According to a new AI report by Dublin-based professional services firm Accenture, 49% of UK executives are expecting artificial intelligence to result in net job losses over the next decade. At a time when companies as significant as Meta, Snap, Oracle, Block, Atlassian, and Salesforce are all shedding employees amid significant AI investment, such a prediction may seem distressing to anyone hoping to hold onto or secure employment. The thing is, most available data on AI’s actual impact on workforces isn’t as gloomy, and may point to a future where LLMs and other technologies enable a net increase in jobs.
The percentage of leaders specifically expecting AI to reduce entry-level roles has climbed from 22% to 37%
Looking at certain details of Accenture’s report, we’d be forgiven for believing that this future will not be bright for younger generations, who UK execs fear will bear the brunt of automation. Their outlook has actually worsened since last year, with only 15% expecting AI to boost entry-level positions, compared to 40% in 2024. At the same time, the percentage of leaders specifically expecting AI to reduce entry-level roles has climbed from 22% to 37%, underlining how the mood is only becoming more negative as far as more traditional career pathways are concerned.
Yet while there is an ongoing expectation that AI will ultimately reduce the number of jobs available to flesh-and-bone people, much of Accenture’s report isn’t as pessimistic, particularly when it comes to recording what’s happening on the ground right now. And its data resonates with other recent studies and surveys, which suggest that AI may be having a positive impact on job numbers.
‘Intensive’ AI use is boosting jobs
Delving further into Accenture’s report, it becomes apparent that many UK companies are still in an exploratory or experimental phase with AI, which has yet to bear on their bottom lines. For example, 46% of execs report that AI has had “little to no positive impact on profit and loss,” while 44% admit that “at least some” of their AI budgets are “wasted.” Meanwhile, 31% revealed that halting all use of AI would have “no material effect” on their activities, a sentiment which seems a far cry from prophecies that automation will eliminate entire job categories.
46% of execs report that AI has had “little to no positive impact on profit and loss, while 44% admit that “at least some” of their AI budgets are “wasted.”
In terms of job numbers, a section of Accenture’s studies examines listings, in particular looking at increases or decreases in positions that involve certain skills. In total, occupations whose postings mention, for example, such skills as “basic technical knowledge,” “language competency,” “business management,” “writing and editing,” “office productivity technology,” and “web design and development” declined by 473 between 2023 and 2025.
Accenture’s report also finds that occupations involving skills such as “people management,” “AI and machine learning,” “regulation and legal compliance,” “quality assurance and control,” “business strategy,” “teaching,” and “process improvement and optimisation” increased by 1,149 in total over the same period.
In other words, the implication of this data is that AI is already having a net positive impact on total job numbers, despite the fact that LLMs such as ChatGPT have been around for only three and a half years. This may seem like a counterintuitive conclusion, especially when much of the reporting on Accenture’s report is latching onto the ‘net job losses’ prediction, but it is supported by other data.
Moderate use of AI isn’t altering a company’s propensity to hire or fire employees, while ‘intensive’ AI use is increasing the likelihood of hiring
Most notably, the ECB published a survey in March of 5,000 European companies, finding that firms which “make significant use of AI” are 4% more likely to increase their staff numbers, while firms that invest in AI are 2% more likely to take on more workers. Speaking to Raconteur, report author David Sondermann confirmed it was “broadly correct” that the study also found “no significant difference” between enterprises that make moderate use of AI and those that don’t. That is, moderate use of AI isn’t altering a company’s propensity to hire or fire employees, while ‘intensive’ AI use is increasing the likelihood of hiring.
Similar data is available in other regions, with the UK’s Association of Professional Staffing Companies (APSCo) — a trade body representing the recruitment industry — identifying growing confidence among recruiters, particularly in the IT and engineering sectors.
“APSCo UK data presents a slightly more positive picture than the ONS headline figures,” said Global Public Policy Director Tania Bowers, speaking to Raconteur. “This is likely because our research focuses on skilled, professional roles, which have so far proved more resilient than entry‑level positions.”
Bowers explains that isolating the impact of AI on jobs is still a difficult ask, not least because the employment market is facing a cocktail of pressures, including post‑pandemic adjustment, the war in Ukraine and the war in the Middle East. However, she does confirm that APSCo’s members are reporting accelerated adoption of AI within their own businesses, which has caused drops in staff numbers in such areas as administration and early‑career recruitment.
“APSCo UK data presents a slightly more positive picture than the ONS headline figures”
Tania Bowers, APSCo Global Public Policy Director
She added, “In some cases, this has reduced demand for entry‑level graduate and support roles as technology replaces or reshapes those functions.”
Conversely, Bowers also notes that AI has resulted in “the emergence of new job titles and teams, particularly around AI governance,” as businesses look to manage risk and accountability more effectively. This is exactly what Accenture’s report suggests, and it supports claims that AI may create more jobs than it destroys.
Long-term future remains uncertain
Again, this conflicts with the long-term outlook of many executives, with the World Economic Forum’s AI report from January revealing that 54% of business leaders globally expect AI to displace workers, while only 23.5% believe it will create “a large number” of new jobs. Such percentages are certainly revealing of what execs may hope to achieve via AI adoption, but the current data would suggest that a jobs apocalypse is some distance away.
In fact, a March report from Claude developer Anthropic found “no impact on unemployment rates” for US workers in occupations (theoretically) most exposed to AI. Similarly, a January study from Oxford Economics concluded that evidence of AI-induced losses is “patchy,” and that productivity growth has been “weak and volatile” in the US since LLMs emerged, contradicting what you’d expect if super-smart AIs were replacing workers en masse.
Of course, whether job numbers hold up in the more distant future is another question, and it’s on this point that many researchers hold back from offering bold predictions. In the conclusion of its blog post on the subject, for instance, the ECB’s authors acknowledge that AI hasn’t (yet) transformed production and operational processes in significant ways, but predicts that “this is set to change,” meaning that the “longer-term impact of AI on employment remains less clear.”
“If we get that right, this isn’t a story of job loss. It’s a story of job transformation.”
Ali Bebo, Pearson’s Chief Human Resource Officer
Speaking to Raconteur, Pearson’s Chief Human Resource Officer, Ali Bebo, suggests that the growth of AI could lead to higher employment levels, but only if organisations invest in people “as seriously” as they’re investing in LLMs.
“The real barrier isn’t the technology,” she said. “It’s how we redesign work, build skills, and help people adapt.”
Bebo adds that the companies that thrive in the future won’t be those that simply deploy AI faster than others. It will be those that also “build capability,” meaning training and hiring people to make the most of new tools.
“If we get that right, this isn’t a story of job loss, she adds. “It’s a story of job transformation.”
The WEF report also refrains from tendering any definitive forecasts, instead outlining four scenarios that range from “stalled progress” to “supercharged progress,” whereby exponential AI innovation results in displacement but also new occupations that “emerge and scale fast.” In the face of this uncertainty, the report advises enterprises to take various steps that can help them prepare for any scenario, including scaling what works, investing in human-AI collaboration, strengthening in organisational culture and transparency, and ensuring that talent evolves in parallel with technology.
According to a new AI report by Dublin-based professional services firm Accenture, 49% of UK executives are expecting artificial intelligence to result in net job losses over the next decade. At a time when companies as significant as Meta, Snap, Oracle, Block, Atlassian, and Salesforce are all shedding employees amid significant AI investment, such a prediction may seem distressing to anyone hoping to hold onto or secure employment. The thing is, most available data on AI’s actual impact on workforces isn’t as gloomy, and may point to a future where LLMs and other technologies enable a net increase in jobs.
The percentage of leaders specifically expecting AI to reduce entry-level roles has climbed from 22% to 37%
Looking at certain details of Accenture’s report, we’d be forgiven for believing that this future will not be bright for younger generations, who UK execs fear will bear the brunt of automation. Their outlook has actually worsened since last year, with only 15% expecting AI to boost entry-level positions, compared to 40% in 2024. At the same time, the percentage of leaders specifically expecting AI to reduce entry-level roles has climbed from 22% to 37%, underlining how the mood is only becoming more negative as far as more traditional career pathways are concerned.
