AI Could Replace 3 Million Low-Skilled Jobs in the UK by 2035, Research Finds

A Transformative Shift in the Labour Market

A new study by the National Foundation for Educational Research (NFER) warns that the UK may lose as many as three million low-skilled jobs by 2035 as artificial intelligence and automation advance into sectors long considered stable. The research identifies roles in trades, machine operations and administrative support as especially vulnerable, with tasks in these categories being rapidly augmented — and in many cases replaced — by increasingly capable AI systems.

This shift does not necessarily mean an overall decline in employment. In fact, the UK is projected to add around 2.3 million new jobs during the same period. However, the types of jobs being created differ sharply from those at risk. Many emerging roles require higher levels of technical expertise, advanced digital fluency or analytical skills that low-skilled workers often have limited access to. Without deliberate intervention, this mismatch could leave millions displaced and unable to transition smoothly into new opportunities.

High-Skilled Roles Set to Grow, Not Shrink

The report challenges a common assumption that high-skilled professionals are equally exposed to automation. Instead, it argues that demand for highly skilled workers will increase in the short to medium term. As AI tools expand the capabilities of professionals, certain high-skilled roles may actually become more complex rather than less necessary. Instead of being automated away, these occupations may see expanded workloads, heightened responsibility and new forms of collaboration with AI.

Jude Hillary, one of the study’s authors, emphasises that AI’s impact will not be uniform. Some roles will be eliminated, others transformed and many new ones created — but the transition will not be straightforward. Those already in secure, high-skilled positions may benefit from AI’s productivity boost, while those in low-skilled or repetitive roles may face far greater instability.

Implications for Workers and Society

For workers in low-skilled positions, the threat is immediate and substantial. If their roles are automated, moving into the new growth sectors may require retraining or education far beyond their current qualifications. Individuals already struggling to secure stable employment could find themselves at a severe disadvantage.

For policy makers, the findings represent a call to rethink education and training systems. Traditional pathways into employment may no longer be sufficient. Instead, the UK will need more flexible, accessible and lifelong learning models that allow workers to continually adapt as the labour market evolves. Without such systems, the divide between those who benefit from AI and those who are harmed by it may deepen considerably.

Businesses too face strategic choices. While automation offers efficiency and cost savings, companies must consider their role in supporting employees whose jobs are changing or disappearing. Firms that manage this transition responsibly may help stabilise the workforce and reduce social disruption; those that focus solely on automation could exacerbate inequality.

Economic and Regional Impact

Although job numbers may rise overall, the uneven distribution of losses and gains could widen economic disparities. Regions already dependent on low-skilled labour may be hit hardest, particularly those outside major cities where access to retraining is limited. The report suggests that without targeted regional investment, AI-driven disruption may reinforce existing inequalities between urban and rural communities.

The shift in required skills also raises questions about mobility and access. Workers may be forced to move geographically or change sectors entirely — transitions that are often expensive, disruptive and difficult for those with family or financial constraints.

What Comes Next

Several factors will shape the scale of AI’s impact over the next decade. The speed of adoption across industries will influence how quickly jobs disappear or evolve. Government and corporate investment in upskilling will determine how well workers adapt to the new landscape. Regional policymaking will affect whether some parts of the country fall further behind. And as AI continues to change the nature of work, the definition of what counts as an “essential skill” will shift as well, making adaptability and continuous learning more important than ever.

The NFER report underscores a key message: AI is not simply a technological shift — it is an economic and social turning point. How the UK responds now will determine whether the coming transformation leads to opportunity, instability or a widening divide.