Fully Automated Luxury Communism A Manifesto

The concept of exponential growth, given its rarity in nature, is dificult to immediately grasp. It is most clearly explained in the ‘wheat and chessboard problem’, first outlined by Ibn Khallikan in the thirteenth century. Some claim this ‘problem was in fact a historical event involving the Emperor of the Gupta Empire and an encounter with the inventor of the game of chess, or a similar precursor.

Supposedly, the Emperor, impressed by the demanding nature of the game he had been shown, told its creator to name their reward. The response he received was as simple as the game was complex: ‘place one single grain of rice on the first square of the board, two on the second, four on the third and so on’. With each successive square the number of rice grains was to double – 1, 2, 4, 8, 16, 32 – until the final square of the board was reached. The Emperor, surprised at such a humble request, happily agreed.

The Three Disruptions

41

It quickly became clear, however, that such a prize was far greater than he had anticipated. After thirty-two squares, only halfway up the board, the game ’s architect had earned 4 billion grains of rice. While a large number, that was still only equivalent to the amount produced by a large field, and this only served to place the inventor in even higher esteem – after all, a field or two of rice was a perfectly satisfactory reward for such a captivating game. That was to change, though, when by the final square the tally was 18 quintillion grains of rice, a pile larger than Mount Everest and more rice than had been produced in history. The Emperor, enraged by the temerity of a subject who had asked for more wealth than even he could ever ofer, ordered the inventor to be executed.

This allegory captures the swift, and often unexpected, dividends of exponential growth, especially compared to linear forms of progress which the human mind is far more inclined to expect. So what happens when such prodigious growth occurs in human afairs? The answer can be found in the history of computing over the last half a century.

In 1965 Gordon Moore, who would later co-found Intel, wrote an article for Electronics Magazine detailing recent improvements in the performance of computer chips. At that time the most complex circuit still only had around thirty components, but progress appeared to be accelerating. In fact, Moore observed that the recent rate of development had been so rapid that the number of transistors that could fit on a circuit had doubled every year since 1959. That discovery got him thinking. What would happen if that same trend, of annual doubling, prevailed for another decade?

After some quick calculations, Moore was shocked by the answer. His forecasts showed that by the end of 1975 the average circuit would have gone from having thirty transistors to 65,000. Moore speculated about the kinds of technology such mesmerising advance could make possible, contemplating a world with ‘portable communications equipment’, ‘home computing’ and perhaps even ‘automatic controls for automobiles’. Unfortunately for Moore, his prediction proved wildly inaccurate. The trend he outlined didn’t persist for another ten years – it’s been going for a half century and counting.

42

chaos under heaven

When Moore wrote his seminal article, a single transistor spanned the width of a fibre of cotton and cost eight dollars in today’s prices. Now, by contrast, billions of transistors can be squeezed onto a chip the size of a fingernail, with a single human hair 10,000 times thicker than Intel’s next generation of products. And the cost per unit? That’s plummeted too, falling to a tiny fraction of a cent. So while you’ll often hear clichés of how modern smartphones are more powerful than the computers used for NASA’s Apollo missions, even that fails to convey how dramatically transistors have transformed over the last few decades.

A more useful comparison can be found between the supercomputer ASCI Red and the PlayStation family of games consoles. The former, built by the US government in 1996, was the first machine able to process a teraflop – a trillion floating-point calculations per second. Costing \$55 million and measuring the size of a tennis court, its purpose was to predict and model nuclear explosions, something it did with ease as it remained the world’s fastest computer until the turn of the millennium – staying in use until as recently as 2005. And yet just one year later, the same processing power was available to consumers in a PlayStation 3, a games console available for as little as \$600. The PlayStation 4, released in 2013, was almost twice as powerful as both its predecessor and ASCI Red. Coming in at \$400 it cost 1/100,000th of the world’s leading supercomputer only two decades earlier.

The Three Disruptions

43

Such a rapid rate of development is only possible because improvements in processing speed have experienced exponential rather than linear gains over the last sixty years. It is this quality, first observed in computing by Moore, that is powering the Third Disruption far more quickly than many thought possible. Its consequences reach far beyond video games.

While progress over the last half century has been dizzying, the parable of rice grains on the chessboard remains instructive. If such trends persist for another six decades, the results – like the pile of rice bigger than Everest – are almost beyond comprehension. If that single field of rice halfway up the board represents global real-time communication and millions of industrial robots, then what is the mountain?

Can Moore’s Law Endure?

The transformative power of Moore ’s Law, should it persist, is inarguable. The key question, then, is how much longer it can endure. In 2015 researchers at Intel foresaw it prevailing for at least another ten years, although by the standards of a trajectory more than five decades old, that hardly counts as optimistic. A year later William Holt, the company’s CEO, was less confident, claiming it might only carry on for another five years and, at best, would significantly slow down thereafter (although he believed progress elsewhere, in areas such as energy eficiency, were likely). That would seem a formidable challenge to more optimistic projections, and if Holt is right our present field of rice will only grow to five or six by the middle of this century. An immense improvement, but certainly not exponential.

44

chaos under heaven

Yet there have been Cassandras predicting the demise of Moore ’s Law for decades. So far they have been proven consistently wrong, with new avenues for improvement opening up just when it seemed any hope for further advance was blocked. Until 2004, increases in the clock speed of chips significantly contributed to enhanced performance, the downside being that overheating placed a limit on how far that innovation could persist. In response, manufacturers incorporated more processor ‘cores’ as the primary means for accelerating power, with processors now working on diferent operations in parallel with one another.

It will take similar kinds of innovation to maintain Moore ’s Law, even if it continues to slow down slightly – something which, in his defence, Holt conceded. While within a decade it may become impossible to miniaturise individual transistors any further, simply because of physical limits, adaptations such as 3-D circuitry and quantum computing – both proven concepts – could mean exponential growth continues. Perhaps even beyond the last square of that chessboard.

More than Processing

Because digitisation is a general-purpose phenomenon, it is not just computer chips that have been subject to its incredible powers of transformation. A similar trend is in evidence with internet bandwidth, where user capacity has grown by between 25 and 50 per cent every year since 1983. The same holds true with data storage, which has likewise enjoyed an exponential function in space-to-cost ratio, with a gigabyte of storage falling from around \$200,000 in 1980 to just \$0.03 in 2014.

More than anywhere else, however, it is in storage that progress has visibly started to slow down. Even if Toshiba’s

The Three Disruptions

45

3-D Magnetic Recording technology – where a magnetic head writes and reads data on stacked layers using microwaves – is commercially scalable, that would still mean maximum storage drives of hundreds of terabytes. Again, that may be impressive, but it certainly is not exponential.

But while a paradigm shift might be needed in storage, which slows progress in the short term, that could mean little in the broader picture. As impressive as digital storage is, we know that compared to storing data as DNA – which can be presumed as a hypothetical limit – we have barely scratched the surface. While technology like that might not be on your laptop anytime soon, the potential is astonishing, with a single gram of human DNA able to store 215 petabytes (215 million gigabytes) of information. This is not the realm of abstract speculation, and humans have been able to store data as DNA since 2012, when Harvard University geneticists encoded a 52,000-word book using strands of DNA’s fourletter alphabet of A, G, T, and C to encode the 0s and 1s of the digitised file.

While such progress might not have applications in the foreseeable future, here too discounting the possibility of going well beyond the final squares of the chessboard – in bandwidth and storage as well as processing speeds – appears unwise. It seems increasingly reasonable to presume that the primary constraints on technological advance are the laws of physics. For now, they remain a long way of.

The Power of Experience

Change doesn’t have to be exponential to be transformative in the context of the Third Disruption. Around the same time Gordon Moore made his forecast about the future of computing, Bruce Henderson – founder of the Boston Consulting

46

chaos under heaven

Group – developed a concept that would come to be referred to as the Henderson Curve (more recently the Experience Curve). Based on observations he made while working with his clients, this soon became a sophisticated predictive model, outlining how the costs of a manufactured good decline by as much as 20 per cent every time capacity is doubled. The variables driving that behaviour are relatively simple, ranging from greater labour eficiency to improvements in product design. While the experience curve does not ofer the same rapid transformation one sees in the exponential improvement of digital technologies, its dividend is of critical importance to extreme supply – specifically when it comes to renewable energy.

That’s because the most important area where one sees the experience curve at work is with the price of photovoltaic (PV) cells, the main consumer technology for generating solar power. Here progress correlates almost perfectly to what Henderson would have predicted, with the cost of PV falling 20 per cent every time capacity has doubled over the last sixty years. When the technology was deployed for the first time aboard NASA’s Vanguard 1 satellite in 1958, each panel was able to generate a maximum half a watt of energy at a cost of many thousands of dollars each. By the mid-1970s, that figure had fallen dramatically to \$100 per watt, still uncompetitive with fossil fuels but impressive nonetheless. Yet by 2016 the price–performance ratio of solar had been transformed, with a watt of energy from a solar array costing as little as fifty cents, making it a genuine alternative to fossil fuels in countries with abundant sunshine.

Few disagree that this trend will continue, and with global solar capacity doubling every two years – it increased by a factor of one hundred between 2004 and 2014 – a virtuous cycle between increased capacity and ever-falling prices has been established. The critical question, as with Moore ’s Law, is how much longer that will continue.

The Three Disruptions

47

What we know for certain is that, in principle, solar is more than capable of meeting the world’s expanding energy needs. Given that the same amount of potential energy hits the Earth in ninety minutes as the whole of humanity consumes in a year then, even in the event of demand doubling over the coming decades, solar might not just be the greenest means of powering our world, but the cheapest one too.

Fortunately the same changes in the price–performance ratio of solar cells also apply to the mainstream technologies of renewable energy storage, lithium-ion batteries. There, recent falls in cost only serve to strengthen the conclusion that it is a question of when, rather than if, the world transitions to renewable energy.

From Crisis to Utopia

Ours is a finite world marked by constraints. To a large extent, these constraints define the five crises set to radically shape the course of the coming century.

Together, these crises – encompassing climate change, resource scarcity, ever-larger surplus populations, ageing and technological unemployment as a result of automation – are set to undermine capitalism’s ability to reproduce itself. That is because they could dissolve some of its key features like the presumption of constant expansion and infinite resources, production for profit, and workers having to sell their labour.

In 1984 the futurist Stewart Brand made the now-iconic declaration ‘Information wants to be free.’ He would later clarify what that meant, saying,

48

chaos under heaven

On the one hand information wants to be expensive, because it’s so valuable. The right information in the right place just changes your life. On the other hand, information wants to be free, because the cost of getting it out is getting lower and lower all the time. So you have these two fighting against each other.

As we shall see, information is the basis of value under modern capitalism – far more than we think. And yet technologies under that same economic system now paradoxically tend towards destroying the scarcity of information, and therefore its value.

It’s unlikely Brand was aware of it in 1984, but Marx said something similar about the tendency of information towards extreme supply more than a century earlier:

Forces of production and social relations – two diferent sides of the development of the social individual – appear to capital as mere means, and are merely means for it to produce on its limited foundation. In fact, however, they are the material conditions to blow this foundation sky-high.

More than three decades after Brand stated his elegant observation we now know he was right – its plummeting cost shows information does want to be free. But by the middle of this century it will be increasingly clear that this also extends to labour, energy and resources too. This is the basis for a diferent set of social parameters underpinned by changes we can already observe around us: a world beyond jobs, profit and even scarcity.

The Three Disruptions

49

3 What Is Fully Automated Luxury Communism?

The goal of the future is full unemployment, so we can play.

Arthur C. Clarke

Why FALC?

Why ‘fully automated luxury communism’? Why those words and in that sequence? After all, many see communism as nothing more than a failed experiment of the twentieth century undeserving of our attention save learning from its mistakes. Some may admit that capitalism has numerous flaws, and may indeed end one day, but if communism is what comes next, that wouldn’t be an improvement.

While it is true that a number of political projects have labelled themselves communist over the last century, the aspiration was neither accurate nor – as we will go on to see – technologically possible. ‘Communism’ is used here for the benefit of precision; the intention being to denote a society in which work is eliminated, scarcity replaced by abundance and where labour and leisure blend into one another. Given