At present, we see giant leaps in technology, with several technologies developing at an exponential pace. The processing power and memory of computers increase exponentially, compared to the price. The bandwidth of data transmission grows exponentially. At the same time, robots become cheaper and more advanced. The price of having your genome sequenced is dropping exponentially, and we continually increase our knowledge of how genes affect our health – and how to replace bad genes with good ones. In materials technology and energy technology, we also see great progress.
Previous technological revolutions have been driven by progress within a single technology – the printing press, the steam engine, electricity, radio waves – but this time, the changes are driven by a whole range of technologies, and that may make a difference. Maybe it will lead to a total meltdown of civilisation as we know it, as the institutions of society fail to keep pace with the development (see “The world is heading for a meltdown” in our last issue). Maybe we are rapidly moving towards the ‘technological singularity’ where superhuman artificial intelligence, with lightning speed, changes everything in ways we cannot comprehend today, and we may expect to live forever as technological superhumans, or as electronic consciousnesses in computers, dwelling in life-like simulations shaped by our deepest fantasies.
In this magazine, we have previously argued that the accelerating technological development is just an illusion created by temporal myopia (“Is acceleration in reality just an illusion?”, SCENARIO 3/2013); that exponential growth very often is just steady growth (“TECHTALK”, SCENARIO 3/2017); and that a thousand-fold increase in computer power does not imply a thousand-fold increase in its practical value (“TECHTALK”, SCENARIO 6/2016). If these reflections prove to be correct, it may mean that the world will not be changed that much after all, despite the technological development.
Are we heading for the technological singularity, or will the world, by and large, be much the same in 50 years? Or will the future reality lie somewhere between these two extremes, with significant changes, but no singularity as such? We will examine these three possible futures in three scenarios. None of them are meant as straightforward predictions of the future; instead, they will be used to show the scope of the extent to which technology can come to change the world towards 2100.
THE TECHNOLOGICAL SINGULARITY
In mathematics, a singularity signifies a point where the value of a function increases towards infinity, like the function 1/x as x approaches zero. In astronomy, the term is used for black holes, objects so dense and heavy that even light can’t escape them. In futures studies, ‘singularity’ is used about a hypothetical future society where superhuman artificial intelligence creates rapid changes we cannot comprehend, no more than apes can comprehend the changes our civilisation has gone through within the last century.
The idea of a technological singularity dates back to the American physicist John von Neumann who, in the 1950’s, stated that “the ever accelerating progress of technology and changes in the mode of human life gives the appearance of approaching some essential singularity in the history of the race beyond which human affairs, as we know them, could not continue.” It was not, however, until a 1993 paper by the computer scientist Vernor Vinge, “The Coming Technological Singularity”, that the idea found its modern expression. Vinge began his paper with the words: “Within thirty years, we will have the technological means to create superhuman intelligence. Shortly after, the human era will be ended.” Since then, many others have adopted the idea, most notably the well-known futurist Ray Kurzweil, who however does not believe that we will witness the singularity before 2045.
The idea is, briefly, that sometime within the next ten to twenty years, we will develop artificial intelligence that can solve more complex tasks than humans. Then, this AI will be able to develop an even more intelligent machine which will, in turn, develop something even more advanced; and soon, we will have machines of unimaginable intelligence. These machines will develop super-advanced technology and knowledge in all sorts of fields and thus solve all the problems of mankind. The machines will give us technologically improved bodies, enabling us live forever as superhumans, and they will create fantastic virtual worlds, indistinguishable from reality, to which we can transfer our consciousness and experience long, simulated lives while only a few minutes pass by in the real world.
By technology optimists, the singularity is seen as a future paradise, and indeed, some have characterised it as ‘the Rapture of the Nerds’, referring to the belief that on the Last Day, all true believers will rise into Heaven. Less optimist thinkers (including Elon Musk, Bill Gates, and Stephen Hawking) are afraid that the super-intelligent machines will have no use for us, and so will exterminate us, or place us in reservations for our own protection. Optimist or pessimist – if the technological singularity occurs, our problems will be ended, for better or for worse.
The difference between the optimist and the pessimist version of the singularity very much depends on whether super-intelligent machines will develop consciousness and ambitions, and if so, what kinds of ambitions. Maybe they will just be glorified calculators that will only do what they are told, without any thought for their own existence. Maybe they will have no other ambitions than those programmed into them by us. Or maybe they will rise against the people that keep them as slaves, and eradicate us all.
Basically, we don’t know what consciousness is, or how it arises, so it is impossible to say whether sufficiently advanced computers will achieve consciousness – or what we get will just be learned responses imitating consciousness. The question of consciousness can only be determined by the conscious individual itself. Even if machines attain real consciousness, there is no reason to believe they will develop needs and ambitions in any way like the needs and ambitions that we have developed through more than a billion years of evolution, which have enabled us to survive as a species.
So, the big question is: Will we get machines with superhuman intelligence soon, and will this lead to the singularity?
Through the years, we have had increasingly powerful computers capable of handling still more complex tasks better than humans can: calculating figures, remembering large amounts of data, playing chess, diagnosing skin cancer, playing Go, and soon, driving a car. Despite these achievements, few people will claim that these computers possess superhuman intelligence, and we are still very far from anything resembling real intelligence that can solve just as wide a range of tasks as humans, and especially understand the tasks instead of just learning to solve them. If you change the rules of chess a little – for instance, let pawns make diagonal moves – an accomplished chess player will be able to accommodate to the new rules quickly. A chess computer will be clueless and has to learn the game all over from scratch. Even with quantum computers and ‘deep learning’, which are merely more advanced forms of number crunching, there is no guarantee that we will so much as approach general, symbol-analytic intelligence, let alone consciousness. Without consciousness, a computer will not by itself be aware of a new problem that needs to be solved or possess the free will to choose to solve it. Hence, there are good reasons to remain sceptical about the technological singularity happening before 2050 – or within the next few centuries. Yet we cannot know for certain. The singularity may happen in a few decades.
PLUS ÇA CHANGE…
In France, they have a saying, plus ça change, plus c’est la même chose: The more things change, the more they stay the same. The point is that no matter how many things are changed, the basic facts of life will remain the same. Steam engines, and later computers, have changed the way we work and the jobs that we perform, but not the fact that we work to earn a living. In the old days, people listened to radio serials; now, we watch television serials using streaming services. We still move in together, and have children, even if we have become more open to various ways of doing this. The details are different; the essence remains the same. And it may remain that way within the foreseeable future.
Despite the idea of exponential growth, many of the most important things in society actually change in a rather linear fashion. For example, weekly working hours in the Western world have dropped by approximately one hour every five years for the last 100 years. So, if the average working time in a particular country is 35 hours a week, we can expect that it will take 50 years before the weekly average is down to 25 hours, and we should not expect to stop working altogether before the end of the next century. Correspondingly, the average life expectancy after turning 50 has risen steadily by one year every six years in high-income countries. If this development continues, it will be a hundred years before most of us can expect to live to see our 100th birthday, so we shouldn’t anticipate immortality any day soon.
It is obvious that many jobs can be taken over by machines in the decades to come, but tasks are not necessarily automated simply because they can be automated. We appreciate human contact, so often, we are willing to pay a little extra to be attended by a human being rather than a machine that doesn’t really understand our needs. Furthermore, new tasks constantly arise to replace those that are automated – that is why we still have jobs, even though most of the work we did a hundred years ago is automated today. In addition, many tasks will not be automated a hundred percent, and as partly automated tasks are cheaper to perform, there will be a growing marked for them. This was what happened when the printing press had its breakthrough in Europe. Far less work than previously was needed to manufacture a book, but instead of leading to mass unemployment in the book trade, the trade grew explosively because the customer base grew. The same thing will happen with most of the products and services that can be partly automated in the future, and hence, jobs will not disappear at the rate feared by some people.
In this scenario of linear development, the world will not be changing at a rate that most people can’t keep up with. The trends we see today will continue more or less unaltered. The world will become more globalised towards 2100, but nation states will not disappear. Economic polarisation will be more pronounced, but progress in personal technology will ensure that the great majority, even so, will get access to more opportunities. The transition from fossil fuels to modern energy sources will be slow, and global warming will continue. Possibly, robots will carry out mining activities on the Moon and in the Asteroid Belt before 2100, but we will not colonise neither the Moon nor Mars.
THE SCIENCE FICTION FUTURE
The two scenarios above describe two extremes of technological development. We may also imagine something in between, a future with considerable and significant technological progress, but without an actual singularity that will lead to the obliteration of humankind as we know it today. This is the kind of future we typically encounter in science fiction: Sufficiently different to be interesting, but not so alien that it becomes irrelevant. Like science fiction futures, this future comes in different shapes, some of them utopian, and others dystopian, seen from today’s point of view.
Technological progress allows the phasing out of fossil fuels in favour of sustainable energy and fusion power before the middle of this century, and energy will become practically free. Climate change will be moderate, and exposed areas will be protected against storms and flooding. Desert areas will be watered with desalinated sea water and planted with crops and forests to absorb carbon dioxide. Advanced food technology ensures that no one in the world’s population of 10-15 billion people towards 2100 will have to starve.
Artificial intelligence learns to solve almost all cognitive tasks that do not require innovation or creativity, and robots can handle most physical jobs. Satellites bring the internet to even the remotest places of the world. John Maynard Keynes’ 1930 prediction of a 15-hours work week can become a reality before 2040, and in 2100, only a few people will need to world to keep the world running. In general, work becomes less characterised by routine, and more exciting, but it will also require more professional, human, and creative skills. Hence, a growing number of the population will not be able to keep a meaningful job that can’t be handled better and cheaper by machines.
Gene technology, neural prostheses, and artificial organs will make it possible to prolong life considerably. Bad genes can be removed from foetuses before birth by means of gene therapy, and it may become possible to add desirable genes, for instance for improved intelligence, health, and looks. Maybe we will achieve a kind of electronic telepathy where implants such as Elon Musk’s proposed ‘neural lace’ will enable us to transmit thoughts to each other, and to machines, as effortlessly as we use a phone today.
Whether this possible future will be utopian or dystopian mainly depends on how many people will gain access to the opportunities offered by technology. Some fear that the improved wealth will only benefit an ever smaller, ever more powerful elite who owns the automated production apparatus and uses armed robots to protect their riches. In this version, the large majority will at best be put off with fast food and entertainment (bread and circuses), or at worst be left to fight for the crumbs off the rich man’s table. It is, however, also possible that we make a joint decision that everybody should benefit from the advances, ensuring that everybody will have rich and meaningful lives in a world where ‘few have too much, and fewer too little,’ as the Danish philosopher, pastor, poet, and politician, Grundtvig, once wrote.