The Second Machine Age
Erik Brynjolfsson and Andrew McAfee
50 min
9 Key Points
4.3 RateWhat's inside?
Explore the impact of digital technologies on our lives and economy, and learn how to navigate and prosper in this new era of innovation and automation.
You'll learn
1. How tech changes our economy and society2. Thriving in the tech era3. The money side of AI and machine learning4. Keeping your business ahead in the digital age5. What tech advancements mean for jobs6. How policy and education can shape a tech-driven future.Key points
01The New Era of Exponential Growth
Human history is a fascinating study in stagnation followed by explosive, sudden growth. For thousands upon thousands of years, the trajectory of human development was essentially flat. If you trace the line of human progress, wealth, and population from the dawn of civilization through the ancient empires of Egypt, Greece, and Rome, all the way into the Middle Ages, you will notice a startling reality. Life for the average person did not change much from one century to the next. A peasant living in the fifteenth century wielded the same basic tools, suffered from the same diseases, and produced roughly the same amount of agricultural output as a peasant living in the fifth century. Human power was strictly limited by the physical strength of our muscles and the muscles of our domesticated animals. We were bound by the biological constraints of flesh and bone. Everything changed dramatically in the late eighteenth century with the invention of the steam engine, brilliantly refined by James Watt. This singular invention sparked the Industrial Revolution and ushered in what Brynjolfsson and McAfee refer to as the First Machine Age. Over a remarkably short period, the trajectory of human progress shot upward at an almost vertical angle. The steam engine allowed us to overcome the physical limitations of muscle power, generating massive amounts of energy to run factories, power locomotives, and cross oceans. For the first time in history, human physical capabilities were multiplied almost infinitely. This era of physical augmentation brought about unprecedented surges in population, wealth, and living standards. The world as we know it today—with our towering cities, global supply chains, and mass-produced goods—was built on the foundation of this First Machine Age. We are now crossing the threshold into a fundamentally different era. We have entered the Second Machine Age, a period defined by the augmentation of our mental capabilities rather than our physical ones. Just as the steam engine multiplied physical power, computers and digital technologies are now multiplying our cognitive power. We are building machines that can calculate, analyze, and even "think" in ways that were previously the exclusive domain of the human brain. To understand the sheer velocity of this new era, we must come to grips with the concept of exponential growth, a force that is incredibly difficult for the human mind to intuitively grasp. The driving engine of the Second Machine Age is Moore’s Law. In 1965, Gordon Moore, a co-founder of Intel, observed that the number of transistors on an integrated circuit was doubling roughly every year to two years. This meant that computing power was doubling at a predictable rate while the cost was simultaneously plummeting. Many experts assumed this trend would hit physical limitations within a decade or two. Yet, decades later, engineers and scientists have continually found brilliant new ways to keep Moore’s Law alive. This relentless doubling has transformed computers from massive, room-sized machines that could barely calculate missile trajectories into sleek smartphones that fit in our pockets and possess more computing power than all of NASA did during the Apollo moon landings. To truly appreciate the power of exponential growth, the authors share a profound fable about the invention of chess. According to legend, the inventor of chess presented his beautiful new game to a powerful emperor in India. The emperor was so captivated by the game that he offered the inventor any reward he desired. The inventor’s request seemed remarkably humble. He asked the emperor to place one grain of rice on the first square of the chessboard, two grains on the second square, four grains on the third, eight on the fourth, and so on, doubling the amount of rice on each subsequent square until all sixty-four squares were filled. The emperor, thinking this was a trivial amount of rice, readily agreed and ordered his treasurer to fulfill the request. During the first half of the chessboard, the amounts of rice are perfectly manageable. By the thirty-second square, the emperor would owe about four billion grains of rice, which is roughly the yield of one large field. It is a substantial reward, but certainly within the means of a powerful ruler. However, as they cross into the second half of the chessboard, the nature of exponential growth reveals its terrifying power. As the doubling continues, the numbers become astronomically large. By the sixty-fourth square, the total amount of rice required would be more than eighteen quintillion grains. This is a mountain of rice larger than Mount Everest, far exceeding all the rice produced in the history of the world. The emperor, realizing he had been tricked by the mathematics of exponential growth, likely had the inventor executed. The brilliant insight of The Second Machine Age is that human technological progress has now entered the second half of the chessboard. We have been doubling computing power for decades. In the early days, those doublings produced noticeable but manageable improvements—faster calculators, basic spreadsheets, and pixelated video games. Now, however, because the baseline numbers are so massive, every single doubling produces a staggering leap in capabilities. We are seeing leaps in processing power, memory storage, and sensor technology that defy our traditional expectations. This exponential growth explains why technologies that seemed impossible just a few years ago are suddenly becoming part of our daily reality. We are living in a moment where the sheer volume of computing power available allows for real-time language translation, incredibly realistic computer-generated imagery, and algorithms that can sequence human DNA in a matter of hours. The second half of the chessboard is characterized by continuous, jaw-dropping breakthroughs. We can no longer rely on linear thinking to predict the future. If we expect technology to progress at the steady, predictable pace of the past, we will be entirely unprepared for the massive disruptions and incredible opportunities that lie just ahead. The Second Machine Age is not just a modest upgrade to our existing tools; it is a fundamental rewriting of the rules of human potential.
02When Technologies Combine and Multiply
Innovation is often misunderstood as a solitary endeavor. We have a romanticized vision in our minds of a lone genius, locked away in a laboratory, suddenly experiencing a brilliant "eureka" moment that changes the world from scratch. While those moments of singular invention definitely happen, they are incredibly rare. The vast majority of technological progress does not occur through isolated breakthroughs, but rather through a process called recombinant innovation. This concept is absolutely crucial to understanding why the pace of change in the Second Machine Age feels so overwhelmingly fast. Recombinant innovation works exactly like building with Lego bricks. When you have a small set of Lego bricks, there are only a limited number of ways you can snap them together to create something new. However, as you add more and more unique bricks to your collection, the number of potential combinations does not just grow linearly; it explodes exponentially. Every new piece you introduce can be combined with all the existing pieces in countless different ways. In the realm of technology, every new software protocol, every new piece of hardware, and every new data set acts as a new building block. As our collective toolbox grows, innovators have an ever-expanding array of components to mix, match, and recombine into entirely novel solutions. To see recombinant innovation in action, consider the phenomenal success of the navigation app Waze. Before Waze existed, GPS navigation was a top-down, static industry. Companies would spend millions of dollars sending specialized vehicles equipped with expensive surveying equipment to drive every road, mapping out the terrain. They would then sell this digital map data to consumers in the form of expensive standalone GPS devices for their cars. Updating these maps was a painstakingly slow and expensive process, meaning your navigation device was often out of date the moment you purchased it. The founders of Waze did not launch their own fleet of surveying vehicles, nor did they invent a new satellite system. Instead, they looked at the digital building blocks that already existed in the world. They saw that millions of people were walking around with smartphones in their pockets. These smartphones already contained GPS sensors, which could track location. They contained cellular radios, which could transmit data instantly to the internet. And they contained screens to display information. The brilliant recombination was realizing that every single smartphone user could simultaneously act as a sensor for a massive, dynamic map. By writing a clever piece of software, Waze combined the existing hardware of the smartphone, the existing infrastructure of the cellular network, and the existing concept of crowdsourcing. As people drove with the app open, their phones silently transmitted speed and location data back to Waze’s servers. If a dozen cars on a specific highway suddenly slowed down, the algorithm instantly recognized a traffic jam and automatically rerouted other users to avoid it. Waze also gamified the experience, allowing users to actively report accidents, police traps, and road closures. They built a multi-billion-dollar company not by inventing new physical infrastructure, but by recombining existing digital technologies in a highly innovative way. This brings us to a critical distinction between the First Machine Age and the Second Machine Age: the nature of the resources we are combining. In the physical world of the First Machine Age, resources are rival and excludable. If a factory uses a ton of steel to build a bridge, that exact same ton of steel cannot be used by someone else to build a car. Physical goods get used up. They degrade over time. Furthermore, reproducing a physical good always incurs a marginal cost. It costs real money to manufacture a second car, print a second book, or press a second vinyl record. Digital technologies, however, play by an entirely different set of economic rules. Digital resources are non-rival. If you use a piece of software, a line of code, or a digital map, it does not prevent a million other people from using that exact same data at the exact same time. The data does not wear out or degrade with use. Even more powerfully, digital goods have a marginal cost that approaches zero. Once a software program is written or a digital song is recorded, creating the second, thousandth, or millionth copy costs virtually nothing. You simply duplicate the bits and bytes. This combination of non-rivalry and zero marginal cost creates a combinatorial explosion of innovation. Innovators today do not have to worry about running out of digital raw materials. A developer in a small bedroom in Estonia can access the exact same open-source code libraries, cloud computing networks, and artificial intelligence frameworks as a massive tech corporation in Silicon Valley. They can combine machine learning algorithms with digital cameras and internet databases to create something entirely new, instantly deploying it to a global audience. General Purpose Technologies, or GPTs, are the ultimate catalysts for this kind of growth. Throughout history, there have only been a handful of true GPTs—technologies so profound that they transform every single sector of the economy. The steam engine was a GPT. Electricity was a GPT. They were not just single-use tools; they were foundational platforms upon which countless other innovations were built. Today, Information and Communication Technology ICT is our era's defining General Purpose Technology. Because ICT provides the building blocks for almost every modern industry, advancements in computing do not just improve the tech sector; they improve agriculture, medicine, finance, logistics, and entertainment simultaneously. A breakthrough in artificial intelligence does not just mean better video games; it means more accurate medical diagnoses, more efficient global shipping routes, and stronger financial fraud detection. As these diverse fields become increasingly digitized, they feed new data and new building blocks back into the system, accelerating the cycle of recombinant innovation even further. We are living in a continuous loop of creation, where every new discovery serves as the foundation for tomorrow's combinatorial breakthrough.
Continue reading with LeapAhead app
Full summary is waiting for you in the app
03Artificial Intelligence Steps Out of Sci-Fi
04The Bounty and the Widening Spread
05Why the Middle Class is Disappearing
06Measuring the Unseen Digital Economy
07How to Thrive in a Changing World
08Conclusion
About Erik Brynjolfsson and Andrew McAfee
Erik Brynjolfsson is an economist and MIT professor specializing in the economics of information technology. Andrew McAfee is a scientist, author, and MIT professor focusing on how technologies are changing business, the economy, and society. They co-authored several books on the impact of technological advances.
