The UK government is urging companies to share data on how artificial intelligence is affecting jobs, skills, productivity, and work quality, as ministers seek a clearer evidence base around one of the most consequential labour-market shifts facing the economy.
The request is linked to the AI Economics Institute, a unit under the Department for Science, Innovation and Technology. Companies are being asked to provide aggregated data on what is changing inside organisations as AI moves from pilots and productivity tools into workflows, staffing models, and operating decisions.
The data-sharing request is voluntary. It is aimed at improving government understanding of how AI is changing business performance, job design, and workforce demand before policy is forced to respond to displacement, skills shortages, or uneven adoption after the fact.
A January government assessment said the available evidence did not yet provide clear answers to many of the questions that matter most for policy. AI capabilities were progressing rapidly, but evidence on labour-market effects remained incomplete and difficult to interpret.
The same assessment pointed to potentially significant productivity gains. It cited OECD estimates suggesting that UK labour productivity growth from AI could reach 0.4 to 1.2 percentage points annually over the next decade, reflecting the UK’s high exposure to knowledge-intensive services.
Productivity, however, is only part of the policy challenge. AI may automate tasks, redesign roles, increase demand for new skills, alter entry-level career pathways, or raise performance expectations in existing jobs. In many cases, the effects may not appear as simple job losses or hiring increases, but as changed workflows, changed skill requirements, and changed routes into work.
Company-level data could help close that gap. Headcount figures alone do not show whether work has been reorganised, whether junior tasks have disappeared, whether productivity has improved, whether training has kept pace, or whether gains are concentrated among certain functions and roles.
The data request sits alongside a widening skills debate. Skills England has warned of a workforce gap across priority growth sectors, highlighting rapid AI adoption, weakening employer investment in training, and the risk that young people could face fewer traditional entry routes if lower-level tasks are absorbed by technology.
Companies are investing in AI to improve productivity and reduce cost, yet the labour-market consequences are still being measured unevenly. If AI improves output without reducing employment, policy will need to support adoption and upskilling. If it reduces entry-level demand, education, apprenticeships, and early careers models may need to change. If it shifts value towards experienced workers with domain knowledge and AI fluency, progression routes will need closer attention.
Recent business evidence is mixed. The British Chambers of Commerce reported earlier this year that AI adoption among SMEs had risen sharply, from 25% in 2024 and 35% in 2025 to 54% in 2026. Its report also found that most companies were using generic AI tools rather than deeply integrated systems, suggesting current use was still largely supporting existing work rather than transforming operations. More than nine in ten reported no impact on headcount over the previous 12 months.
That pattern may change as adoption moves from individual productivity tools to integrated systems. The largest effects are likely to emerge when AI is embedded into customer service, finance, software engineering, legal operations, HR, marketing, and business analysis, rather than used as an optional assistant by individual workers.
There is also a governance problem. Companies may be willing to share aggregated data, but many will be cautious about exposing commercially sensitive information, workforce plans, or early evidence of job displacement. Others may not yet have reliable internal data, particularly where AI use is fragmented across departments or happening through unapproved tools.
The request therefore tests both policy design and corporate measurement. Companies that cannot identify where AI is being used, what work it is changing, or how it is affecting skills and productivity will struggle to provide meaningful evidence. The same lack of visibility also weakens internal decision-making.
Government will need a useful national evidence base without creating a reporting burden that discourages participation. Employers, meanwhile, will need to treat AI deployment as a workforce strategy as well as a technology roll-out, with clearer data on which roles are being augmented, which tasks are being automated, and where new capability is needed.
The labour-market debate around AI has often swung between optimism and alarm. A request for company data points to a more practical stage: policy now depends on evidence from inside organisations, while employers need to show that AI deployment is being measured as carefully as it is being promoted.
You must be logged in to post a comment.