Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Code, Capital, and Clusters: Understanding Firm Performance in the UK AI Economy

Waqar Muhammad Ashraf, Diane Coyle, Ramit Debnath

TL;DR

The study investigates how AI specialization and local socioeconomic contexts shape firm performance in the UK AI economy (2000–2024) using a merged dataset from WAIFinder, glass.ai, and Companies House, augmented with postcode-level census indicators. It employs CatBoost with SHAP to identify drivers of operating revenue and revenue per employee, revealing that firm size drives total revenue while specialization dominates productivity, with local factors contributing meaningful variation. Time-series forecasting (ARIMA, Theta, ETS, MFLES) with conformal prediction projects the UK AI sector toward maturation and consolidation by 2030, estimating approximately $4{,}651$ total entities (CI: $4{,}323$–$4{,}979$) and a rising dissolution rate near $2.21\ ext{%}$; London remains the dominant hub, intensifying regional policy considerations. The findings support place-sensitive policy to cultivate regional AI capabilities, differentiate scaling from specialization, and manage ecosystem consolidation to sustain growth and resilience in the UK AI economy.

Abstract

The UK has established a distinctive position in the global AI landscape, driven by rapid firm formation and strategic investment. However, the interplay between AI specialisation, local socioeconomic conditions, and firm performance remains underexplored. This study analyses a comprehensive dataset of UK AI entities (2000 - 2024) from Companies House, ONS, and glass.ai. We find a strong geographical concentration in London (41.3 percent of entities) and technology-centric sectors, with general financial services reporting the highest mean operating revenue (33.9 million GBP, n=33). Firm size and AI specialisation intensity are primary revenue drivers, while local factors, Level 3 qualification rates, population density, and employment levels, provide significant marginal contributions, highlighting the dependence of AI growth on regional socioeconomic ecosystems. The forecasting models project sectoral expansion to 2030, estimating 4,651 [4,323 - 4,979, 95 percent CI] total entities and a rising dissolution ratio (2.21 percent [-0.17 - 4.60]), indicating a transition toward slower sector expansion and consolidation. These results provide robust evidence for place-sensitive policy interventions: cultivating regional AI capabilities beyond London to mitigate systemic risks; distinguishing between support for scaling (addressing capital gaps) and deepening technical specialisation; and strategically shaping ecosystem consolidation. Targeted actions are essential to foster both aggregate AI growth and balanced regional development, transforming consolidation into sustained competitive advantage.

Code, Capital, and Clusters: Understanding Firm Performance in the UK AI Economy

TL;DR

The study investigates how AI specialization and local socioeconomic contexts shape firm performance in the UK AI economy (2000–2024) using a merged dataset from WAIFinder, glass.ai, and Companies House, augmented with postcode-level census indicators. It employs CatBoost with SHAP to identify drivers of operating revenue and revenue per employee, revealing that firm size drives total revenue while specialization dominates productivity, with local factors contributing meaningful variation. Time-series forecasting (ARIMA, Theta, ETS, MFLES) with conformal prediction projects the UK AI sector toward maturation and consolidation by 2030, estimating approximately total entities (CI: –) and a rising dissolution rate near ; London remains the dominant hub, intensifying regional policy considerations. The findings support place-sensitive policy to cultivate regional AI capabilities, differentiate scaling from specialization, and manage ecosystem consolidation to sustain growth and resilience in the UK AI economy.

Abstract

The UK has established a distinctive position in the global AI landscape, driven by rapid firm formation and strategic investment. However, the interplay between AI specialisation, local socioeconomic conditions, and firm performance remains underexplored. This study analyses a comprehensive dataset of UK AI entities (2000 - 2024) from Companies House, ONS, and glass.ai. We find a strong geographical concentration in London (41.3 percent of entities) and technology-centric sectors, with general financial services reporting the highest mean operating revenue (33.9 million GBP, n=33). Firm size and AI specialisation intensity are primary revenue drivers, while local factors, Level 3 qualification rates, population density, and employment levels, provide significant marginal contributions, highlighting the dependence of AI growth on regional socioeconomic ecosystems. The forecasting models project sectoral expansion to 2030, estimating 4,651 [4,323 - 4,979, 95 percent CI] total entities and a rising dissolution ratio (2.21 percent [-0.17 - 4.60]), indicating a transition toward slower sector expansion and consolidation. These results provide robust evidence for place-sensitive policy interventions: cultivating regional AI capabilities beyond London to mitigate systemic risks; distinguishing between support for scaling (addressing capital gaps) and deepening technical specialisation; and strategically shaping ecosystem consolidation. Targeted actions are essential to foster both aggregate AI growth and balanced regional development, transforming consolidation into sustained competitive advantage.
Paper Structure (12 sections, 3 equations, 5 figures, 3 tables)

This paper contains 12 sections, 3 equations, 5 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Methods
  3. Data collection
  4. Data processing
  5. Model development and feature importance
  6. Timeseries models for forecasting AI growth
  7. Results
  8. The evolution of AI entities since 2000
  9. Revenue generation by AI firms
  10. Drivers of AI firm performance
  11. Forecasting AI firm growth in the UK
  12. Discussion and Conclusion

Figures (5)

  • Figure 1: Evolution of AI Entities in the UK from 2000 to 2024. Variation in the count of AI entities corresponding to active and dissolved status is depicted with temporal-geographical details.
  • Figure 2: The evolution of keywords utilisation and their co-occurrences, respectively, (a) and (c) for Active, and (b) and (d) for Dissolved AI entities.
  • Figure 3: Association of revenue generated by AI entities with the firm-level variables and geographical footprint. (a) Scatter plot of operating revenue with service years, broken down with respect to sectors. (b) Binning the operating revenue against the employee count with respect to the similar count of AI entities. (c) The yearly dynamics of operating revenue per employee in the selected cities in the UK.
  • Figure 4: CatBoost-based modelling for (a) operating revenue and (b) operating revenue per employee of AI entities. SHAP framework provides the percentage contribution of input variable to predict (c) operating revenue and (d) operating revenue per employee for the trained models.
  • Figure 5: Historical trends and model forecasts for AI sector dynamics in the UK, 2000--2030. (a) Total registered AI entities. (b) Active AI entities. (c) Annual dissolutions of AI entities. (d) Annual dissolution rate (dissolutions/active entities). The blue series represents the training data (2000--2019), the green series the validation period (2020--2024), and the orange series the model forecast with 95% prediction interval (2025--2030). The dashed vertical lines separate these periods.