• All governments lie, but disaster lies in wait for countries whose officials smoke the same hashish they give out.

  • I.F. Stone

zondag 7 mei 2017

The Robot Economy

The Robot Economy: Ready or Not, Here It Comes

Sunday, May 07, 2017 By JP Sottile, Truthout | News Analysis 
Duc Tran, an automation engineer, observes the fully autonomous robotic truck loader during a test at the Wynright Robotics facility in Arlington, Texas, July 18, 2012. (Photo: Brandon Thibodeaux / The New York Times) Duc Tran, an automation engineer, observes the fully autonomous robotic truck loader during a test at the Wynright Robotics facility in Arlington, Texas, July 18, 2012. (Photo: Brandon Thibodeaux / The New York Times) 
September 17 changed everything.
On that day in 2013, Oxford University published an innocuously titled academic paper by two mostly unknown economists. But "The Future of Employment" wasn't just another number-crunching exercise in opacity by a couple of dreary scientists. No, their bombshell report portended a coming robot apocalypse that could change the nature of human civilization, and perhaps even human beings themselves.
Thankfully, the forthcoming carnage described by Carl Benedikt Frey and Michael A. Osborne isn't a doomsday scenario where Skynet systematically wipes out humankind, or a darkly lit near-future where attractive Replicants violently struggle to make sense of their emerging emotions in a perpetually damp Los Angeles.  
Instead, the economists previewed an all-too-real world where the second-richest man on the planet -- Amazon's Jeff Bezos -- gleefully parades around like Sigourney Weaver in a massive robotic exoskeleton built by Hankook Mirae Technology.
They presaged the impending doom from robots like Handle, the Michael Jordan-esque robot built by Boston Dynamics. Handle can leap like a superhero, can run a marathon in under three hours and, if Softbank CEO Masayoshi Son is right, will probably be smarter than you in just a few decades.
They foresaw a future with the likes of Gordon, the "first robotic barista in the U.S." Gordon can serve "about 120 coffees in an hour." They also predicted the likes of Otto, the self-driving big-rig designated by Uber to deliver truckloads of beer to thirsty consumers. And then there's Pepper, the empathic, "day-to-day" companionthat is not just working in airports and banks, but being "adopted" into Japanese homes … and even "enrolling" in school.
The Future Is Now
This is the "next economy," and, ready or not, it is coming at the double-time speed of Moore's Law. This rapid acceleration of the Fourth Industrial Revolution is transforming "The Future of Employment's" apocalyptic premonition -- that 47 percent of all jobs in the United States may be lost to automation over the next two decades -- into a solemn epitaph for the rapidly fading era of manufacturing-based, consumption-driven economics.
Dire warnings have come from Bill GatesStephen HawkingElon Musk and, most dauntingly, from cybersecurity experts who recently warned of the threat of hacked robots violently turning against people and their pets in a souped-up scenario reminiscent of The Purge. However, long before a haywire Roomba or a disgruntled Pepper comes calling, millions of workers will struggle to fend off the brutal reality of unplanned obsolescence.
This is an economy where manufacturing jobs require a college degree, artificial intelligence replaces administrative workers, automated kiosks dislodge food service workers and driverless vehicles threaten the livelihoods of up to 10 million Americans who take the wheel for a living.
This is "Industry 4.0." It's an economy where Amazon's dirigible-based distribution center perpetually floats over cities, effortlessly deploying drones to deliver stuff built by robots or 3-D printers or both. F rankly, it's Amazon's jungle out there. Its plan to "disrupt " the grocery business with almost human-free stores is just the next phase of an ongoing, Amazon-led "retail apocalypse " that's driving human-staffed brick and mortar stores into extinction. Amazon also dominates the AI-assistant market with its self-teaching, suspiciously spooky personal assistant Alexa. And now Amazon's poised to pair its increasingly automated warehouses with a move to become the leader in the self-driving future of trucking and overland shipping.
Even low-paying farming jobs could be completely upended by robotic fruit pickerswith the deft touch needed to harvest food in American and European fields. Robots are already replacing cheap migrant workers shut out by anti-immigrant policies. And new robot-staffed factories are producing modular houses, while robotic bricklayers promise to do to the construction trades what automation did to coal mining.
This is an economy where there's "an 83% chance that workers who earn $20 an hour or less could have their jobs replaced by robots in the next five years" and "those in the $40 an hour pay range face a 31% chance of having their jobs taken over by the machines," according to a 2016 report by then-President Obama's White House.
And it's not coincidental that this robot apocalypse comes at the very moment the epic economic growth we've enjoyed throughout the post-World War II era seems to be coming to an end. Peak demand in oil and "peak stuff" for consumer products may signal more than just growing energy efficiency and market saturation for cheap stuff. This may signal the spread of Japan-style economic stagnation around the developed world. This could be the end of the ecologically untenable assumption of limitless growth.
The End of the American Century
The post-war idea of never-ending growth emerged from the manufacturing boom that came with the "American Century." Perpetual government investment in the military-industrial complex created a baseline of well-paying, low and high-skilled jobs that helped raise the floor -- and consumer expectations -- for workers, while also sustaining huge corporations like Boeing, Westinghouse and General Electric. Military Keynesian was a reliable tide that lifted a lot of boats moored to an American dream of endless economic expansion. Sadly, it also conjured up a series of costly, nightmarish wars for those who paid the ultimate price both at home and abroad.
An often overlooked element, though, is the way automation helped maintain continued growth in productivity, even as wages lagged. As the Guardian recently noted, "As of 2015, a typical production worker in the US earned about 9% less than a comparable worker in 1973. Over the same 42 years, the American economy grew by more than 200%, or a staggering $11tn." This divergence between wages and productivity drove wealth inequality. But it also presented a problem for producers facing the declining purchasing power of their customers -- a.k.a. the workers not getting the wages they needed to drive, or at least sustain, continued growth in consumption.
To keep up, the manufacturing economy needed 1) a massive flood of consumer credit to create artificial purchasing power; 2) ever-lower prices on ever-more mass-produced goods so wage-stagnant workers could afford to consume the way they did when one good job per family made for a middle-class life.
To get there, jobs went to places like China. But one of the underappreciated drivers of this move to cheap labor was the coming of automation, which made American humans a far less efficient way to make things. BuzzFeed reported on a recent study by the National Bureau of Economic Research, which found that since 1990, each new robot installed in an American factory "reduced employment in the surrounding area by 6.2 workers." The only thing cheaper than replacing American humans with better technology was exploiting cheap labor markets in China and the developing world. But that's no longer true.
To wit, Oxford paired with Citibank in 2015 to produce a fuller picture of the global threat to jobs. In "Technology at Work," researchers found that 75 percent of the world's robots are "geographically concentrated" in just five countries -- "China, Germany, Japan, Korea, and the US." Of those, China is by far "the fastest growing market" with the "most significant room for future growth." In fact, they determined that "on average 57% of jobs are susceptible to automation," but the "number rises to 69% in India and 77% in China."
As if on cue, China's new "five-year plan" to be a global leader in robotics has already produced results. China leads the world in production of industrial robots, and robots are becoming a cheaper, more productive option to Chinese labor. That's the simple, 21st-century reality undermining the hollow promise to "Make America Great Again" by bringing back jobs from China. It's a great plan for growth … if you've got a time machine big enough for 320 million people.
Not only is China automating jobs at an ever-faster pace, but 3-D printing technology is radically altering the way everything gets made. Even as we begin to wrap our heads around a future filled with 3-D printed guns, researchers are moving on to 4-D printing techniques with evolving materials that can self-assemble, repair and even remake themselves at will.
Perhaps most tellingly, artificial intelligence is starting to lay waste to college-educated workers in non-manual jobs previously thought to be exempt from automation. Here are just some of white-collar casualties of the robot apocalypse:
Goldman Sachs "employs" Marcus -- a fully automated lending platform that's part of an industry-wide AI-makeover displacing humans in equities "sales, trading, and research."
Fukoku Mutual Life Insurance is turning over policy payouts to IBM's Watson Supercomputer, which has 6,000 corporate clients ranging from Hilton Hotels to Whirlpool and Visa.
* Coca-Cola's global senior digital director plans on using AI bots to crank out ads with "automated narratives" and AI is already creating commercial music and jingles.
* Jeff Bezos-owned Washington Post used Heliograf to pump out stories about the 2016 Rio Olympics before tweaking it to "cover" the 2016 Election campaign.
* The Smart Tissue Autonomous Robot (STAR) recently "outperformed" a human surgeon in a test of skill. This breakthrough augurs a near-future world where robots assist and perform a variety of surgical procedures.
Just like elsewhere, medical robots will ultimately prove they can do the job quickly, efficiently and without human burnout. The high price of doctors will be weighed against robots that perform thousands of surgeries without complaint or error, particularly since "self-taught" AI already outperforms humans in predicting heart attacks and matches them in diagnosing skin cancer. Like it is doing in manufacturing, this could mean a wholesale revolution in health care that lowers costs and increases access to highly specialized care. It's that level of relentless, Terminator-like efficiency that is disrupting the existing economic model of everything from retail and restaurants, to education and warfare.
What's Next?
The Fourth Industrial Revolution may be an era of human obsolescence for which the previous three Industrial Revolutions may not be a guide. That's because artificially created, human-modeled deep learning neural networks already generate cognitive processes so complex that even the scientists who created those networks cannot determine how, for instance, an autodidactic self-driving car makes the decisions it does when driving itself around New Jersey. In other words, we're not talking about industrial looms, steam engines or punch-cards. We're talking about cheaper, efficient, tireless and infinitely upgradeable workers that (or who?) are ready to take on nearly every human endeavor.
That's why tech leaders increasingly talk about the "New Collar" economy, where human beings not only work with intelligent machines, but also begin physically merging with machines in an effort to keep pace. While some see this coming singularity as a metaphysical event horizon for human consciousness, Elon Musk sees it as the only way we'll will be able to keep up with what is, in effect, a whole new class of forced labor … and he's got the start-up to prove it.
Sadly, forced labor may be the economic model most applicable to the next economy. As more and more jobs are turned over to AI, robots and algorithms, more and more wealth will accumulate in the hands of those already at the top of a steep pyramid. Like the pharaohs of old, these masters of the universe will profit as the cost of labor declines precipitously thanks to the robots they "employ."
As MIT Technology Review recently pointed out, this will generate windfalls for firms like Goldman Sachs where "pay of the average managing director … will probably get even bigger, as there are fewer lower-level people to share the profits with." Since they are in the business of leveraging money to make more money, they can thrive -- at least for a while -- in the whiz-bang, highly financialized world of high-speed, algorithmic market trades.
But it will also be an era when inequality hits not only human beings, but also companies. The tech world is filled with hi-tech unicorns like Uber that gallop to higher and higher valuation with just a fraction of the workforce (6,700) supported by Ford (201,000), Hertz (30,000) or one of the nation's leading trucking firms (18,000). Successful companies with the fewest employees and the most robots will complete the long process of "de-industrialization" often blamed on globalization. On the surface, outsourcing and offshoring look like the main disruptive forces opening up the income and wealth gaps. But they're really just the logical outcome of driving growth through increased productivity, instead of driving growth (and consumption) through increased wages and broader employment. Robots are simply the cheapest, most efficient and most productive phase of all.
However, the next economy doesn't have to be a doomsday scenario. It could be a sustainable economy where technology drives a stake into the heart of the hydrocarbons, drives down the cost on heath care and expands banking to underserved communities. The next economy could also see the "gig economy" transition to open-source, peer-to-peer, micro-entrepreneurial and microgrid-powered networks that eliminate economic middlemen. And the "Maker Movement" could inspire a "curated economy" of artisans, urban farmers and hyperlocal bartering that runs parallel to the robot-run, mass-produced economy. These human-made goods and services may be worth far more in a world where every mass-produced need is only a click and a hovering Amazon blimp away.
As our needs are increasingly met by hyper-efficient, AI-driven systems, the only acquisition that will truly matter is knowledge. And knowledge must become more than just a "means" to the end of getting that coveted or expected job out of high school or college -- because the means of production will no longer require human hands, human sweat or human tears. Perhaps that's why Alibaba's Jack Ma not only lamented a coming world of "pain" as the tech-driven economic disruption lays waste to the global economy, but also said education must be reformed to "raise children to be more creative and curious or they will be ill-prepared for the future." It's a future where knowledge -- and the creativity it sparks -- may have to be an end unto itself.
And while Musk is plotting for the day when augmented humans rise up to stop his vision of a Terminator-filled future, the true test will be how we as a society choose to manage -- or whether we simply continue to ignore -- the widening pain of displacement as we move from this economy to the next one. Either way, it's not only coming … it's already here.
Copyright, Truthout. May not be reprinted without permission.

JP SOTTILE

JP Sottile is a freelance journalist, published historian, radio co-host and documentary filmmaker (The Warning, 2008). His credits include a stint on the Newshour news desk, C-SPAN and as newsmagazine producer for ABC affiliate WJLA in Washington. His weekly show, "Inside the Headlines With The Newsvandal," co-hosted by James Moore, airs every Friday on KRUU-FM in Fairfield, Iowa. He blogs under the pseudonym "the Newsvandal."

Geen opmerkingen:

Een reactie plaatsen