Feb 28, 2015
Feb 27, 2015
Retail giant Amazon has floated the concept of equipping trucks with 3D printers to create products on demand and on location in an effort to them to customers faster.
According to patents filed this week with the US Patent and Trademark Office, first reported by3Dprint.com, Amazon is looking for a way to reduce delivery times by utilising 3D printing.
Customers could order an item from Amazon and have it delivered that same day - without Amazon needing to stock the inventory or ensure the product's availability.
In its patent filing, Amazon highlighted current challenges raised by being forced to store a large inventory, time delays resulting from locating items its warehouses, and their effect on the speed of delivery and resulting customer satisfaction rates.
RECESSION is when your neighbour loses his job. Depression is when you lose yours, Ronald Reagan quipped. Catastrophe, Reagan might have added, is when you, your neighbour and half the other people in your street lose their jobs too.
We are, I’m afraid, on the brink of such a catastrophe. A 2013 study of 700 professions by two Oxford researchers, Carl Frey and Michael Osborne of the Martin School, warned that 47 per cent of all jobs in the US and Britain are at risk because of computerisation.
The race against the machine has begun. And we are being outrun, outgunned and outflanked by today’s increasingly widespread network of digital devices and algorithms.
“It is an invisible force that goes by many names,” wrote Derek Thompson, of the magazine The Atlantic. “Computerisation. Automation. Artificial intelligence. Technology. Innovation. And, everyone’s favourite, robots.”
Or maybe we should just call it big data. “We identified several key bottlenecks preventing occupations from being automated,” Frey and Osborne noted ominously. “As big data helps to overcome these obstacles, a great number of jobs will be at risk.”
“The robots are coming and will terminate your jobs,” warns the normally cheerful British economist Tim Harford in typically forthright language. Harford is right. The robots are indeed coming, and computerisation, automation, artificial intelligence and big data are about to destroy many of our livelihoods.
You’ve probably heard this warning before. But this time it’s different. This time there are billions of reasons — about 50 billion reasons within the next five years, to be precise — why we should all be terrified of computerised artificial intelligence.
I’m not alone in my fears, which The Wall Street Journal columnist Daniel Akst calls “automation anxiety”. Some of our leading scientists and technology entrepreneurs are taking this anxiety to an apocalyptic conclusion.
Stephen Hawking, Britain’s highest profile living scientist, for example, warns that what he calls “full artificial intelligence” could “take off on its own” and “spell the end of the human race”. Things could get so bad, Hawking fears, that we may need to “expand our horizons” to another planet if, indeed, “we are to have a future”.
Elon Musk, a Silicon Valley entrepreneur who is chief executive of Tesla electric cars, is equally fearful, warning artificial intelligence is “summoning the demon” and, being “more dangerous than nukes”, represents humanity’s “biggest existential threat”.
Microsoft co-founder Bill Gates concurs with Musk and admits he does not understand why more people are not concerned about the impact of artificial intelligence on jobs.
Google’s executive chairman, Eric Schmidt, told the 2014 World Economic Forum in Davos that the “race between computers and people” will be the “defining one” for the next quarter-century.
The study of this imminent techno-apocalypse has now captured the attention of leading Oxbridge boffins. Cambridge recently opened a Centre for the Study of Existential Risk, funded by the Skype co-founder Jaan Tallinn, which studies risks to our entire species with a particular focus on artificial intelligence.
Oxford’s Future of Humanity Institute has published a study envisaging the ways in which the world could be destroyed. The probability of artificial intelligence bringing the world to an end was estimated at as high as 10 per cent.
The Oxford study — like Hawking’s prediction that a disaster to our planet will be a “near certainty” in the next 1000 or 10,000 years — was speculative, perhaps even science fictional. But robots aren’t just inchoate threats, a distant horde of mechanical orcs on the theoretical horizon.
As Frey and Osborne warn, not only are they at our gates, but they are also in our gates, in our homes, in our cars and, most eerily, in our pockets. Like so much else about our hyperconnected and technologically saturated world, the reason for this lies with Moore’s law — the prescient 1965 observation by Gordon Moore, the co-founder of Intel, that the number of transistors on a processor of a given size will double every two years.
Moore’s law has enabled the ubiquity of network computing, driving our reliance on desktop computers, then laptops and increasingly powerful mobile devices. Today Moore’s law is creating what in Silicon Valley is described as the “internet of things”: a networked world in which increasingly intelligent inanimate objects — from cars to clothing to buildings to cities — are connected.
By 2020, according to Swedish telecommunications giant Ericsson, there will be 50 billion connected devices in the world. These are the 50 billion reasons we should be terrified of computerised artificial intelligence.
In their bestselling 2014 book, economists Erik Brynjolfsson and Andrew McAfee have called this epoch The Second Machine Age. We are on the brink today of the age of thinking machines, of a networked society in which everything and everyone will be connected on a ubiquitous global electronic grid.
“The exponential digital and recombinant powers of the second machine age have made it possible for humanity to create two of the most important one-time events in our history,” Brynjolfsson and McAfee note. “The emergence of real, useful artificial intelligence and the connection of most of the people on the planet via a common digital network.”
Brynjolfsson and McAfee argue that each of these changes in its own right has a profound historical significance. But when combined, they explain, “they are more important than anything since the Industrial Revolution, which forever transformed how physical work was done”.
The problem may not be quite as full of cinematic drama as Hawking or Musk fears. We are not on the brink of “singularity” — that moment when machines become more intelligent than humans and, as if in a digital remix of Mary Shelley’s Frankenstein, reinvent themselves as our master.
It might not yet be time to flee to another planet. Particularly since it’s not clear what the employment situation is on Mars or Jupiter for the accountants, legal experts, technical writers and other white-collar occupations that, according to Frey and Osborne, will be most vulnerable to digital destruction.
The transformation to the second machine age is dominated by the shift from an economy based on human expertise to one dominated by intelligent machines. The human meritocracy of the 20th-century information economy is being replaced by a machine-centric capitalism.
Marc Andreessen, co-founder of Netscape and a prominent Silicon Valley venture capitalist, boasts that “software is eating the world”. But, in truth, software is eating many of our jobs and failing to replace them. “The prevailing methods of computerised communication pretty much ensure that the role of people will go on shrinking,” notes influential American technology critic Nicholas Carr in his latest book, The Glass Cage.
Carr describes a digital age in which attorneys, business executives and doctors are being usurped by algorithms. He explains that legal firms are using software from companies such as Lex Machina that replaces the expertise of the senior litigator with algorithms able to predict the outcome of patent lawsuits.
“Society is shaping itself to fit the contours of the new computing infrastructure,” Carr warns. “The infrastructure orchestrates the instantaneous data exchanges that make fleets of self-driving cars and armies of killer robots possible. It provides the raw materials for the predictive algorithms that inform the decisions of individuals and groups. It underpins the automation of classrooms, libraries, hospitals, shops, churches and homes.”
Brynjolfsson and McAfee echo some of Carr’s concerns. It’s “not implausible”, they say, that “Dr Watson”, the medical version of IBM’s Watson — the “cognitive system” that participated in the US television quiz show Jeopardy! — “might one day be the world’s best diagnostician”. What becomes of the human doctors who are replaced by Dr Watson? What becomes of diagnosticians replaced by an IBM cognitive system?
Oddly enough, it’s the most skilled workers who will be most vulnerable in the second machine age. This irony — known as Moravec’s paradox in homage to Austrian robotics expert Hans Moravec — is based on the disconcerting reality that what we once considered “high-level reasoning” requires little computational sophistication to replicate.
“The main lesson of 35 years of AI research is that the hard problems are easy and easy problems are hard,” Canadian cognitive scientist Steven Pinker explains. “As the new generation of intelligent devices appears, it will be the stock analysts and petrochemical engineers and parole board members who are in danger of being replaced by machines. The gardeners, receptionists and cooks are secure in their jobs for decades to come.”
But despite Moravec’s paradox not all unskilled jobs are safe. Automated, self-driving cars will replace cabbies and delivery drivers. Machines are already replacing workers in factories around the world. Foxconn, the gigantic Chinese electronics manufacturer, has said it will replace a million workers with robots.
Jeff Bezos, Amazon’s chief executive, is pioneering robots in his distribution centres and promised investors he would be “employing” 10,000 robots by the beginning of this year. “Amazon’s warehouse jobs are gradually being taken over by robots,” warns George Packer of The New Yorker, thereby completely “eliminating the human factor from shopping”.
Bezos is also experimenting with automated drones that would make tens of thousands of delivery drivers redundant. “I know this looks like science fiction,” he says, “but it’s not.”
What really looks like science fiction is the future of labour. “You’ll be paid in the future depending on how well you work with robots,” according to Kevin Kelly,Wired magazine’s “senior maverick”. But for every “senior maverick” able to work with computers there will be a legion of teachers, lawyers, accountants and diagnosticians whose skills will be increasingly redundant in the age of the intelligent machine.
The political ramifications are particularly troubling. Financial Times economist Martin Wolf warns that intelligent machines will hollow out traditional middle-class jobs, compound income inequality, make the wealthy indifferent to the fate of the rest of society and make a mockery of democratic citizenship.
“Average is over,” notes American economist Tyler Cowen about a new world in which the “key divide” is between 10 to 15 per cent of people who can “manage computers” and everyone else. Cowen describes this new elite as a “hyper-meritocracy” of people who can work effectively with artificially intelligent machines.
In today’s increasingly automated economy, the relative egalitarianism of our industrial age will be replaced by a social order more akin to feudalism. “We can expect job growth in personal services,” Cowen predicts. “This will mean maids, chauffeurs and gardeners for the high earners.” It will be a world of “billionaires and beggars”.
Cowen’s feudal vision is replicated in the broader networked economy. The network effect has created a winner-takes-all economic system in which a tiny proportion of Silicon Valley companies such as Google, Facebook and Amazon are dominant.
The real race today in Silicon Valley is to control this new robot economy. In 2012 Amazon paid $775 million for Kiva Systems, the maker of intelligent machines for servicing warehouses, whose robots it is also using in its own distribution centres. Facebook, too, is aggressively pursuing opportunities. In 2014, for example, it acquired Oculus VR, a virtual reality company, and British-based pilotless drone company Ascenta.
Facebook’s Mark Zuckerberg has also invested in Vicarious, an artificial intelligence company that, according to its founders, will “learn how to cure diseases, create cheap renewable energy and perform jobs that employ most human beings”. What isn’t clear, however, is what exactly we humans will do all day when every job is performed by Vicarious.
It is Google, the dominant technology company of our networked age, that has been the most aggressive in controlling the robot economy. In the second half of 2013, Google acquired Boston Dynamics, a producer of militarised robots, as well as seven other robotics companies. At the beginning of 2014 it paid $500m for DeepMind, a British company with a strong focus on AI. Last year Google also acquired the leader in smart home technology, Nest Labs, for $3.2bn.
Then there is Google’s championing of self-driving cars as well as its $250m investment in Uber, which, some speculate, could be used as a global transport platform for automated vehicles. As a company with a $550 billion market cap that employs fewer than 50,000 people, Google’s core value is based on its algorithm rather than its labour force. It may not represent the planet’s biggest existential threat or spell the end of the human race, but Google exemplifies the way ordinary working people are losing the race against the machine.
“Don’t fear artificial intelligence,” says Ray Kurzweil, Google’s director of engineering and an evangelist of robot technology. He believes AI is making the world a better place by improving the diagnosis of disease, developing renewable clean energy, cleaning up the environment and providing high-quality education to people.
“We have the opportunity to make major strides in addressing the grand challenges of humanity,” Kurzweil argues. “AI will be the pivotal technology in achieving this progress.”
Kurzweil is wrong. Rather than saving the world, the automated economy is provoking a crisis of unemployment and inequality. In 1930 economist John Maynard Keynes wrote: “We are being afflicted with a new disease of what some readers may not yet have heard the name, but of which they will hear a great deal in the years to come — namely technological unemployment.”
Today, technological unemployment is back — only this time there are no world wars to give people work. As Keynes’s biographer, Robert Skidelsky, has warned about the contemporary capitalist system: “Sooner or later we will run out of jobs.”
Kurzweil argues that we have a “moral imperative” to realise the promise of artificial intelligence. But in today’s increasingly automated networked economy the real moral imperative is to create a world of full employment. There can be no progress when we are being outrun, outgunned and outflanked by the machine. There can be no progress when it appears that we have finally run out of jobs.
Feb 26, 2015
China has dropped some of the world’s leading technology brands from its approved state purchase lists, while approving thousands more locally made products, in what some say is a response to revelations of widespread Western cybersurveillance.
Others put the shift down to a protectionist impulse to shield China’s domestic technology industry from competition.
Chief casualty is US network equipment maker Cisco Systems, which in 2012 counted 60 products on the Central Government Procurement Center’s (CGPC) list, but by late 2014 had none, a Reuters analysis of official data shows.
Smartphone and PC maker Apple has also been dropped over the period, along with Intel security software firm McAfee and network and server software firm Citrix Systems.
The number of products on the list, which covers regular spending by central ministries, jumped by more than 2,000 in two years to just under 5,000, but the increase is almost entirely due to local makers.
The number of approved foreign tech brands fell by a third, while less than half of those with security-related products survived the cull.
An official at the procurement agency said there were many reasons why local makers might be preferred, including sheer weight of numbers and the fact that domestic security technology firms offered more product guarantees than overseas rivals.
China’s change of tack coincided with leaks by former US National Security Agency (NSA) contractor Edward Snowden in mid-2013 that exposed several global surveillance programmes, many of them run by the NSA with the cooperation of telecom companies and European governments.
“The Snowden incident, it’s become a real concern, especially for top leaders,” said Tu Xinquan, Associate Director of the China Institute of WTO Studies at the University of International Business and Economics in Beijing. “In some sense the American government has some responsibility for that; (China’s) concerns have some legitimacy.”
Cybersecurity has been a significant irritant in US-China ties, with both sides accusing the other of abuses.
US tech groups wrote last month to the Chinese administration complaining about some of its new cybersecurity regulations, some of which force technology vendors to Chinese banks to hand over secret source code and adopt Chinese encryption algorithms.
The CGPC list, which details products by brand and type, is approved by China’s Ministry of Finance, the CGPC official said. The list does not detail what quantity of a product has been purchased, and does not bind local government or state-owned enterprises, nor the military, which runs its own system of procurement approval.
The Ministry of Finance declined immediate comment.
“We have previously acknowledged that geopolitical concerns have impacted our business in certain emerging markets,” said a Cisco spokesman.
An Intel spokesman said the company had frequent conversations at various levels of the US and Chinese governments, but did not provide further details.
Apple declined to comment, and Citrix was not immediately available to comment.
Industry insiders also see in the changing profile of the CGPC list a wider strategic goal to help Chinese tech firms get a bigger slice of China’s information and communications technology market, which is tipped to grow 11.4per cent t to $465.6 billion in 2015, according to tech research firm IDC.
“There’s no doubt that the SOE segment of the market has been favouring the local indigenous content,” said an executive at a Western technology firm who declined to be identified.
The executive said the post-Snowden security concerns were a pretext. The real objective was to nurture China’s domestic tech industry and subsequently support its expansion overseas.
China also wants to move to a more consumption-based economy, which would be helped by Chinese authorities and companies buying local technology, the executive said.
Policy measures supporting the broader strategy include making foreign companies form domestic partnerships, participate in technology transfers and hand over intellectual property in the name of information security.
Wang Zhihai, president and CEO of Beijing Wondersoft, which provides information security products to government, state banks and private companies, said the market in China was fair, especially compared with the United States, where China’s Huawei Technologies, the world’s largest networking and telecoms equipment maker, was unable to do business due to US. security concerns.
Local companies were also bound by the same cybersecurity laws that US. companies were objecting to, he added.
The danger for China, say experts, is that it could leave itself dependent on domestic technology, which remains inferior to foreign market leaders and more vulnerable to cyber attack.
Some of those benefiting from policies encouraging domestic procurement accept that Chinese companies trail foreign competitors in the security sphere.
“In China, information security compared to international levels is still very far behind; the entire understanding of it is behind,” said Wondersoft’s Wang.
But Wang, like China, is taking the long view.
“In 10 or more years, that’s when we should be there.”