Insurance can be admin heavy, due to compliance factors, the complexity of claims, or manual updates to policies mid-term. AI can help a great deal here, but will it ultimately replace lots of jobs? Here’s the word.
The UK is among the countries most exposed to the next wave of artificial intelligence (AI) driven transformation with almost 20% of total workplace tasks – such as drafting documents, analysing data and processing information – at risk of automation, according to new research co-produced by Coface and the Observatoire des Emplois Menacés et Émergents (OEM).
The report analyses 923 occupations across 12 countries and finds that AI is likely to have a fundamentally different impact from previous technological waves – disproportionately affecting high-skilled, high-paid, and cognitive roles rather than routine or manual work.
The UK ranked as the most exposed advanced economy alongside the Netherlands
The UK ranks at the top end of global exposure, alongside countries such as the Netherlands, reflecting its concentration of jobs in finance, professional services, and technology.
The UK has historically benefited from a strong base in white-collar roles, with many international corporations choosing to make the UK their headquarters for access to well-educated and talented workers able to support business operations. This is being turned on its head, and Coface has identified that the UK suffers from a “headquarters trap” whereby employment is concentrated in highly exposed, cognitively intensive occupations – most notably in finance, IT, legal and media jobs. The tasks most at risk tend to fall into data processing, analysis, and content generation, which AI systems are capable of automating.
While much of the disruption is still ahead, early signs of impact are already emerging in labour markets in the UK. Evidence suggests that AI is beginning to reshape hiring patterns, it seems to affect entry level positions more, where employers are slowing recruitment in roles most exposed to automation. Unemployment has risen more among 16–24-year-olds than overall unemployment since the widespread adoption of generative AI.
By contrast, more senior roles have so far proved relatively resilient, reflecting their greater emphasis on oversight, judgment and decision-making. This divergence risks narrowing traditional career pathways, as fewer entry-level opportunities make it harder for younger workers to gain a foothold and progress into higher-skilled roles over time.
Significant disruption ahead
The study suggests the next phase of disruption will be driven by the rise of “agentic AI” – systems capable of executing multi-step workflows rather than assisting with isolated tasks.
This shift could automate entire job functions rather than individual tasks, reduce demand for cognitive, white-collar roles and further increase reliance on human oversight rather than execution.
Contrary to previous automation waves, the most exposed occupations include:
- Engineering and IT roles (29%)
- Legal, financial, and creative professions (27%)
- Management and administrative functions (24%)
By contrast, physically intensive and in-person roles, such as construction, transport, and care, remain relatively protected.
Implications for the UK economy
The potential economic implications for the UK are significant. Coface expects that AI-driven automation would result in a large impact on employment activity, as well as potential for increased productivity which could result in a positive for GDP growth over the coming years.
As the UK’s most exposed roles are also among the highest contributors to tax revenues and economic output, the consequences could extend well beyond employment. Coface warns of pressure on wage growth in high-income professions, a shift in income from labour to capital and a potential erosion of the UK tax base, which relies heavily on income taxation. The impact could lead to a shortfall in tax revenue. They note that the shift could also lead to a partial leakage of capital income toward foreign countries where AI profits are concentrated.
The report suggests that AI could create a “double fiscal challenge” for the UK. As highly paid, white-collar roles come under pressure, government revenues from income tax and National Insurance could decline. At the same time, public spending may rise if displaced workers require unemployment support or retraining.
More structurally, a shift away from high-paying professional roles towards lower-paid service sector jobs could further erode the tax base, as these roles typically generate less tax revenue. This could force a rethink of how public services are funded, particularly if a growing share of AI-driven profits accrues to large international technology firms rather than being taxed domestically.
Global comparison
While the UK sits at the higher end of exposure, the study finds similar patterns across advanced economies:
- Germany: ~17%
- United States: ~17%
- France: ~16%
- Turkey: ~12%
Higher exposure correlates strongly with GDP per capita, reflecting the greater concentration of knowledge-based, white-collar roles in advanced economies. By contrast, lower-income economies tend to have a higher proportion of manual, in-person or informal jobs, which remain more resistant to automation. As a result, while AI is a global phenomenon, its impact is likely to be felt most acutely in advanced economies. The study estimates that around one in eight occupations could undergo deep structural change as AI adoption accelerates, with roles reshaped rather than eliminated as significant portions of day-to-day tasks are automated.
Jonathan Steenberg, Lead UK Economist at Coface, commented: “The real economic risk for the UK isn’t a sudden loss of jobs, but a gradual erosion and concentration of its highest-value roles. If that plays out, it could weigh on productivity, employment and ultimately the tax base that underpins public finances. This, in turn, would raise longer-term questions about how the UK sustains its economic model and it works in an AI-driven economy.”
Be the first to comment