Now Begins The Revolution

“Travis Solorio, a 26-year-old of Latin American origin, was shot dead on Sunday, February 5 on Tamarack Avenue in Sun Valley, according to information from the medical examiner in Los Angeles.” Does there appear to be something odd about this text passage, taken from the LA Times in February 2012? Is there any eagle-eyed person out there who could see that it was not written by a person? This piece about the fatal shooting of Travis Solorio was produced by a computer that automatically retrieves facts from the Los Angeles coroner and converts the data to new prose. According to the British website,, news robots are now making inroads into the increasingly digital news world: ”In five years, a computer program will win the Pulitzer Prize”, a scholar was quoted on the site. But surely it is one thing to publish minimal hack prose, and quite another to create great fiction? Well, it depends on whom you ask. Inside Science reported in January on a new computer algorithm that could be used to predict which books would be commercially successful. From this plateau, of course, the leap is even shorter to a program that itself can write these successful books without too many significant human interventions. (An interesting observation here is that the algorithm found that successful books use many adjectives – a claim that is likely to have traditional writers and writing teachers tearing their hair out.) On the radio station P1’s ”Stakeout” show, Jessika Gedin said recently about how creatively set computers have started writing Shakespeare sonnets. The program used so-called machine learning, and produced – completely without otherwise necessary poetry gear such as a quill, red wine and unrequited love – lines like these:

“When I in dreams behold thy fairest shadeWhose shade into dreams doth wake the sleeping mornThe daytime shadow of my love betray’dLends hideous night to dreaming ’s faded form.”

The next step is maybe to teach the machines to avoid careless assonance of the type ”morn” and ”form”. Reasoning such as above is rampant in the newly published The Second Machine Age. Work, Progress, and Prosperity in a Time of Brilliant Technologies, which tries to lock on to one of the most elusive targets humanity has ever known: societal development as a result of applying digital technology. The first machine age was the age of the steam engine – the latter half of the 1700s. The technological advances that took place before that may have been important, but this was an innovation that ultimately, and for the first time, fundamentally changed the human condition. An invention that, when it generated the industrial revolution, ”allowed us to overcome muscle power’s limitations” and brought us into the modern age.

Now, we are living through the second machine age. Computers have been with us for a while, but, just as it took a few generations for the steam engine to start the industrial revolution, it is only now that computers and digital technology are becoming really revolutionary. What is striking, and only a little nerve-wracking, about The Second Machine Age is the authors’ wide-eyed, uninhibited enthusiasm. It is doubtful that even the most well-trimmed processor could create a text so saturated by ”astonishing”, ”unprecedented”, ”life-changing potential”, ”faster, cheaper, smaller, lighter… ”. Yes, computers are beginning to truly become really really fast, but where is the critical gaze? People today have unprecedented opportunities to ”share their insights” online, cheer Erik Brynjolfsson and Andrew McAfee, but seem to forget that the connected mostly seem to want to use their new superpowers to post ‘selfies’, where they eat pulled pork. But, after further reading, it is not difficult to think of an idea. Is this really a time for cynicism? Is there any other reasonable approach to what is currently being done in the digital evolution than just wide-eyed enthusiasm? ASCI Red was presented in 1996, it cost $55 million and was the world’s fastest supercomputer. Its size is about 80 percent of a tennis court, it is used to simulate nuclear weapons tests, and it can reach a speed of 1.8 teraflops. If your knowledge of computing capacity is limited, just take my word for it: it was really, really wailing a lot. Nine years later, another PC reached 1.8 teraflops; it bore the name of Sony Playstation 3. Today, it costs $1,700 online. With accessories included. We live in a time of exponential growth. The cost of computing power is decreasing with dizzying speed, the computers get faster and data is everywhere abundant. And the more data available, the better and more useful the computer programs. One could write algorithms everywhere it was alleged that machines would never be able to cope: landing an airplane, defeating a human at chess, driving a car through traffic, writing poetry. Time and again, the have computers convinced us. People talk about the death of ideology, but a heated global ideological battle is going on right now. Partially, but not slavishly, it follows a right/left line. On one side, usually the left, stand the development pessimists – we can somewhat bluntly call them the Nina Birch School. Because of the techno-capitalist complex, we will all be unemployed, inequality will increase, the earth will warm like a sourdough loaf in the oven and, for those who know their Terminator: on August 29, 1997, ‘Skynet’ becomes aware of its own existence and starts its war against humanity.

The other side is optimistic. We can call it the Hans Rosling School. All the evidence shows that the world is just becoming more peaceful, more diseases are being cured, capitalism is leading to unprecedented global prosperity and the disadvantages caused partly because of technology will, by far, be offset and eliminated by other technologies. The Second Machine Age primarily supplies ammunition to this second group, to all us optimists out there. “Every generation has perceived the limits to growth that finite resources and undesirable side effects would pose if no new […] ideas were discovered. And every generation has underestimated the potential for finding new […] ideas. […] Possibilities do not merely add up; they multiply.” (Quoted from a key passage by economist Paul Romer.) The second machine age differs from the first in that, this time, it is not only people in traditional manual occupations that are losing their jobs; now, it is even those with white cuffs, including, notoriously, news hacks and sonnet poets. The authors reason on this and note that it should be okay – but the parties feel dutiful and do not differ much from other theses that are adopted, and abandoned in the field. It’s much more fun to go with them in their childish delight at being in the eye of the technology hurricane, at the heart of computer action. It is always from far away that those living in the miracle can see the miraculous. In the book’s most powerful metaphor, they take up the well-known story of the emperor and the inventor of chess. The latter wanted to ‘just’ be rewarded with a grain of rice on the board first square, two on the next, four on the next and so on, with a doubling in each box. After half a board, 32 squares, the emperor is up to four billion grains; this is manageable and only corresponds to approximately one rice field. But then, in the second half, things really start to happen. If the Emperor were to fulfil the wish, he would need more rice than mankind has ever produced since the first rice plant stretched toward the sun. The time of exponential growth. One can also, like Erik Brynjolfsson and Andrew McAfee, call it the chessboard’s second half. It’s exciting terrain – albeit very demanding for sonnet writers.

Erik Helmerson

Ledarskribent i Dagens Nyheter.

Mer från Erik Helmerson

Läs vidare