On 17 December 2025, the FCA published a study (the Study), following an earlier opinion published in June 2024 on the competitive functioning of the generative AI sector (Opinion No. 10 24-A-05) (the Opinion) – you can view the FCA’s English press release here. The Opinion and the Study are of particular interest for all stakeholders in the AI sector, but also energy suppliers and generators, as they start to map out how European competition authorities might think about enforcement priorities in this space in the near future.
Energy access as a specific area of focus
In its Study, the FCA focused on the energy and environmental footprint of AI, highlighting access to electricity and land as potential barriers to entry on this market. The FCA points to:
- grid congestion and length of grid connection delays (up to 5-7 years), especially in congested areas;
- preferential energy supply agreements secured by large digital players;
- excessive reservation of electricity or land capacity, potentially foreclosing access to these inputs for competitors; and
- risks of discriminatory long-term supply contracts (the Power Purchase Agreements or PPA) by incumbent energy suppliers.
According to the FCA, these practices are liable to produce foreclosure effects and raise prices to the detriment of other market participants. This places data centre siting strategies, long-term PPAs and grid capacity reservations within the scope of potential competition scrutiny.
Environmental performance and standardisation
The FCA identifies ‘frugal AI’ (i.e. minimisation of the environmental impact and resource consumption of AI) as a new parameter of competition, capable of fostering innovation and entry. At the same time, it warns against:
- unsubstantiated environmental claims relating to frugal AI (‘greenwashing’);
- refusals to disclose environmental footprint data, in particular by dominant players or through parallel behaviour, which may hinder the emergence of frugality as a parameter of competition by preventing customers from assessing and comparing AI solutions on environmental performance; and
- anti-competitive conduct within standardisation processes – the FCA supports the development of national, European and international standards to measure AI’s environmental footprint, noting potential pro-competitive benefits such as transparency and comparability.
As a result of these risks, environmental claims and participation in standard-setting initiatives require robust competition compliance safeguards.
AI: A growing priority for public authorities
Generative AI has rapidly emerged as a key policy priority worldwide, prompting public authorities to balance innovation, economic competitiveness and risk mitigation. In France, this has led to an early and sustained policy effort, starting with the national AI strategy launched in 2018 and followed, in March 2024, by 25 recommendations from the AI Commission launched by the French Prime Minister, notably aimed at strengthening computing capacity, improving access to data and contributing to global AI governance.
At the European and international levels, this momentum has translated into an increasingly dense regulatory landscape. The EU AI Act (1), which will largely be applicable from 2026, introduces new obligations for providers of large generative AI models, alongside broader digital regulations such as the Data Act (2). Other initiatives such as the UK Bletchley Declaration and international events, for example the February 2025 AI Action Summit in France, reflect the strategic importance which public authorities now attach to AI.
What it means for you
The FCA’s Opinion and Study underline that competition authorities are ready to intervene early to prevent lock-in effects and the reinforcement of barriers to entry in the AI sector, even in the absence of formal dominance findings. Access to key inputs – computing power, cloud infrastructure, data, energy and talent – is already under active scrutiny (you can view our November 2024 briefing on this related topic here).
As a result, competition law must be embedded in the structuring of AI-related deals, including partnerships, cloud agreements, minority investments and vertical integrations. It should also be factored into market entry and growth strategies, as enforcement is likely to combine ex ante digital regulation with traditional antitrust tools.
References
1. Regulation (EU) 2024/1689 of the European Parliament and of the Council of 13 June 2024 laying down harmonised rules on Artificial Intelligence.
2. Regulation (EU) 2023/2854 of the European Parliament and of the Council of 13 December 2023 on harmonised rules on fair access to and use of data.
