A review of Radical Technologies by Adam Greenfield

One Sentence Summary: Current and near-future technologies offer great potential for enhancing our lives, but we need to consider the inherent trade-offs in adopting products and services that dictate an increasing portion of our everyday experiences.

“We live in a society exquisitely dependent on science and technology, in which hardly anyone knows anything about science and technology”- Carl Sagan

In the nearly three decades since Carl Sagan voiced his concern, we have adopted an amazing array of technologies into our daily lives — smartphones, laptops, the World Wide Web, digital maps, Siri, Alexa, Facebook, the Internet of Things — which have increased convenience and expanded our connectivity with other individuals around the world. Yet, we still do not grasp the fundamentals behind these technologies or realize what we sacrifice when we adopt them without a moment’s hesitation. We imagine technology as a beneficial force with only positive effects — greater ease and access to information — while we overlook the trade-offs — decreased privacy and autonomy — implicit in upgrading to the latest model. Advances promised to us in the coming decades — augmented reality, 3D printing, machine learning, cryptocurrency — are designed to satisfy our needs and bring us novel forms of entertainment. Before we blindly accept these products and services, it is critical that we understand the implications of relinquishing ever more control of our day-to-day experiences to the companies that provide them. This argument forms the premise for Adam Greenfield’s Radical Technologies: The Design of Everyday Life, which explains not only how modern and near-future technologies work, but the concessions we make when these developments are woven into the fabric of modern life.

Ceding Control

The modern smartphone has become our constant companion — you probably spend more time with your phone than with any individual — and the tools it offers us are indispensable to many of our daily routines. What we see in our smartphones is a sleek exterior housing and a 4.7 inch screen radiating crystal clear images of anything that we desire. What we experience with our devices is an ability to communicate with anyone in the world at a moment’s notice, or to navigate in an unfamiliar city and receive recommendations tailored to our interests. What we fail to observe is the $250 worth of precious metals and raw materials making up our phones that are extracted at an enormous environmental cost, or the $4 in labor required to assemble the phone in Shenzhen, a city of 18 million in China where 90% of the world’s electronics will pass through at some point in the manufacturing process. Also escaping our purview is the immense amount of data we generate in the process of every action involving our phones. When we receive directions from Google Maps, we don’t stop to consider how incredible it is that Google can provide us with real time maps of the entire world at no cost. In reality, Google has become one of the most valuable and influential companies in history through the accumulation of data, and every service they offer is provided not for free, but in exchange for access to the information our daily activities generate. When we stop at a store or drive faster than the speed limit while using Maps, Google records that information and can then sell it to advertisers or auto insurance companies that might be interested in your driving habits. Greenfield refers to this as the monetization of everyday experience. After companies realized they could extract valuable information from the activities we do on a regular basis, they developed seemingly free platforms to entice us to share more information. Facebook does not exist because it believes in the ultimate good of allowing humans to connect with one another, it exists because there are billions of dollars to be made from capturing our social interactions, our likes, and our interests. We are happy to exchange vast quantities of data about our personal preferences for a curated news feed with content that Facebook knows will keep us coming back day after day.

Facebook, Google (now under Alphabet), Amazon, and Apple are becoming more than companies that offer us services and devices. By creating ecosystems that control ever more parts of our lives, they want to become decision-makers for the majority of our daily choices while gathering vast amounts of information in the process. Greenfield refers to these ecosystems as “Stacks,” or institutions that want to make themselves into the sole providers of services/products in a domain. The most efficient way to accomplish these aims is through the aggregation of user data which allows the stacks to provide you with better personal recommendations and sell that information to other companies. When we log into our social networks, we see precisely the kinds of things that Facebook has identified interest us based on our past tendencies. From too many personal experiences, I know the incredible effectiveness of Amazon product offerings, and if you are still unwise enough to use the web without an ad-blocker, you might have noticed ads for a product you researched on Amazon minutes before. In effect, both of these companies have become information filters. We give them access to our history, and they sort through the mounds of products and news stories and show us only what they have learned we want to see.

To some degree this is a necessity; the amount of information in the modern age is overwhelming, and without a means to sift through it, we can waste time on irrelevant content. Yet, what are we missing in the news stories we never see or the search results that Google does not show us? The extreme polarization in modern politics is at least partly because the pre-filtering of content has made it possible for us to read only the information that supports our views. If you rely on Facebook as your sole source of news, you can go the rest of your life without seeing a contrary opinion. It might be nice to wake up on a Sunday morning and read stories that confirm your pre-existing beliefs, but this results in an inaccurate conception of human life that misses all of our nuances and conflicting opinions. Furthermore, society progresses not when individuals gather in echo chambers, but when all sides of an argument are allowed to intermingle and to discuss the issue. The only place these debates might occur for most of us is online, but they have no chance to occur when we artificially limit the diversity of information to which we are exposed. In effect, we have created is a system in which we are censored by our own likes and search history, which has real-world consequences for the political functioning of our country.

Not only do the stacks determine the news we read and the search results we see, but over the past few years we have gradually off-loaded more and more decisions to their algorithms (an algorithm is a system that takes in some information and executes a set of instructions based on that information to produce an output). These algorithms have become astonishingly accurate thanks to the maturation of machine learning. The capability of machine learning algorithms is directly proportional to the amount of data they have access to because these models are “trained” on past data. For example, the Amazon product recommendation software is a machine learning algorithm that has learned from the shopping history of millions of users for almost two decades. Anytime you are logged in to your Amazon account, the algorithm compares your shopping and search history to all other users and recommends the products you are most likely to be interested in based on what other users with similar profiles have purchased. As you provide more information to the system, it gets better at predicting what you want. According to a recent study, with just 150 likes, Facebook knows your preferences better than close family members, and, with 300 likes, Facebook can predict your choices better than your spouse. This incredible ability to forecast our future behavior means we no longer need to make conscious consumer decisions. The music we listen to, the products we buy, the movies we watch, the route we take to work, the restaurants where we eat, and even the people we date are now routinely selected not by sentient humans, but by machine learning systems that might know us better than we know ourselves. What we gain in convenience we lose in autonomy, or the ability to make our own choices. Moreover, for now, these algorithms are still developed by humans and they reflect the biases of their creators. In 2017, those developers were overwhelmingly white, male, 20-somethings, living in a particular region of California. With each passing year, we pass more and more decisions to this small group of individuals employed by the Stacks.

Engineering Our Downfall

In addition to recommending us the perfect restaurant to try or the optimal route to take home, machine learning algorithms are rapidly developing the ability to perform tasks previously accomplished only by humans. Nearly every job held by humans today is based on patterns and algorithms: we encounter a situation, respond to it via a set of learned rules, and produce useful output as a result. Even the job of a doctor is an algorithm: they ask a patient a series of questions, check the patient’s medical history and symptoms, and then generate a treatment plan based on a set of established guidelines. There are nuances between the interpretation of a patient’s condition, but the overall process is always the same. Creative endeavors such as writing and music composition come down to recognizing and recreating the patterns humans find most enjoyable. Machine learning algorithms excel at pattern recognition and hence have conquered many human domains. Machine learning systems can now diagnose MRI scans more accurately than doctors, navigate city streets better than human drivers, and compose music that is indistinguishable from that of the best biological composer. Past threats of machines taking human jobs went unfulfilled because humans found a way to stay a step ahead. When agriculture was mechanized, we taught ourselves to operate machinery in factories. When robots that never needed to rest or lost concentration pushed us out of industry, we went to college and contributed to the service economy. However, the new machine intelligence leaves no room for humans to outsmart or outwork mechanical systems. A recent study concludes that 47% of jobs in America are vulnerable to automation within the next ten years. Humans could be at real risk for becoming economically irrelevant and current government systems are too slow to deal with this potential future. The only institutions with the ability to rapidly adapt are the stacks themselves, and they are the ones who most stand to gain from the adoption of machine intelligence.

Even though we may literally be engineering our own irrelevance, it is inconceivable that research into machine intelligence will cease or even slow. Humans have an insatiable desire for progress, even when that progress comes with a demonstrable risk. Researchers are working to develop artificial intelligence, which can be thought of as conscious machine learning algorithms with their own motivations and values. An artificial intelligence that could create its successor would likely be the last invention humanity would ever need to make. Even without any malicious intent, AI could cause the end of humanity. In our quest to transform the world to our liking, we have caused the extinction of numerous species not out of harmful intent, but because we think that our objectives are more important than the survival of a lesser species. Who is to say that an AI would not feel the same way towards us? Although achieving AI could result in the eclipse of humanity, researchers will never give up the dream because of the difficulty of the challenge and the human nature to continue forging ahead into the unknown. This drive has resulted in all of the wonders we enjoy today and the current golden age of civilization, but it also poses an existential risk. While this may seem like science fiction and AI has been “two decades away” for 50 years, it is important that we discuss the implications of our increasing trust in algorithms and recognize that every decision we give up brings us a step closer to a world with no need for humans. Political discourse is rightly dominated by everyday concerns, but it also needs to encompass long-term planning that accounts for current scientific research. Rather than a reactive government, we need a rule-making body that can anticipate future needs and draft legislation in anticipation for the world of decades from now.

Technologies of the Future

Radical Technologies is not a pessimistic work. As an advocate for human-centered technology design, Adam Greenfield is best described as a techno-skeptic, and his book is a balanced assessment of new advances compared to the unbridled optimism emanating from Silicon Valley. Therefore, when Greenfield places his approval behind a new technology, his words carry more weight than an author who praises every new phone model. Foremost among the technologies that Greenfield believes will have a positive effect on society in the coming decades are 3D printing and cryptocurrency. News coverage of both of these subjects tends towards extreme praise (“3D printing will change the world by allowing everyone to be their own manufacturer”) or gloom (“cryptocurrency will enable criminals to conduct transactions with no traces”) which underscores the mundane but steady progress in both fields. 3D-printing, also known as additive manufacturing and embodied by companies such as MakerBot, could distribute ownership of the means of production, but it has so far failed because of high system costs and the availability of cheap products produced overseas. Why buy your own additive manufacturing system and filament to make a chair when you can simply go on Amazon and have any one of thousands delivered to your house in 2 days for under $100? The real people who could benefit from 3D-printing are exactly those who currently lack access to the technology, such as people living in developing countries where the global supply chain extracts from rather than benefits the poor. Additive manufacturing could create a sustainable circular economy, where once a product has reached the end of its useful life, it is simply melted down into material used to create another product. However, as long as the technology remains solely in the hands of mostly upper-class western individuals, it will not profoundly change the world. 3D-printing will be most useful not where it is an expensive hobby for tinkerers, but where it can create necessities of daily life by harvesting waste streams and open-source designs.

Cryptocurrency, the most notable example of which is Bitcoin, is another development with great unrealized promise. Greenfield points out that both 3D-printing and cryptocurrency are brilliant concepts that have failed at the interface between the technology and the real world. Digital currencies are designed to enable greater freedom of transactions because they do not have to go through a central clearinghouse. Bitcoin records information on the Blockchain, a global record of all transactions conducted in the currency that is available for download by anyone with an Internet connection. The verification and recording of a transaction is not conducted by a single entity, but by every computer on the network. This means that theoretically, Bitcoin is both less susceptible to price fluctuations than traditional currencies because it is not pegged to a physical entity, and it is not as vulnerable to fraud because faking a transaction would essentially require fooling every single computer verifying the transaction. In the real world, Bitcoin has failed because it does not fulfill the basic function of a currency: it cannot be easily spent. Anyone willing to purchase and spend Bitcoin must first find a broker who exchanges American dollars for Bitcoin, obtain a wallet that can store the currency, and then find a vendor willing to accept the currency in exchange for a product. In an age with one-click checkout, this will never be a feasible option for the majority of consumers who value convenience over privacy. Bitcoin will ultimately be unsuccessful because of difficulty of use (and technical details such as the energy cost ) but the idea of a Blockchain will live on in other currencies and distributed autonomous organizations. When trust in government is eroding (whether or not that is justified), the idea of distributing responsibility to all users of a system could be invaluable to operations such as organizing protests or creating a smarter energy grid.

Recommendation

The future is shaped not by those who merely know a technology works, but by committed groups of individuals who understand the societal implications of new advances. Knowing how technology functions is important, but knowing how it is used is more critical. Radical Technologies examines both the fundamental building blocks of current and near-future developments and the potential ways in which they may impact society. This book is a worthwhile read in its entirety because it forces us to consider the concessions we make when we allow ever more devices and products to control our daily lives. Once we are aware of the extent to which our existence is dictated by algorithms, it is not difficult to find ways to gain back a little control. You can start by disabling notifications on your phone or choosing only select times to receive alerts to reclaim your time and attention. Next, stop self-censoring yourself and read a different news source or follow entities you might not agree with in order to at least understand their argument. Finally, be skeptical when making any consumer decision. Ask yourself: “Is this really what I want, or is this only what Google/Facebook/Amazon/Apple want me to believe that I want?” Often, the benefits of a new technology more than outweigh the trade-offs, and I would claim that the modern connected world is a much better and enriching place to live because of the computer revolution. Nevertheless, the question that we must ask with regards to any advance is how can it be incorporated into our lives without concentrating power in the hands of those select few who understand the technology?