X Close

The jobs that survive computerisation will require a human touch. Are we still teaching that?

Credit: Lupuca/Google Images

Credit: Lupuca/Google Images

November 1, 2017   7 mins

It seems so obvious. Technology plays a larger part in our lives with every passing day. So education, at every level, should be increasingly focused on technology. Together with the cognate disciplines of science, engineering and mathematics, the four are handily grouped together by their initial letters as “STEM”.

Every time someone raises the question of tomorrow’s jobs, the answer seems obvious. The T-word. They will be tech-focused. So let’s get people prepared. In the words of the UK government’s education department, “If we want the UK to remain a world leader in research and technology we will need a future generation that is passionate about, and skilled in, science, technology, engineering and maths (STEM).”1

Like other fashionable assertions that seem obviously true, this one depends on a series of assumptions that are actually open to question. It’s surprising how little critical assessment there is of the STEM mantra, and the fallacy behind it.

When one of the witnesses at the UK’s House of Lords recent committee inquiring into Artificial Intelligence dared to raise a question about STEM, it was noticed. In Computer Weekly’s report on the hearing, Andrew Orlowski, who edits the tech site The Register, stood out:

“In a rare demurral from the conventional stress on science, technology, engineering and mathematics (Stem) education among technology commentators, he said: ‘Kids are taught algorithms every week now, but hardly taught music or history at all.’ ”2

Tech future, human jobs

There’s no doubt that many of tomorrow’s jobs will involve technology. But what if jobs with a high level of technology in them prove to be just the kind of jobs most likely to disappear? Won’t they be the ones most susceptible to computerisation?

The jobs that will survive are not the ones with a high “tech quotient” but those with a high “human quotient”

Which jobs are most likely to survive is an urgent and complex question. One of the unfortunate effects of the STEM fetish has been to douse that debate. Whether we are looking to re-train ourselves, advise children and grandchildren, or re-shape our vast systems of public education, we need a smarter and wider-ranging discussion of future jobs than “more STEM because more tech”.

Because there’s a good argument that the key feature of the jobs that will survive (and new ones that will emerge) is precisely the opposite. Not the jobs with a high “tech quotient” but those with a high “human quotient”.

Looking ahead, we don’t know if the total number of jobs will actually go down as robots take over many traditionally human functions, or if people will merely shift into new kinds of jobs.  Either way, there will be serious turbulence in labour markets. It plainly makes no sense to prepare people to do jobs that have no future.

The question is how we should be preparing average students for the jobs they will be doing in 10 and 20 years’ time

Of course, there are currently shortages of skilled candidates for various tech-focused jobs (from computer programming to welding). We need to prepare people to do them. And in the long term, assuming the robots haven’t completely taken over, there will always be a need for extremely smart people at the top of the tech tree. The question I’m asking here is about the kind of work more average workers will do, and how we should be preparing average students for the jobs they will be doing in 10 and 20 years’ time.

This baseline report on the likely future of jobs3 from Oxford University’s Future of Humanity Institute is distinguished by its painstaking analysis. Its authors review every single job category there is. They start with the US Department of Labor job classification list, which runs to over 900 separate kinds of jobs. Then they pare the list down to 702, as some jobs are very similar to each other. Then they assess all 702, one at a time, for how likely it is they will be done by machines. The report is a treasure-chest of practical examples.

At the end of the report there is the actual list of all 702 jobs. If you scan it for STEM-type jobs, the result is stunning. They start (No 1) with jobs least likely to go to machines. They end with those most likely.


At number 699 (out of 702) are mathematical technicians, and not far away five other kinds of technicians – library (692), agricultural and food science (665), prepress (660), ophthalmic (656) – all ripe for computerisation. Also watch repairers (at 697), tax preparers (695), milling and planing operators (678), locomotive engineers (638), construction operators (617) and nuclear power operators (614). Even welders come in at 598. The point is that low and mid-level tech jobs will likely be the very first to go. The same is true of low and mid-level jobs involving numbers. Insurance underwriters come in at 698, tax preparers at 695, accounts clerks, brokerage clerks, and order clerks  at 693, 698 and 697 respectively. These jobs involving tech and numbers are at the very top of the list of ones to be replaced by machines.

Then skip up to the other end, and look at the jobs least likely to be computerised. The first dozen jobs include healthcare social workers (8), substance abuse counsellors (4), recreational therapists (1), occupational therapists (6). At number 13 we have choreographers. While physicians and surgeons and dentists (15, 19) also feature, a little further down, as well as some other jobs that are STEM-related, these are high-level occupations – and ones that involve interface with people.

It’s staring us in the face: average-worker tech jobs will be first to go. It’s non-tech jobs such as social work and therapy that are the very safest.

The new leisured class

Second, there seems little doubt that in the decades to come there is likely to be less work to go around. Whether there’s a wholesale “jobocalypse”4 or merely a more gradual replacement of human labour by machine labour, there are good reasons to believe that in the future some combination of automation and increasing wealth holds out more leisure for most people. The old debate as to whether education is chiefly for “life” or “work” is re-emerging in fresh fashion. It was recently put in stark form to me by a student, who asked what she should be learning at university if when she graduated, at the age of 22, she would also retire.

There is an increasingly pressing need for education to prepare us not so much for work as for the new “leisure” that advances in technology will likely bring

Economist John Maynard Keynes, in his tantalising essay Economic Possibilities for our Grandchildren,5 looked forward to a time when machines would free us from much of the work we do. He spoke whimsically of the emergence of a “new leisured class” – in other words, ordinary folk with no jobs. And he noted the real problem that such “leisure” could create: boredom. “For those who sweat for their daily bread leisure is a longed-for sweet – until they get it… It is a fearful prospect for the ordinary person… to occupy himself.”

In other words, there is an increasingly pressing need for education to prepare us not so much for work as for the new “leisure” that advances in technology will likely bring. Yet the opposite is happening. As Andrew Orlowski noted in his evidence to the House of Lords, schools are teaching less music and history in order to up their technology curriculum.


STEM and longevity

Third, the extraordinary increase in the human life-span that has differentiated our modern experience from that of our grandparents continues to press ahead. And, of course, it is mainly the result of advances in medical science and nutrition. So we can chalk it up to STEM. It’s taken place gradually, so there is surprisingly little public awareness of its revolutionary impact. Life expectancy has more than doubled since 1900, and the process is keeping moving6. To take one recent projection as an example, “up to two thirds” of children born in the UK now can expect to live until they are over 1007.

So, whatever the impact of automation on the future of jobs, remarkable improvements in health and longevity mean that today’s children will spend many years of their lives outside the workforce in greatly extended “retirement”. These are years for which they need to be prepared.

How to occupy the leisure?

Keynes puts it well: “Thus for the first time since his creation man will be faced with his real, his permanent problem how to use his freedom from pressing economic cares, how to occupy the leisure, which science and compound interest will have won for him, to live wisely and agreeably and well.”

Yet this is the very moment that the STEM screws are being tightened. And the disciplines that do indeed prepare us best for spending time outside the work environment – the humanities, the arts – are being placed under increasing pressure in every educational institution. As Orlowski said, things like music (long under siege from those pressing for better performance in “academic” subjects) are being pushed out.

Even if our concern is limited to preparing the rising generation for the kind of jobs most likely to survive the vast impact of automation on the workplace, it’s simply a fallacy to believe that for the general mass of the workforce the answer lies in more STEM. There’s no question that every worker needs to understand the basics of technology, not least as many new jobs are likely to be focused on managing the interface between people and machines (for example, the elderly in their increasingly smart homes). But these will not be “tech” jobs.

For the great majority of people the need is for a vast revival of the humanities and the arts at the centre of our educational enterprise

What’s more, it’s looking increasingly likely that less of our time will be spent working in the future – whether we are all down to 15 hours a week (Keynes’ suggestion), or retirement gets earlier as governments use it to manage the job pool (my own proposal8), or whether there is simply a general increase in unemployment as automation-driven “rust belts” break out all across the economy, so there will be more “leisure”. And as Keynes noted, leisure is something we are not good at managing – especially after the first hour or two or week or two.

And then there’s longevity. Whatever the outcome of the automation process, we are living longer and the generation born today will likely live a great deal longer. The need grows to prepare us for the kind of life in which the rigours of work no longer provide us with meaning.

So in place of STEM, for the great majority of people the need is for a vast revival of the humanities and the arts at the centre of our educational enterprise. And let’s name the STEM fetish for what it is – raised upon a fallacy, and for all its good intentions set to mislead young people as to the kind of careers to which they should aspire, and help deprive them of due preparation for the lives that lie ahead, outside and after the workplace.


As the reference from Computer Weekly to Orlowski’s evidence shows, almost everything written about the future of jobs buys what we’re calling the technological fallacy. Some key websites:

US government:  https://www.ed.gov/stem

UK: https://www.stem.org.uk/about-us



  1.   https://www.gov.uk/government/publications/2010-to-2015-government-policy-public-understanding-of-science-and-engineering/2010-to-2015-government-policy-public-understanding-of-science-and-engineering
  2.  Brian McKenna, in Computer Weekly for October 11, 2017: “House of Lords committee hosts clash between AI enthusiasts and sceptics.” http://www.computerweekly.com/news/450428020/House-of-Lords-committee-hosts-clash-between-AI-enthusiasts-and-sceptics. Examples of the STEM push: In the United States, the STEM Coalition: http://www.stemedcoalition.org/about-us/our-purpose/ . In the UK: https://www.gov.uk/government/publications/2010-to-2015-government-policy-public-understanding-of-science-and-engineering/2010-to-2015-government-policy-public-understanding-of-science-and-engineering
  3.  Carl Benedikt Frey and Michael A. Osborne, “The Future of Employment. How Susceptible are Jobs to Computerisation?” September 17, 2013. Oxford Martin School. http://www.oxfordmartin.ox.ac.uk/downloads/academic/The_Future_of_Employment.pdf
  4.  I explore this in my book Will Robots Take Your Job? (Polity Press, 2017), summarised in this UnHerd video: https://youtu.be/QABMVTWVDRM
  5.  J.M. Keynes, “Economic Possibilities for our Grandchildren,” originally published in 1930, available here:  http://www.executiveshift.org.uk/images/site_graphics/downloads/John_Maynard_Keynes.pdf
  6.  E.g. Max Roser, “Life Expectancy,” at Our World in Data. https://ourworldindata.org/life-expectancy/
  7.  John Bingham, December 12, 2013, in The Daily Telegraph: “Two thirds of today’s babies could live to 100.” http://www.telegraph.co.uk/news/health/news/10511865/Two-thirds-of-todays-babies-could-live-to-100.html
  8.  https://staging.unherd.com/2017/09/flexible-retirement-ages/

Nigel Cameron writes about technology, society, and the future. In 2007 he founded the Washington think tank The Center for Policy on Emerging Technologies. His most recent book is Will Robots Take Your Job?


Join the discussion

Join like minded readers that support our journalism by becoming a paid subscriber

To join the discussion in the comments, become a paid subscriber.

Join like minded readers that support our journalism, read unlimited articles and enjoy other subscriber-only benefits.

Notify of

Inline Feedbacks
View all comments