Seven in 10 of the UK public are worried about the economic impacts of AI, six in 10 think it will eliminate more jobs than it creates, half think its impact will be worse than a normal recession – and one in five think it will create civil unrest.
People are particularly worried about the impacts on entry-level jobs and young people more generally: nearly six in 10 agree with Anthropic CEO Dario Amodei’s 2025 prediction that AI could eliminate half of all entry-level white-collar jobs within five years.
New research by King’s College London's Institute for Artificial Intelligence and Policy Institute finds that while fear of AI's impact on jobs and society is widespread and cuts across every group, there is somewhat less concern about the impact on people’s own jobs, and some groups are more excited about the positive impacts, particularly employers, men generally and male undergraduate students in particular.
The findings are being launched at the King's AI Summit: Workforce Futures on 19-20 May, where Baroness Martha Lane-Fox and Dame Chi Onwurah MP will speak alongside senior leaders from industry, academia, policy and the media to address urgent questions about the impact of AI on the future of work.
The study, the first wave of a major new tracker on attitudes to AI and The Future of Work, surveyed four groups: a 2,000 representative sample of the general public; a 1,000 sample of young people aged 16-29; 1,000 university students; and 500 employers.
Four in 10 (41%) of the public say they are afraid of AI, and only 24% think it is positive for humanity, compared with 39% who disagree. Almost half (48%) say they would rather avoid AI-based technologies altogether. Around a quarter to a third say they neither agree nor disagree with these statements, suggesting public opinion remains highly unsettled.
The gender gap is pronounced. Women are more likely to disagree than agree that AI is positive for humanity (44% vs 18%), positive for the UK (34% vs 23%), and that AI will improve their lives (41% vs 22%). Men are more evenly divided, and on whether AI is positive for the UK are slightly more likely to agree (33%) than disagree (29%).
University students are more positive than the wider public, with 47% saying AI is positive for the UK and 43% that it is positive for humanity, compared with just 28% and 24% of the general public respectively. But the gender gap persists even among students: a majority of male students believe AI is positive for both the UK (58%) and humanity (52%), while only 38% and 39% of female students agree.
Despite broadly negative views, more people agree (43%) than disagree (26%) that they will use AI in the future, suggesting that people are making pragmatic choices even where their attitudes remain sceptical.
Worry about the economic impact of AI-driven job losses is widespread and remarkably consistent: 69% of workers, 68% of university students and 63% of employers all express concern. A majority (57%) believe AI will eliminate far more jobs than it creates, leading to widespread unemployment.
One in five (22%) go further, saying AI will eliminate jobs fast enough to cause civil unrest – a share that rises to a third (34%) among university students.
If AI does lead to widespread job losses, half (50%) the public say the economic consequences would be worse than a normal recession, because AI would keep improving and taking more jobs before workers could recover. University students are the most pessimistic, with 56% taking this view.
Looking ahead five years, the public’s expectations of AI’s role in the job market darken considerably: while 32% currently say AI is mainly replacing workers, this rises to 47% when asked about the situation in five years’ time. Employers remain far more optimistic, with 49% expecting AI to be mainly assisting rather than replacing staff (23%) even in five years.
Across all groups, around six in 10 predict that AI’s economic benefits will flow mainly to wealthy investors and large companies. Just 7% of the public think the benefits will be shared fairly across society.
Despite the scale of general concern, people are noticeably less worried about AI’s impact on their own job. A majority of workers (55%) say they are not worried about their own job being replaced by AI. However, 38% are worried, including 16% who are very worried. Graduate workers (45%) are more likely than non-graduates (34%) to express concern about their own role.
University students show the same pattern of anxiety about their futures more broadly: while 8 in 10 would still go to university (77%), 30% say they would choose a different undergraduate degree given the growth of AI, and 12% say they would choose not to go to university at all. A majority (60%) expect AI will have made the job market much tougher for them by the time they graduate.
The public have a reasonably accurate sense of recent declines in entry-level job vacancies, with nearly half (48%) saying the number of advertised vacancies has decreased since 2022 – broadly in line with the reality. In June 2025, Azuna reported a 32% decline in entry-level roles since the launch of ChatGPT, and Indeed reported a 33% year-on-year drop in graduate jobs advertised. However, university students – the group most directly affected – are more than twice as likely as the overall public to incorrectly say vacancies have increased (19% vs 9%).
When asked what proportion of UK workers are in jobs AI could potentially perform or enhance, the public’s median guess is 35%. The International Monetary Fund’s estimate is 70% – twice as high – reflecting the large role of the service sector in the UK compared with other advanced economies. Workers, young people and university students guess only slightly higher at 40%.
The most common view on what has driven the decline in entry-level vacancies is that no single factor is to blame, with 31% of the public citing a combination of all causes. However, university students are nearly twice as likely as the wider public to single out AI as the main factor (19% vs 10%).
Three-quarters (77%) of university students use AI at least a few times a month, compared with 46% of workers. More than one in four students (27%) use AI daily or almost daily. Students use AI most commonly to help write or edit text (36%), gather and summarise information (32%), and prepare for exams (31%).
But almost nine in 10 students (85%) who use AI say they have encountered problems with AI-generated content, compared with two-thirds (65%) of workers who use AI. The most common problems for both groups are factual errors or inaccuracies (37% students, 31% workers) and made-up sources, quotes or statistics (31% and 19%). Fewer than half of students (43%) say they usually or always check and verify AI output before using it.
Almost all employers (92%) are already using AI in some area of their organisation. More than half (55%) use it for data analysis, and two in five for research (40%) and administrative tasks (39%). The overwhelming majority (86%) say AI has led to productivity improvements, with significant and modest improvements reported in equal measure (43% each). Increasing productivity is the most common reason given for AI adoption (63%), comfortably ahead of improving quality or accuracy (50%).
Employers who have seen productivity gains most commonly say AI has freed staff up for higher-value work (40%) or enabled further investment in AI (39%). But a fifth of all employers (22%) say they have already made roles redundant or reduced hiring specifically because of AI – a figure that rises with organisation size, from 9% of small employers to 29% of large ones.
Almost nine in 10 employers (88%) say they are confident they understand how their employees are using AI at work. The biggest reported barrier to wider adoption is concern about errors and bias (35%), followed by the cost of implementation (30%) and lack of relevant skills (25%).
Just one in five (20%) believe the education system is preparing young people well for a world shaped by AI. Almost half of the public (49%) say young people should prioritise vocational and technical careers over university as better protection from AI – a view shared by two-thirds of employers (67%).
Public appetite for government intervention is strong. Two-thirds (66%) say AI companies should be closely regulated even if it slows development. Majorities across all groups back both government-guaranteed retraining for workers displaced by AI (53%) and a tax on companies that replace workers with AI to fund retraining (53%), though employers are the most resistant to the latter, with 43% opposing it.
A clear majority (58%) of workers say they should benefit from AI productivity gains through higher pay. A similar proportion of employers (58%) agree in principle, though 36% say companies should be free to keep the financial gains.
Professor Elena Simperl, Director of The King’s Institute for Artificial Intelligence, King’s College London, said:
“These findings tell us something important: the British public isn't asking us to slow down on AI, they're asking us to do it better. People want these tools, they want more of them, and they've used them enough to know where they fall short. Employers see creative thinking as the top benefit AI can offer, ahead of productivity, but the public and the experts both doubt that today's tools deliver this. That puts a real onus on those of us building and deploying AI to make systems that genuinely support learning, creativity, and critical thinking, and honestly, too few people in the sector are working seriously on this. Moreover, women seem to be more cautious about AI than men. That should make us ask who we are designing for, and who is being left out of the conversation.”
Professor Bobby Duffy, Director of The Policy Institute at King’s College London, said:
“The public, workers, young people and university students are watching the rapid development of AI with more fear than excitement, with real concern for what it will do to jobs, particularly at entry levels, and, therefore, the prospects for our young people and the economy in general.
“This is perhaps no surprise when key figures, such as Anthropic’s CEO, Dario Amodei predicted that AI could eliminate half of all entry-level white-collar jobs within 5 years. Amodei has since painted a more optimistic picture of the labour market adapting and creating new opportunities. However, the public are much less convinced about similar claims: only a quarter agree with the World Economic Forum that AI will create twice as many jobs globally as it will eliminate by 2030.
“This, therefore, is a vision that will need to be explained, and demonstrated, to the public. It is still early days, and our baseline study shows that many don’t yet have firm views or much direct experience of AI’s impact – but that’s likely to change quickly, and we’ll need to outline clear plans on how we will adapt and support people in the transition.
“On that, the public’s instinct is move more carefully, with the majority favouring regulation and protection of jobs over fast adoption, alongside clear government and employer-backed plans for retraining. People mostly look to the government, schools and universities to help our young people adapt, but there is clearly much more to do here: for example, while a majority of university students say their university can prepare them well for an AI-shaped job market, only 36% say they currently are being well prepared.”
Dr Bouke Klein Teeselink, Lecturer in Philosophy, Politics, and Economics, King’s College London, said:
“This survey gives a really interesting window into how British students, workers, and employers feel about AI. Some of the main concerns held by the public, such as fewer job openings, a contraction in entry-level roles, and increased pressure on white-collar work, echo what I find in my own research on AI and the UK labour market. But none of these effects is fixed. With the right training, policies, and institutional support, there is a clear path forward to a more hopeful future, with rising productivity, broader opportunity, higher incomes, and faster scientific progress.”
Study details
Fieldwork was carried out by Opinium across 16-29 April 2026, consisting of four studies: a nationally representative sample of 2,000 UK adults aged 16+ (weighted on age, gender, education, region, ethnicity, work status, 2024 vote, EU referendum vote, and political attention); a sample of 1,002 GB adults aged 16-29 (weighted on age, gender, region, ethnicity, work status, and education); a sample of 1,000 GB university students (weighted by gender, age, and course level); and 506 UK employers (decision-makers in UK businesses with 11+ employees).
[i] https://www.gov.uk/government/publications/assessment-of-ai-capabilities-and-the-impact-on-the-uk-labour-market/assessment-of-ai-capabilities-and-the-impact-on-the-uk-labour-market#fn:6
此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。 原文来自 — 版权归原作者所有。