Ever since Oxford’s Carl Benedikt Frey and Michael Osborne published their paper on the potential for jobs to be automated in 2013, a groundswell of concern has emerged about the impact of the various technologies of the 4th industrial revolution might have on the jobs market.
Given that for many, the prevailing narrative has been that the relentless march of digital technology will consume all jobs in its wake, a COVID-19 pandemic that has in many ways expedited digital transformation might have also expedited the automation of the workforce.
It’s a narrative that has an element of truth to it. A recent report from Dell highlighted the way COVID has accelerated the digital transformation efforts many organizations had been undertaking before the pandemic, albeit with areas such as cybersecurity getting as much investment as the rebuilding of business models. It’s also noticeable that technologies such as AI and robotics were still not a major focus for organizations.
That firms don’t appear to be investing in AI to any great extent is shown up for the folly that it is by research from Berkeley Haas, which highlights how beneficial such investments are in terms of productivity, efficiency, and market share.
What is perhaps important to note in discussions around the impact of technology on jobs is that the companies investing so heavily in AI actually grew their workforce by 15%. What’s more, this increase in employment was not just found at the firm level, but across entire industries. So AI-savvy firms were not benefitting at the expense of rivals. The researchers suggest that this is because firms were using AI to become more productive, which helped them to grow.
This expansion was often into new regions or product markets, as firms capitalized on their newfound productivity and efficiency. Obviously, this does tend to feed into concerns about excessive market concentration among a few huge firms, but the researchers believe this shouldn’t concern us as it is simply a reflection of market share going to the most productive firms.
It’s a finding echoed by additional research from the Stanford Institute for Human-Centered Artificial Intelligence (HAI), which found that there appears to be a clear relationship between AI-related jobs and economic growth, which in turn results in greater well-being across society.
This general optimism is not uniformly felt, however. For instance, the Global Attitudes Report from Oxford University’s Internet Institute reveals a public that is broadly speaking divided down the middle on the impact of AI in public life, with those in the East seemingly far more comfortable than those of us in the West.
The research was based upon the data generated by the 2019 World Risk Poll produced by the Lloyds Register Foundation. Their survey examines the public perception of global risks, with 142 countries participating.
“Understanding public confidence in AI and machine learning is vital to the successful implementation of such systems in government,” the researchers explain.
This perception of vulnerability was reinforced in the latest report from The Royal Society for Arts, Manufactures and Commerce’s (RSA) Future Work team on how COVID is affecting the automation landscape.
The report attempts to develop a risk register of jobs that are vulnerable to any fallout from COVID-19 and indeed the digital transformation that society is undergoing as a result of the pandemic. The authors suggest that those industries that are most reliant on furlough support from the government are also most likely to be at risk from automation.
They identify young people as the most vulnerable, which is perhaps not that surprising, as research from the University of Western Australia found that young people were most vulnerable to job losses during COVID in general.
The paper highlights how the youth labor market is typified by high levels of consumer-facing roles in sectors such as hospitality, fitness, and retail. What’s more, such jobs are often part-time or casual, and youth unemployment was already extremely vulnerable due to this.
While it’s far from clear from the evidence to date that automation is playing anywhere near as big a role as COVID itself in the labor market challenges faced by young people, it is nonetheless crucial that support is provided to young people suffering as a result of the pandemic.
It’s well known that in previous crises, such as in 2008 and the early 1990s, unemployment tended to crowd the labor market after the recession, which made it extremely challenging for young people. As such, the authors believe governments should offer support to help young people back into secure work.
“Personalised support of our young people that can address multiple barriers to employment such as qualifications, transport, disadvantage, job readiness and communication skills is important,” they say. “It’s also important for policy leaders to address the increasing casualisation of the workforce and ensure they support those who are most vulnerable.”
The RSA do provide some solid advice in terms of the nature of this support, especially in terms of helping people to gain new skills and transition into new careers. These are almost certainly going to be hugely important, even if their suggestion that AI is to blame is wide of the mark.