We spoke to network expert Terence Hogarth to find out what the history of technological change can tell us about the potential future impact of artificial intelligence (AI) on the labour market.
AI has been around, in one form or another, since the 1950s. However, more recently the emergence of generative pre-trained transformers (GPTs) such as ChatGPT has sharpened attention on the potential impact of AI on employment and skill demand.
AI in this emerging form looks likely to automate tasks previously considered beyond the reach of computers and robots. Indeed, some suggest that AI could completely transform the world of work, significantly impacting both the nature of work people undertake and overall employment levels.
The historical impact of automation
Before looking in more detail at how AI might affect the future labour market, it is instructive to look at the historical evidence on the impacts of technological change on labour and skills demand. Terence highlights that “Autor’s excellent survey provides the interested reader with a succinct summary”. (Autor, 2022)
In thinking about any kind of automation, we typically find three types of labour market impact:
- automation – where machines take over some or all of the tasks undertaken in a job;
- augmentation – where machines increase the productivity of certain tasks undertaken by workers; and
- task reinstatement – where new tasks or jobs emerge as a consequence of new technologies.
Historically, those tasks most at risk of being replaced have been in skilled tasks, which nevertheless follow a clear set of rules. These have tended to be in the middle of the occupational hierarchy (e.g. skilled trades jobs in the International Standard Occupational Classification). In contrast, low-skilled jobs have remained relatively immune to automation – as have high-skilled jobs that often require abstract thinking and interpersonal skills, such as in management roles.
At the same time, the ‘task reinstatement’ effect of previous technological developments has been significant. Autor et al (2021) estimate that around 60 per cent of employment in the USA in 2018 was found in jobs that had titles which did not exist in 1940.
The potential future impact of AI
Looking to the future then, it seems that AI may well change the relationships set out above, since AI is significantly different to historical forms of automation. With sufficient computing power and a suitable database, tasks which were previously out of the reach of computers or robots are now within their grasp. The ability of GPTs, for instance, to draft high quality text has already been widely demonstrated.
Analysis by Goldman Sachs estimates that up to two thirds of jobs in the USA could be affected by AI. Other findings suggest that 19% of workers in the USA could see more than half their tasks affected by GPTs (Eloundou et al, 2023). The ILO estimates that 0.4% of employment in low-income countries is potentially exposed to the effects of automation, rising to 5.5% in high-income ones (Gmyrek et al., 2023).
Does this mean that the world of work is about to be taken over by machines and robots? There is every reason to be believe that AI, like previous generations of technological change, will have a range of impacts. It will likely replace some jobs through automation, whilst augmenting others, and creating new tasks and jobs not yet seen in the labour market.
Whilst focusing on the range of jobs likely to be replaced can feel quite scary, the ILO study cited above actually suggests that the augmentation effect could be greater than the automation one. It estimates that AI has the potential to augment 10% of employment in low-income countries and 13% in high-income countries (Gmyrek et al., 2023).
Nevertheless, even if the impact of AI on employment and skills might be less than is sometimes imagined, there will be winners and losers. The evidence shows that this was certainly the case for earlier forms of technological change. There is therefore a critical need to identify those at risk of losing their jobs because of AI and to put in place, in good time, the requisite re-skilling and up-skilling to assist them in accessing the new kinds of job that AI will generate.
References
Autor, D. (2022) The Labor Market Impacts of Technological Change: From unbridled enthusiasm to qualified optimism to vast uncertainty. Washington: NBER Working Paper 30074
Autor, D., Salomons, A., and Seegmiller, B. (2021). “New Frontiers: The Origin and Content of New Work, 1940 – 2018.” MIT Working Paper, July.
Eloundou, T., Manning, S., and Mishkin, P. (2023) GPTs are GPTs: An Early Look at the Labor Market Impact Potential of Large Language Models. OpenAI Working Paper
Gmyrek, P., Berg, J., Bescond, D. (2023). Generative AI and jobs: A global analysis of potential effects on job quantity and quality. Geneva: International Labour Organisation Working Paper 96
Please log in or sign up to comment.