The European Union is moving toward a less demanding version of its emerging climate rules for data centres, according to a draft text reviewed by the Financial Times. The change is subtle in wording but significant in effect: earlier proposals that would have tied compliance more tightly to renewable energy certificates appear to have been scaled back after intense lobbying from large technology companies and parts of the energy market.
For years, the EU has faced a familiar policy dilemma. Data centres are expanding rapidly—driven by cloud computing, artificial intelligence training and inference, and the broader digitisation of industry. At the same time, electricity demand is rising faster than many grids can comfortably absorb it, and the climate impact of that demand is under scrutiny. The EU’s answer has been to require new and existing data centres to demonstrate that their power use aligns with decarbonisation goals. But the question at the centre of the current debate is not whether data centres should be cleaner; it is how the EU should verify “cleaner” in practice.
In the draft now circulating, the approach to renewable energy certificates—often used as a mechanism to prove that electricity consumption is matched with renewable generation—has been weakened compared with earlier drafts. That shift matters because certificates can function as a bridge between where electricity is consumed and where renewable power is produced. They can also become a lightning rod: critics argue that certificates may allow companies to claim progress without guaranteeing that additional renewable capacity is actually built or that the electricity used is genuinely low-carbon at the time and place it is consumed. Supporters counter that certificates are a pragmatic tool in a world where grid constraints and build-out timelines make direct sourcing of renewable power difficult.
The draft’s movement away from stricter certificate-linked requirements suggests the EU is trying to balance two competing realities: the need for credible climate outcomes and the political and commercial pressure to avoid rules that could slow investment or increase costs for operators.
Why certificates became the battleground
Renewable energy certificates are designed to separate the physical flow of electricity from the environmental attributes associated with renewable generation. In theory, a company can purchase certificates corresponding to renewable output and use them to support claims that its electricity use is matched with renewables. In practice, the credibility of those claims depends on the certificate type, the accounting rules, and whether the certificates represent “additional” renewable generation rather than simply reallocating existing renewable output.
Earlier EU proposals reportedly leaned toward a more stringent framework in which certificates would play a larger role in determining whether data centres meet climate requirements. That direction would have forced operators to rely more heavily on specific certificate categories and potentially on tighter eligibility criteria. For Big Tech and other large electricity buyers, that prospect raised immediate concerns: the supply of certain certificate types may be limited, prices could rise, and compliance could become complex across multiple jurisdictions.
Lobbying, in this context, does not necessarily mean a single dramatic intervention. It often looks like sustained engagement over months—technical submissions, consultations, and negotiations aimed at shaping definitions and thresholds. The draft text indicates that such efforts were effective enough to alter the substance of the proposal, at least in the area of certificate-linked compliance.
The EU’s challenge: verification without freezing the market
The EU’s climate policy for data centres is being developed in a landscape where electricity markets are already under strain. Grid operators face constraints, permitting for new generation and transmission can take years, and renewable build-out is uneven across member states. If the EU were to impose very strict requirements that effectively require data centres to secure low-carbon electricity in a way that is not widely available, the result could be delays in construction and expansion—or a wave of legal and political pushback.
At the same time, the EU cannot simply accept that data centres will grow while emissions remain high. The sector’s footprint is becoming too visible, and the political appetite for “greenwashing by accounting” is limited. That is why the certificate question is so central: certificates can be a legitimate instrument for decarbonisation, but they can also be criticised as a substitute for real-world emissions reductions if the rules are too permissive.
The draft’s weakening of certificate-linked requirements can be read as an attempt to keep the policy workable while still steering the sector toward genuine decarbonisation. Instead of making certificates the decisive factor, the EU appears to be moving toward a framework that gives more room for alternative compliance pathways—or at least reduces the extent to which certificates alone determine eligibility.
What the draft change could mean for data centre operators
If the final rules follow the direction suggested by the draft, data centre operators may face less stringent requirements around how they evidence renewable alignment. That could translate into several practical outcomes.
First, the EU may treat the “renewable energy certificate” approach differently than originally proposed. That could mean certificates remain relevant but are no longer required to meet the strictest thresholds, or that only certain certificate types would count fully. It could also mean that the rules will allow a broader set of evidence—such as direct procurement arrangements, long-term contracts, or other mechanisms—without forcing operators into a narrow certificate category.
Second, compliance costs could be lower than under the earlier, stricter framework. Certificates can be expensive, and the cost of compliance is ultimately borne by customers, investors, or both. A softer approach may reduce the risk that data centre expansion becomes financially unattractive in regions where certificate markets are thin or where renewable procurement options are constrained.
Third, the change could affect how quickly operators shift from “paper decarbonisation” to operational changes. If certificates are less central, companies may have stronger incentives to pursue direct renewable procurement, on-site generation, or grid-supporting investments. But there is also a risk: if the rules become too flexible, some operators may continue to rely on certificates without accelerating the build-out of new renewable capacity.
The EU’s negotiating posture: keeping momentum while avoiding a backlash
The draft’s evolution reflects a broader pattern in EU climate regulation: the Commission and member states often start with ambitious proposals, then adjust them through negotiation to account for feasibility, competitiveness concerns, and political realities. Data centres are a particularly sensitive sector because they sit at the intersection of climate policy, industrial strategy, and national economic interests.
Member states may worry about attracting investment. Large technology firms may warn that overly burdensome rules could slow deployment of critical infrastructure. Meanwhile, environmental groups and some policymakers argue that the EU must not dilute standards simply because compliance is inconvenient.
The draft suggests the EU is trying to thread a needle. It wants to ensure that data centres contribute to decarbonisation, but it also wants to avoid creating a compliance regime that is either unworkable or easily gamed. Weakening the certificate-linked element could be seen as a concession to the market—but it could also be a move toward a more nuanced system that better reflects how electricity and renewable attributes actually work.
A unique angle: the policy is really about timing and causality
One reason certificates are controversial is that they can obscure causality. A certificate purchased today may correspond to renewable generation that occurred elsewhere, possibly earlier, and possibly without any direct link to the incremental renewable capacity needed to serve new data centre demand. Critics argue that this breaks the causal chain between increased consumption and increased renewable supply.
The EU’s draft shift may be partly motivated by this causality problem. If the earlier proposal made certificates too determinative, it could have allowed operators to claim compliance without ensuring that their growth leads to additional renewable build-out. By weakening the certificate requirement, the EU may be trying to reduce the chance that the policy becomes a bookkeeping exercise rather than a driver of real-world change.
However, causality is hard to measure. Even direct procurement arrangements can be complex: long-term contracts may not always guarantee additionality, and grid constraints can mean that the physical electricity delivered is not always the low-carbon electricity the contract implies. That is why regulators often fall back on proxies—certificates, contract structures, and accounting rules. The EU’s challenge is to choose proxies that are credible enough to matter, without making compliance impossible.
How this intersects with AI-driven electricity demand
The timing of these negotiations is not accidental. Data centre growth is increasingly tied to AI workloads, which can be electricity-intensive and geographically concentrated. The EU’s policy choices will therefore influence not just emissions accounting but also where AI infrastructure is built and how quickly.
If the EU’s final rules are perceived as too strict, operators might delay expansions or shift investment to jurisdictions with different regulatory regimes. If the rules are too lenient, the EU risks locking in high-emission infrastructure that will be difficult to retrofit later.
The draft’s direction—softening certificate-linked requirements—could be interpreted as a bid to keep investment flowing while still setting a climate expectation. But the real test will come in enforcement and in how the EU defines acceptable evidence. The difference between “less stringent” and “meaningfully weaker” will depend on the final thresholds, the treatment of certificate types, and the degree to which operators must demonstrate progress beyond purchasing instruments.
What to watch next before adoption
Because the draft is not final, several elements could still change. The most important are likely to be:
1) The exact role of renewable energy certificates in the compliance hierarchy
Will certificates be one option among several, or will they remain central but with revised eligibility rules?
2) The definition of what counts as “renewable” evidence
Are there restrictions on certificate vintage, additionality, geographic scope, or verification standards?
3) Whether the rules include performance-based requirements
Some frameworks focus on procurement evidence; others incorporate operational metrics such as energy efficiency, carbon intensity, or reporting obligations. If the EU adds or strengthens performance metrics, the certificate debate may become less decisive.
4) The timeline and transition period
Even a softened rule can still drive change if it includes clear milestones and tightening over time. Conversely, a softened rule without a tightening schedule could reduce urgency.
5) Enforcement and auditability
A policy’s credibility depends on how well it can be checked. If the EU makes compliance easier but also harder to verify, the environmental outcome could suffer.
The broader implication: EU climate policy is becoming more “market-aware