The Hollywood vision of a jobless, authoritarian future has the super-rich minority living forever and the rest of the population in dangerous poverty, not able to do anything of interest or value.
Possibly in response to this, we regularly hear scare stories from some politicians (for example Nick Clegg in the Evening Standard earlier this year) about robots taking over and the imminent need for universal welfare programmes or taxes on robots.
But as automation accelerates and destroys jobs, how well is the UK jobs market responding?
Between 2011 and 2016, the number of jobs in the UK grew by 2.5m. This is a net figure. Looking at ONS data across 369 categories of job, we see that 3.6m were created and 1.1m destroyed. These are large changes in just five years.
We estimate that over 500,000 jobs were lost to automation, including cashiers (down 16 per cent, or 39,000 jobs lost from 2011 to 2016), traffic wardens (down 59 per cent, or 13,000) and typists (34 per cent, or 18,000). Jobs reduced by public policy (and then more than reabsorbed into the private sector on average into more sustainable roles) make up over 250,000.
But what of the future? In a paper entitled The Future of Employment: How Susceptible are Jobs to Computerisation, Carl Frey and Michael Osborne of the Oxford Martin School suggested that nearly half of US jobs (47 per cent) faced a high probability of being automated. They developed a framework for what makes jobs easier or harder to automate based on the tasks within the job.
Put simply: social, creative and complex jobs are harder to automate.
Should UK workers be concerned about this profound shift in employment patterns? Using the 2011-2016 ONS data, we set out to find out.
Overall, the 1.1m jobs destroyed over the period were at high risk (a 61 per cent chance) of automation according to the Frey and Osborne framework.
Within the 3.6m jobs created, meanwhile, about a quarter are clearly the result of bubble growth. There have been huge increases in the numbers in road driving occupations (up by 110,000 since 2011), for example, even though these jobs are at high risk of automation (89 per cent).
Importantly, the other three quarters of jobs created by the market are high quality, in that they have a low risk of automation (21 per cent).
Among the major drivers have been: demographic change (creating over 600,000 new roles for teachers, carers and medics); lifestyle shifts (plus 500,000) leading to new positions for personal trainers, chefs and Airbnb hosts; complexity (plus 600,000) creating roles in project management and management consultancy; and sales and marketing (up 200,000). The final category is automation itself, creating, we estimate, over 300,000 roles.
Such figures provide confidence that, so far, the jobs market is responding well to the challenge of automation, creating new roles that are more complex, social and creative (and so more sustainable with respect to automation) than the jobs lost to automation or public policy.
Each wave of industrial revolution brings new challenges and pushes the jobs market in new directions. There is a lot to do to make the most of these new jobs and trends. Investors and business owners should pursue automation to drive up productivity. Corporates should reassess the cost-benefit of retraining and develop medium-term skills plans. School students should be counselled on these factors as they choose degrees and careers. The state sector should catch up in terms of automation.
Because the new positions are social, creative and complex, retraining is a priority for those losing their jobs so the wealth arising from technology is spread widely.