimal way of leveraging it going forward.
See this note for additional thoughts on the blockchain value proposition and potential use cases.
Science and IT
Research is another area where information technology is quite helpful. While it is hard or even impossible to calculate the gains/cost ratio there, it is clear that it is overwhelmingly favorable. Likely, applying IT (including AI) to do science will prove to be the most impactful thing we ever did. The early results, like proteins folding and controlling plasma in a fusion reactor, look promising.
Since the beginning of time, humans have been exploring and settling the earth’s surface, eventually reaching the depth of oceans, the orbit, and the Moon.
Despite reaching outer space 60 years ago we are at the very early stages of exploring the possibilities it opens to us. While space tourism may be catching a lot of attention, the opportunities created by the rapidly declining price of launching satellites into the low earth orbit (100x decline since the 1960s — from ~$200,000/kg to ~$2,000/kg) are an order of magnitude larger. Earth observation and broadband internet are the two most obvious applications, and additional ones will open up as we advance. Military/defense is another growing area.
We expect the space economy to create new opportunities for businesses worth hundreds of billions of US dollars by 2040.
How about the colonization of other planets? If it sounds far-fetched imagine how the idea of eight billion people spread around the globe would have sounded to our ancestors as they were just starting to venture outside of Africa.
Outside of the thrill of exploration inhabiting other planets is the only sure way to get rid of the human extinction risk — due to the global nuclear war, some deadly epidemic, or a collision with another large object in space — which is the biggest risk faced by humanity.
The energy question is defining here, though, as space travel requires massive amounts of energy. Can we leverage a type of energy uniquely fitted to the needs of space travel? After all until recently we didn’t use the wind to power our travel on the surface , but it became a game-changer on the water.
Global catch up
Innovations have always been spreading around the world driven sometimes by trade and sometimes by war (colonization and empire-building). However, never before had the world been so connected. People around the planet can now learn much faster from each other and implement innovations in their societies. Capital has also become much more agile, making it as easy as ever to marshal necessary resources.
As an example, it took the US 25 years to get to nearly full cellular phone coverage (100 subscriptions per 100 people). It took Russia only 7. In Germany 120 years were needed to raise the formal education level from 2 years of schooling to 10. The United Arab Emirates got there is 45.
Although this is great news for humanity, we still have a lot of work to do to make all the innovations people take for granted in developed countries available to those in developing parts of the world. From basics, like access to electricity, clean water, and vaccines to the internet, digital banking, world-class education, and health care.
This transfer of innovations is a two-way street and developed countries also benefit from the ideas that emerged first in the developing parts of the world.
Another great news is that as the economy becomes more digital (see the virtual world) the borders matter less. People in developing countries have a growing number of options to join the world’s most innovative companies as remote workers or achieve global reach as entrepreneurs.
Innovations in the digital realm are transferred faster than the ones in the physical world. Short-sighted political regimes may slow down the process. There can be temporary setbacks driven by geopolitics or natural causes, like pandemics. However, overall it is accelerating.
The less obvious consequence of improving living standards is a rapidly declining fertility rate. To maintain the population, this rate should be around 2.1. In most developed countries, it is already lower. The rest of the world is following suit. We may soon have to worry about population decline rather than growth.
The global catch-up will not be equal. There will always be better-off and worse-off places in the world, just as there are better-off and worse-off places in every country or city. Talented people will tend to move from the latter to the former, further exacerbating the gap. The brain drain will lead to the inability to compete on a global scale and may push the worse-off states to specialize in the trades their reach counterparts are suppressing at home, like global organized crime. However, unlike a single country where the government can mitigate this dynamic by redistributing resources from the better-off places to the worse-off ones, there is no such force on a global scale. We will have to come up with better global governance institutes to deal with this issue.
 That’s not true for individual societies, e.g. the UK replaced heating with wood with coal and then with natural gas. However, at the same time, the coal and wood usage growth in other parts of the world more than compensated for that shift.
 Note: the deceleration of growth, not the decline of energy supply.
 Take the COVID-19 vaccine, which requires a minute amount of energy to produce/deliver, while the human body does all the work training to fight the virus. Or the miracle drug, fixing a terminal condition with just one dose. Don’t let the price tag fool you. The energy profile of the drug is as low as any while the value is unmatched.
 We do now with wind turbines and electric transport.