Automation driving towards After Capitalism

The fourth in our series of seven key drivers influencing the move to After Capitalism is technology acceleration.

Automation could drive towards After Capitalism from two distinctly different paths.

• The first path, the negative one, is that automation is applied indiscriminately and causes massive job loss and capitalism takes the blame.
• The second path, the positive one, is that automation is applied strategically and improves productivity and enables people to work less and eventually facilitates a post-work future.

Automation is defined here as the use of various technologies for operating equipment or for performing tasks, physical or digital, with minimal or reduced human intervention. Another way to frame the issue is whether “this time is different.” Waves of automation are not new and each time ways have been found to create more new jobs. The Shift Commission raised this question, in particular noting the contribution of AI: “We believe today’s progress in AI is consequential enough to prompt careful consideration of whether this time is different.” Slightly more than half their members (58%) agree that this time is different.Harari suggests that “the most important question in twenty-first-century economics … [that is] what to do with all the superfluous people …” made obsolete by automation.

There is plenty of potential for the first “massive job loss” path:

• Graeber supports the other extreme and suggests that not only will automation lead to net unemployment, but that it already has. He suggests that dummy jobs are effectively made up to mask the difference. His popular Bullshit Jobs suggests that if one eliminates bullshit jobs, and the real jobs that only exist to support them, from the picture, one could say that the catastrophe predicted in the 1930s has actually happened. “Upward of 50 to 60 percent of the population has, in fact, been thrown out of work.”
• Australia has a fully automated farm using robots and artificial intelligence.
• Xiang suggests that if AI remains under the control of market forces, it will inexorably result in a super-rich oligopoly of data billionaires who reap the wealth created by robots that displace human labor, leaving massive unemployment in their wake.
• Once established as successful, Uber announced plans to automate its fleet and put all its workforce out of work. Due to business problems, however, it has abandoned that approach for now.
• The high-profile story of a CEO firing 90% of his staff and replacing them with AI took place in India.

One might argue that the only reason automation has not taken off faster is that people are still cheaper than machines.

There is less current evidence for the “strategic” path leading toward post-work. This is not surprising since post-work has not yet emerged as a large-scale desirable concept. Several AC concept authors talked about the potential for technology acceleration as an enabler of post-work.

• Bastani’s suggests a key role for automation and technology as enablers of his “Fully Automated Luxury Communism” vision.
• Srnicek & Williams in their post-capitalism vision suggest automation can eliminate “huge swathes of boring and demeaning work.”
• Bregman in his vision of a post-work future, cites a RAND study suggesting a future in which only 2% of the population would be able to produce everything society needed.

A third potential path was identified in the research. Most projections identified by the research for this work suggested that automation would lead to net job loss. But not all. Reese is representative of a school of thought that AI will create more jobs than it destroys. He uses the example of the ATM, which for some time counter-intuitively led to more tellers, because it became cheaper to open a branch bank. However, the Bureau of Labor Statistics now projects a 15% decline in teller jobs in the US over the next 10 years. The view here is that this path is less plausible, and even if it were to be true, it would reinforce the Neoliberal Capitalism baselines and not drive towards After Capitalism.

My own study of the future of work aligned with the first path. We identified a baseline scenario in which automation is applied in purely economic terms with little regard for social costs. As the projections above suggest, the full impact of automation is yet to come. But when it does, it is likely to be quite disruptive, and little is being done to prepare for it. Thus, it seems most likely that automation will largely be applied with “full steam ahead.” There is still time for the strategic path to be taken. The work being done here supports that path.