As AI Scales, France’s Competition Authority Raises Energy & Environmental Red Flags

Key Takeaways

• Energy Access as a Competition Risk: Grid constraints and volatile electricity prices could reshape AI markets, particularly as energy costs represent a large share of data-centre operations.
• Frugality Becomes a Market Differentiator: Resource-efficient AI models may lower costs, broaden competition, and create openings for smaller players.
• Standardization Matters: Comparable and scientifically robust environmental metrics are essential to prevent greenwashing and support fair competition.
• Scale Brings Scrutiny: The Autorité is wary of large operators using their market power to secure privileged access to energy, land, or nuclear electricity.
• Transparency Is Foundational: Without reliable data on AI’s environmental footprint, competition based on sustainability cannot function effectively.

Deep Dive

France’s competition authority has turned its attention to a growing tension at the heart of artificial intelligence. As AI systems scale across Europe, the energy and environmental costs of that expansion are becoming a competition issue in their own right.

The Autorité de la concurrence published its first dedicated study examining how the energy and environmental footprint of AI could reshape competition in digital markets. The report builds on the authority’s earlier opinion on generative AI and urges companies, regulators, and policymakers to engage more actively with the risks now emerging around electricity access, resource consumption, and transparency.

While AI’s benefits remain difficult to fully quantify, the Autorité notes that the infrastructure supporting it is already placing visible strain on energy systems and natural resources. The rapid build-out of data centres, which is a strategic priority for both France and the European Union, is driving rising electricity demand, growing pressure on water, land, and rare metals, and a material increase in carbon emissions.

Globally, data centres account for roughly 1.5 percent of electricity consumption today, but that figure could more than double by 2030 as AI workloads expand, potentially reaching as much as 945 terawatt-hours. In France alone, data-centre electricity use could climb from around 10 TWh in the early 2020s to between 12 and 20 TWh by 2030, and as much as 28 TWh by 2035. That would represent close to four percent of national electricity consumption. Against this backdrop, several large operators, particularly U.S.-based firms, have already moved to secure long-term supplies of low-carbon electricity, including nuclear and renewable energy.

According to the Autorité, these trends raise three interconnected competition challenges:

• Access to Energy and Grid Capacity: Difficulties connecting to the power grid and uncertainty around long-term electricity prices risk distorting competition, particularly as energy can account for 30–50 percent of a data centre’s operating costs. In France, the end of the regulated nuclear electricity access mechanism at the end of 2025 is reshaping the market, with new redistribution measures and long-term nuclear supply contracts offered by EDF already being taken up by data-centre operators. The Autorité cautions that these arrangements must not lead to discrimination, foreclosure, or preferential treatment that advantages the largest players over smaller competitors.
• Frugality as a Competitive Parameter: The optimization of resource use to minimize environmental impact is increasingly shaping both demand and supply in AI markets. Public and private buyers are showing greater interest in less resource-intensive tools, while developers are experimenting with smaller models and more transparent reporting of environmental footprints. The Autorité sees frugality as a potential driver of competition, enabling smaller operators to compete through lower costs, lighter infrastructure needs, and efficiency-focused innovation, while warning against misleading claims, inadequate disclosure, or practices that suppress innovation.
• Standardization of Environmental Footprints: As AI’s environmental impact becomes more visible, the Autorité argues that transparent, comparable, and scientifically robust measurement frameworks are essential. Today, information remains limited, methodologies vary, and comparisons are difficult. New tools and emerging French and European initiatives aim to measure energy use, carbon emissions, and lifecycle impacts, and to support eco-design and frugal AI. From a competition perspective, these efforts can strengthen competition if they are inclusive, methodologically sound, and do not facilitate the exchange of commercially sensitive information or exclude certain operators.

Across all three areas, the authority stresses that access to energy, land suitable for data centers, and low-carbon electricity must not become de facto reserved for the largest operators. It also places particular emphasis on the need for reliable, transparent data on AI’s energy and environmental impacts, arguing that meaningful competition on sustainability is impossible without it.

The study concludes by inviting stakeholders to report suspected anticompetitive practices or to seek informal guidance from the Autorité’s General Rapporteur on how sustainability-driven projects can align with competition rules. As AI policy debates increasingly intersect with climate, energy, and industrial strategy, the authority is signaling that competition enforcement will play a central role in shaping how the sector evolves.

The GRC Report is your premier destination for the latest in governance, risk, and compliance news. As your reliable source for comprehensive coverage, we ensure you stay informed and ready to navigate the dynamic landscape of GRC. Beyond being a news source, the GRC Report represents a thriving community of professionals who, like you, are dedicated to GRC excellence. Explore our insightful articles and breaking news, and actively participate in the conversation to enhance your GRC journey.