by Jeff Stibel
In his famous essay “I, Pencil,” Nobel Prize–winning economist Leonard Read points out just how complex even the most simple emergent systems have become. Read demonstrates how something as simple as a pencil is astonishingly hard for one person to make. Humans create billions of them per year, but no single individual knows how to do it on his own. A few people know how to combine the rapeseed oil from the Dutch East Indies with the sulfur chloride and the cadmium sulfide in just the right combination to make the little pink eraser. Others know how to turn petroleum into the paraffin needed for the smooth lacquer coat. Still others know how to make the string that ties together the paper sacks in which the graphite is packed for shipping after it is mined in Sri Lanka. The list goes on ad nauseam. It takes a huge network to create something as simple as a pencil. As scientist Matt Ridley points out, “we’ve created the ability to do things that we don’t even understand . . . We’ve gone beyond the capacity of the human mind to an extraordinary degree.” We expanded our minds, not with technology, but with networks.
This is true of all highly social species. The secret of the ant’s evolutionary success, like ours, is networks. Ants are tremendously diverse, but what they have in common, as biologist Bert Hölldobler explains, is that “they all live in societies; they all are social insects. There is not a single ant species known that lives solitarily. The evolutionary transition from a solitary life to a social life occurred only in about 3 to 5% of all animal species, including our own species, Homo sapiens. But this minority plays an overwhelmingly dominant role in almost all land habitats.” Like those of humans, the social networks of ants enable them to have emergent properties and dominate the landscape.
Through communication and cooperation, ants are able to do things that no other animals can do, even much smarter, bigger-brained animals. In May 2012, Discovery News ran the headline “Human Societies Starting to Resemble Ant Colonies.” This article outlined Smithsonian researcher Mark Moffett’s perspective that while our DNA is most similar to that of chimpanzees, “no chimpanzee group has to deal with issues of public health, infrastructure, distribution of goods and services, market economies, mass transit problems, assembly lines and complex teamwork, agriculture and animal domestication, warfare and slavery.” He added that “ants have developed behaviors addressing all of these problems.”
II
The world is a dangerous place. Of all the species that have ever existed on earth, 99.9 percent have gone extinct. But the success rate seems to be better for the most networked animals. Most of nature’s known social animals still exist—ants, termites, bees, wasps, and us. Even so, survival is tough. More than 90 percent of harvester ant colonies fail in their first year. (Though if they reach the breakpoint, most of the remaining 10 percent will survive another 20 to 25 years on average, sending many reproductive queens out into the world.)
Not only are networks vital to our success, they’re vital to our intelligence. Post-breakpoint, networks are much, much more intelligent than any individual member of the network. It’s true for humans as much as for any other species, and it’s similarly true for technological networks. After all, our technology networks—the internet, the web, Facebook—are just tools to further connect our human network.
The important point is this: social networks after the breakpoint are highly successful both in the short term (the life of the network) and the long term (the survival of the species). This extends to the individuals that make up the network as well: a social animal living outside the network won’t last long. Imagine a baby ant, or even a baby human, somehow forced to live without the support of others. Neither is likely to survive. Social animals do not fare well outside of their networks. On the other hand, an infant grasshopper, snake, or even a kitten stands a much better chance.
Social animals also do not fair well when their networks grow too large. UCLA scientist Jared Diamond argues that we became disconnected from our social networks as a result of the population boom during the agricultural revolution, which he calls “the worst mistake in the history of the human race.” We parted ways with the egalitarian, antlike hunter-gatherer societies in which we had survived and thrived for hundreds of thousands of years. Of course, agriculture allowed our populations to explode (perhaps causing us to overshoot our carrying capacity, the subject of Diamond’s book Collapse). Agriculture enabled large-scale food production in which a few farmers could feed many people, allowing cities to be built. Subsequently, people came in contact with many more people than previously—hunter-gatherer tribes rarely grew beyond 150 members or so. We fundamentally changed the structure of our networks from democratic, antlike “colonies,” so to speak, to a more complex hierarchical system.
The printing press, the industrial revolution, and the digital revolution made the human network more efficient and intelligent. The network revolution is changing us even more drastically. Of course, we are both collectively and individually smarter as a result of our unprecedented access to the world’s information—and thanks to improved search capabilities, we can access that information at unprecedented speed. But the changes brought about by the internet are even more fundamental than that.
The internet both connects us more deeply and levels the playing field between us. For example, ten years ago, it was easier for a dictator to survive with massive disconnects between what he promised and what he delivered; there was no real vehicle to tease that apart en masse. Now there is a vehicle, and in recent years we’ve seen analysts credit social media networks with helping Middle Eastern protesters organize demonstrations, share information, and report to the rest of the world. The web in general and social media in particular, by their very nature, promote democracy and present challenges for dictators attempting to pass off as fact anything other than the truth. Even the dictators know this. In January 2011, the Egyptian government instructed all of the country’s internet service providers to pull the plug. With only four ISPs, all of which were obligated to obey government orders, it was relatively easy for Egypt to cut off its citizens from the outside world.
Of course, this wouldn’t be possible in the United States or Europe as there are hundreds of ISPs with distributed redundancy throughout the internet. As the internet grows in the Middle East, it will become increasingly difficult to do such a thing there as well. Our human network is getting both deeper and broader, and the network itself is more powerful than any dictator. In a sense, the democratization of the web is reversing the hierarchies brought about by agriculture and returning us to the tight-knit networks that have allowed humans to thrive throughout history.
III
Of course, networks must get through a breakpoint and reach equilibrium before they are of real value to us. Just as our brains aren’t wise until they hit a breakpoint and prune neurons and connections, the networks that connect our brains to those of others must be allowed to mature. For technology networks, this means encouraging growth at all costs and avoiding monetization too early; but it also means shifting gears once the breakpoint is reached.
The results of patience are well worth it, both for corporations that must satisfy stakeholders and for humanity as a whole. The result of mature, fully functioning networks is a more tightly knit world with capabilities well beyond the sum of our abilities.
The network revolution has changed the game permanently, and this is just the beginning. What is to come will be more exciting than ever. Technology is on the verge of creating the types of things habitually reserved for humans: consciousness, intelligence, and emotion. The future will be limited only by the limits of the greatest imaginations of our technological and biological networks.
Twelve
Afterword:
The Internet Is a Brain
Throughout this book, the internet has been used as an analog to the brain because both are complex networks. But there are even more fundamental similarities, which I woul
d argue are evidence that the internet is not merely like a brain but is a brain. While not central to the concept of breakpoint, these ideas are significant in developing a full understanding of the potential of the internet, especially as it relates to artificial intelligence.
The important question is this: Could the internet itself be made to perform more like a brain or even perform the functions of a brain—just as a hearing aid performs the function of the inner ear, or a contact lens performs the function of the cornea, or an artificial heart performs the function of that biological muscle?
A computer is generally a poor analogy to a brain. It’s true that semiconductors switch on and off like neurons, and that fibers of glass can transmit messages like synapses and axons, but that’s where the analogy ends. Our computers are not nearly as analogous to the brain as, for example, an artificial heart is to a real heart.
The internet, however, is unlike anything humankind has built before. All of our previous inventions—steam locomotives, television sets, cars—are inert. Chessboards and baseball stadiums may flicker to life momentarily, but go dark when the game is done. The internet is different. It’s unbounded, self-perpetuating, and capable of collective consciousness. It’s more like the crowd watching the baseball game than the stadium itself.
Of course, every innovation that delivers something greater than the sum of its parts is miraculous. Alexander Graham Bell attached two small drums to two wire coils and out of those bits created something beyond the sum of the parts: sound. But the telephone did not go on to replicate and improve itself. The internet can and does. And beyond that, the internet learns. It processes information, shapes it, transmits it. It remembers some things, forgets others, and constantly loops whatever it has again and again, spinning it in as many ways and in as many directions as one could imagine.
I
Without a strong understanding of what a brain actually is, it may seem preposterous to say that the internet is a brain. The internet is not the three-pound wrinkled gray glob that most of us conjure up as an image of the brain. Actually, that’s not even what the brain is. The brain is nearly 60 percent white matter—the tissue that connects neurons—with only the remainder being the gray stuff we typically think of. The gray matter contains the all-important neurons, but the connections are equally important.
Outside of the deep ridges and two hemispheres, most people wouldn’t recognize a brain if presented with one. The brain is very soft, almost jellylike, and ivory in color with deep burgundy-colored veins. The brain doesn’t take on a firm gray appearance until it’s dead, bloodless, and preserved. This visual distinction is important because it tells us that a living brain is consuming massive amounts of energy in the form of blood flow.
But even that description is somewhat misleading. In the way the brain actually functions, it is far more similar to a piece of paper. The paper represents the outermost area of the brain, the cerebral cortex. It is here that most of the magic of thought takes place. Imagine this piece of paper: thin, rectangular, and mostly blank to start. On the paper are bits of information that grow as the brain is formed, like braille embossed on the page. Those are the neurons, and they help to store and process information.
The ingenuity of the brain comes not from the informational elements but from how that information is physically connected. Imagine crumpling the piece of paper into a ball. Two dots at either end of the page are far away from each other initially. But as you crumple the paper, the dots get closer. Crumple it enough times, and every point will be in striking distance of every other point. Our brains are folded and crumpled into our skulls in just this manner, and their unique power comes from the ability to connect disparate pieces of information for quick communication and interdisciplinary learning.
In terms of the computer industry, the human brain is a sophisticated “parallel processing” machine. That means that it does a number of things at the same time, unlike “serial computing” in which one thing happens, then another, and then another. Neuroscientists call this “distributed computing,” meaning that since the functions of the brain are distributed all over the place, things can happen simultaneously. (“Distributed” is a more accurate term than “parallel” because parallel computing conjures up the idea of two unbending parallel lines, like railroad tracks, whereas “distributed” is a freewheeling image that describes more accurately how the brain actually works.)
A neuron consists of the soma, an axon, and dendrites. Think of the soma as the center of the neuron or the information clearinghouse. The axon acts as a transmitter, sending information from one neuron to another. The dendrites receive information from other neurons. Neurons communicate with one another through electrical and chemical transmitters. These tightly packed neurons work together in a distributed network, forming patterns that allow us to perform tasks such as walking, speaking, remembering someone’s name, and even reading this book.
II
The fact that the brain is an ordinary organ is actually a good thing in terms of trying to create an artificial mind. It frees us to speculate that an intelligent internet is possible. We’ve always considered the brain to be a sacred organ in a sacred chalice, but to many philosophers, scientists, and more than a few internet entrepreneurs, the idea of mechanized thinking is no longer beyond the question, “Why not?”
Never before has the idea of a “thinking machine” brought together a greater confluence of thinkers and scientists. They range from neurologists, who are dissecting the brain with greater skill and instruments; to psychologists, who are understanding behavior that emerges from the brain; to linguists, who recognize how thoughts are put into the symbols that we call words; to evolutionary scientists, who are developing a new field called genetic engineering; to computer scientists, who are building machines and algorithms that mimic the mind; and to artificial intelligence types, who are focusing intently on getting machines to actually think.
Of course, even among this group of distinguished thinkers, there is no consensus on how to create artificial intelligence. Tufts University philosopher Dan Dennett notes in his book Consciousness Explained, “No one can keep all the problems clear, including me, and everyone has to mumble, guess, and hand wave about large parts of the problem.” The voices are many, the opinions sharply different, but in this burgeoning world of brain science and technology, we are finally seeing a convergence on the idea that a thinking machine is inevitable. And, consequently, that this newfound intelligence will affect our lives profoundly.
What is bringing this diverse group of people together is the internet. The reason, of course, is that the internet is very similar in structure to the brain, in that it’s mainly a massive storage, calculation, and communication system. The internet is clunkier and smaller than the brain (neurons versus computers, not sheer size or weight), but the fundamental structure is roughly the same. The brain has neurons and memories while the internet has computers and websites, connected together through cables and hyperlinks instead of axons and dendrites.
Of course, something may look like something else and not have a connection. The switching system of a phone network looks something like a neural network, but it isn’t. Yet there are things about the internet that make this similarity too strong to resist.
Remember, the internet is really the combination of two innovations. The first was the telephone, which allowed information to be transmitted electronically. With the advent of the telephone, people could communicate instantaneously across enormous distances and through the most difficult terrain. This seems obvious to us now, but it was unimaginable during the days of the Pony Express.
The second innovation was the computer, which allowed us to process and store large amounts of information. Before the computer, we processed information using calculating devices and then wrote down the results on paper. Complex calculations, the type even the most modest laptop does without
breaking a sweat, were virtually impossible before the computer. And if we needed to store valuable or large amounts of information, like a book manuscript, for example, the best options were to put it either under a mattress or in a bank.
Both the telephone and the computer solved huge problems, but combining the two and creating the internet created revolutionary opportunities. The modern-day internet may simply be a set of computers hooked together by telephone lines, but this simple yet powerful amalgamation allows us to store, process, and transmit information.
The internet is uniquely powerful because it has hundreds of millions of computers connected to each other, all sharing information and working on your behalf as you sit in your house searching for “vegan brownie recipes” on Google. Like the neurons in the brain, the internet processes information in parallel across its millions of computers.
The human brain has around 100 billion neurons. In another 20 years or so, the number of computers connected to the internet will arrive at a comparable total. In time, the internet will approximate the complexity of the brain. Think of it this way: it took hundreds of thousands of years for the human brain to evolve to its current level of complexity and sophistication. The internet will approximate that in a few generations. We will have experienced in cyberspace a replication of biological growth itself, as though it were a living brain. But more specifically, we will not merely replicate the brain itself, we will duplicate the brain’s byproduct: thought.