My last blog post focused on Amazon’s, now confirmed, entry into the Australian market and the potential impact that such a move might have on domestic consumers, retailers and workers. Many of the sources I came across while digging deeper concerned Amazon’s increasing use of automated systems. As such, I’ve decided to shift the focus of my project towards the broader implications of automation on the global workforce. This change means I don’t have to limit myself topically to either Amazon or, necessarily, Australia.
As early as 1967, figures like Marshall McLuhan were criticized (p.237) for believing that ‘total automation is upon us’. So to did William Gibson poignantly state time and again, that ‘the future is already here — it’s just not very evenly distributed’. So to that end, let us assess the current status of automation: What systems have been made obsolete by automation? What specific technologies are emerging today, and who is it displacing? Finally; what is on the horizon, and what professions, if any, will be safe from the process of automation creep? These will be the questions that my research report will engage with, and what I’ll be touching briefly upon in this post.
To talk about automation is to talk about what John Maynard Keynes coined (p.3) in 1930 as ‘technological unemployment’. He described this emerging phenomenon as the unfortunate ‘[availability] of labour outrunning the pace at which we can find new uses for labour’. Keynes added that this is only ‘temporary’, and standards of living will be multitudes better in one hundred years when there’s little work for anyone to do. But it was Keynes belief that ‘everybody will need to do some work if he is to be contented’ (p.6) as work provides meaning to one’s life, a topic for another time.
Since the process of industrial mechanisation saw a decline in production-line jobs that manufacturing industries provided, we haven’t yet seen any mass unemployment from the introduction of new technologies. Aside from the advent of electronic computing decreasing the need for human computers, and automatic exchanges largely making switchboard operators redundant, the workforce has survived. We’re only now seeing the beginnings of the technological unemployment Keynes imagined.
With the introduction of technologies such as the self-checkout machines at supermarkets, many commentators including Barack Obama himself, see automation as ‘relentless’ and ‘killing traditional retail’ jobs. With robots capable of sorting more than 200,000 packages a day in warehouses, and capable of working on cents worth of electricity instead of minimum wage, it’s hard not to be concerned. But importantly, it’s not just blue-collar industry workers who are at threat. White-collar professions relying on skills like decision making, paperwork, and writing are newly susceptible to automation via learning AI.
Platforms like Quill from Narrative Science can analyse large amounts of data and identify meaningful trends, then output a report reflecting these findings in ‘everyday language’, be it finance or sports results. While it’s been criticized for an inability to ‘discern the relative newsworthiness’ of stories, the unmatched speed and lack of bias that an AI system writes with is undeniable.
In addition to AI software, ‘general purpose’ robots are being developed with an ability to ‘learn’ new tasks. ‘Baxter’, from Rethink Robotics and Roomba creator Rodney Brooks, is being developed to fulfill ‘quality assurance or small assembly’ in factories, but still requires a human to initially ‘teach’ it these functions. This universal robot represents a leap in usefulness comparable to the first personal computers. Baxter is capable of fulfilling whatever task is ‘within his reach‘, but perhaps this is an agreeable compromise; there will still be work available for workers on an assembly line, but it will be less laborious and more about oversight and refinement of process.
Other systems are being designed to take over more skilled professions. IBM’s ‘Watson‘ for example is being touted as an AI doctor, networked to be constantly up to date with the newest research and possessing the ability to instantly access and share your medical records as required. Similarly, Enlitic has a program which can analyse medical imaging results and boasts a ‘false-negative rate of zero’.
The impact that automation makes on employment isn’t always clear until years later, however. The Economist reminds that although automated teller machines briefly reduced the number of human tellers in 1988, bank branches became cheaper to operate and so they grew by ‘43% over the same period’. So, will a technology like self-driving cars destroy the transport and hauling industry, or will new, unprecedented roles appear for the millions employed in those sectors?
While time will tell, I’ve plenty of sources to investigate for my final report in the meantime.