..of the words, foolproof and incapable of error......’ has come to me from every electronic device I have tried to live with culminating in the locked-down ‘we know so very, very much better than you’ world of Apple.
We have become inured to the sights and sounds of assembly lines dominated by robots, one or more powerful arms welding cars with frightening speed and accuracy and with no human in sight. To date the reason for the absence of what a friend of mine calls RMUs (Replaceable Meat Units - aka wetware) was the fact that we humans get in the way with sometimes fatal results as hundreds of hydraulic horsepower punch a hole through the unlucky worker who strayed into their pre-programmed path. Now we have people friendly robots that no longer weld the foreman behind the bulkhead of your Kandy-Kolored Tangerine Flake Streamline Baby.
Because we forget how old we really are you may have to be reminded that robots have been a part of automotive manufacturing for over half a century. The first industrial robot—a 4,000-pound arm called the Unimate—attached die castings to car doors at a GM production line in 1961. Now even the Chinese are replacing auto workers with robots – yes they are cheaper and more accurate. China recently opened one of the most advanced automotive plants in Tianjin. The plant, belonging to Great Wall Motors, has 30 workstations occupied by 27 robots that perform more than 4,000 high-precision welding operations. The robots are so fast they can weld an entire Haval SUV in just 86 seconds. It may be rubbish but it’s well-welded rubbish.
But not just cars. Other Chinese employers are increasingly substituting robots for labour, as wages and living standards are rising – the contract manufacturer Foxconn that employs 1.2 million workers, is now investing in robots to assemble products such as the Apple iPhone.
But as an extension of the trend Robocop is being replaced by RoboWorker. Ford has turned their assembly workers into ergonomic cyborgs with robotic arms. These wearable exoskeletons carry the weight of the workers’ arms as they repeatedly grab and assemble small objects. GM is doing their part to keep up with their competitors by adopting their own robotic appendages. With the help of NASA, the Australian exiting firm has developed a robotic glove that reduces the stress of repetitive motions which might help young boys through puberty. Whist still a prototype the automated appendage comes with sensors, actuators, simulated nerves, muscles and tendons.
This is the most spectacular visible side of automation but conceals the fact the nearly every job can be, or has been, replaced by an automated process. In 1992 in one of my more challenging tasks after joining what was soon to become Telstra was to predict how many people it would eventually employ. Given that at that stage it had over 110,000 on the payroll and was unique in its operational scope compared to pretty any other telco in the world my estimate of around 45,000 was alarmingly prescient. And the same fall in employment numbers, despite increased customer size, demand and product complexity would be true of all the white collar businesses such as banks and insurance companies.
It is not just a robot apocalypse and but a potential human apocalypse that could result from so many more workers out of a employment when half of current jobs could be automated out of existence.
Yes, 47% of current American jobs could be automated in "a decade or two," according to the 2013 paper “ The future of employment: how susceptible are jobs to computerisation?” by Carl Benedikt Frey and Michael A. Osborne the question becomes: which half? And would Australia be worse given that we have largely banished manufacturing to the ‘not done in this country’ basket? And as another unrelated thought; what does it mean for a country’s strategic independence when it does not have any form of basic industrial manufacturing self reliance?
Over the past 30 years, software and robots have thrived at replacing a particular kind of occupation: the average-wage, middle-skill, routine-heavy worker, especially in manufacturing and office administration. This captures the essence of the current trend towards labour market polarization, with growing employment in high-income cognitive jobs and low-income manual occupations, accompanied by a hollowing-out of middle-income routine jobs.
But whilst the next wave of computerisation will continue to shred human work where it already has: manufacturing, administrative support, retail, and transportation the rise in big data, advanced algorithms and distributed processing and Artificial Intelligence will make just about any job vulnerable. In the past menial work had been spared as whilst machines could do calculus in a flash they did have problems walking up a flight of stairs – or driving a car. Even now your phone can tell you how to get home, soon it will drive you as well. And this was not meant to happen; despite the evidence of other people’s driving, progressing down a crowded freeway isn't entirely mindless. It requires a deft combination of spacial awareness, perennial anticipation and multi-factorial visual analysis , and the ability to ignore your passengers — skills we thought quintessentially human.
Not surprisingly Frey and Osborne ‘further provide evidence that wages and educational attainment exhibit a strong negative relationship with an occupation’s probability of computerisation.’ In translation this means that high wage jobs that require a good university degree are less likely to be automated away in the foreseeable future. As are most aspects of the creative arts - though it is surprising how well computers are turning out what passes for music these days.
We would be quite anxious if we knew exactly which jobs are next in line for automation. The truth is scarier - we have no idea. Perhaps we should ask a computer?