Concerns have been raised about the potential for market concentration in generative AI (GenAI), if incumbent firms are able to leverage their advantages in data, compute, and distribution to entrench their positions and distort competition in emerging GenAI applications. In a new industry report prepared for the Computer & Communications Industry Association (CCIA), we analyse the competitive dynamics in GenAI globally and in Europe and find there is no current evidence these concerns are likely to materialise.
We examine global GenAI developments in investment, cost, and technical capabilities, then focus on Europe's unique position, identifying both its broader challenges and emerging opportunities in deploying GenAI. Drawing on an economic framework rooted in theory, we assess the current state of competition in GenAI by analysing real-world deployment case studies and the features of strategic partnerships.
Key Findings
The GenAI industry currently shows signs of vibrant competition. Despite early, forward-looking concerns, we see diverse business models, with both specialised firms and vertically integrated solutions competing in the same markets, multi-sourcing strategies across companies of all sizes, and technical solutions lowering switching costs. Open source models play a crucial role as a competitive constraint by providing alternatives to proprietary platforms and facilitating fine-tuning for specialised applications. Partnerships are actively fostering innovation by pooling resources and combining capabilities, and have not been found to raise any competition concerns so far: though each must be assessed on its individual merits. High levels of investment, entry, and innovation continue, with no evidence of the concentration or foreclosure that observers feared thus far.
Europe faces real challenges but has significant opportunities. European companies lag behind their US counterparts in private capital and funding, which hinders their ability to scale and retain talent. Additionally, market fragmentation across different languages and regulatory frameworks bring about even greater complexity. However, Europe's deep industrial expertise in sectors like luxury and automotive engineering, coupled with strong cross-border R&D collaboration, creates significant opportunities for GenAI deployment, productivity gains and growth.
Policy recommendations:
Competition policy: Active monitoring is appropriate given the rapid pace of technological change, but premature intervention could disrupt effective competition or chill innovation during critical development phases. Enforcement action should only be taken upon clear evidence of harm.
Broader policy: Policymakers should consider lowering structural barriers to entry, for instance by facilitating access to capital and simplifying regulatory frameworks that GenAI companies navigate, while promoting widespread GenAI adoption across European industries.
This analysis was led by Katie Curry and Meryem Haraj Touzani, supported by Jimmy Gardebrink, David Henriques, and Johanna Lunghini.
Katie and Meryem will further discuss these findings during a panel session at CCIA’s upcoming conference on 12 June 2025.
The preparation of this paper has been supported by the Computer & Communications Industry Association. The opinions expressed are those of RBB Economics and do not necessarily represent the views of CCIA or of its members.
