Automation vs Augmentation: The impact of AI on jobs

On the 28th of November 2023 the UK government released a report on the impact of AI on UK jobs and training. The report can be summarised as providing information on the exposure of different occupations, industries, and geographies to AI. In the report, the measure of exposure (AI occupational exposure) is calculated based upon the abilities exhibited in different job roles and how they correspond to different types of common AI applications (e.g. image recognition).

The key takeaway is that workers in professional occupations, particularly those in finance, law, and business management, located in London or the South East, are those with the highest exposure to AI.

Figure 10: Exposure to LLM by industry

GOV.UK Impact of AI on UK jobs and training report 

It is important to note that the report clarifies that the analysis measures exposure to AI, and does not distinguish whether a job will be augmented (i.e. aided) or replaced by AI. This is important because there are currently diverging views on how AI will impact jobs in the future. Nevertheless, in Annex 2, the report does provide insight into the effect of exposure to AI on likelihood of automation. Simply put, occupations at most risk of automation are those with the highest exposure to AI. Interestingly, there appears to be a sweet-spot for occupations with moderate exposure to AI that exhibit high augmentation. In plain English, jobs with some exposure to AI may be those set to benefit the most.

Figure 7: Occupations identified as “high augmentation” or “high automation” by ILO

GOV.UK Impact of AI on UK jobs and training report

My personal view is that AI will generally complement existing jobs and boost productivity, or create new jobs to replace those lost to the technology, rather than detrimentally affect the labour market.  However, I do appreciate that there’s always the possibility that Skynet will take over.

Separately, the Advanced Research Invention Agency recently released a document indicating that between 2000 and 2100, the proportion of the UK population aged over 65 is set to double. This raises the concern that developments in robotics/AI may not be capable of supplementing labour shortages in the future, particularly for physical occupations. Thus, we should welcome innovation in this space.

To this end, it is useful to note that advancements in the application of AI to the field of robotics are generally patentable before the European Patent Office. Positively, our latest AI report indicates that from 2000 onwards the number of applications in the field of robotics/AI has been steadily increasing. But will it be enough?