ChatGPT developer OpenAI has paused its flagship UK data center project, known as “Stargate UK,” citing high energy costs and regulatory uncertainty. The project was part of a broader £31 billion ($40+ billion) investment plan aimed at expanding artificial intelligence (AI) infrastructure in the country.
The initiative was designed to deploy up to 8,000 GPUs initially, with plans to scale to 31,000 GPUs over time. It was aimed to boost the UK’s “sovereign compute” capacity. This means building local infrastructure to support AI development and reduce reliance on foreign systems.
However, the company has now paused development. An OpenAI spokesperson stated that they:
“…support the government’s ambition to be an AI leader. AI compute is foundational to that goal – we continue to explore Stargate UK and will move forward when the right conditions such as regulation and the cost of energy enable long-term infrastructure investment.”
Energy Costs Are Now a Core Constraint
The main issue is energy. AI data centers require large amounts of electricity to run GPUs and cooling systems.
In the UK, industrial electricity prices are among the highest in developed markets. Recent estimates show costs at around £168 per megawatt-hour, compared to £69 in France and £38 in Texas. This gap creates a major disadvantage for large-scale data center investments.
AI workloads are especially power-intensive. A single large data center can consume as much electricity as tens of thousands of homes. As AI adoption grows, this demand is rising quickly.
Globally, the International Energy Agency estimates that data centers could consume over 1,000 terawatt-hours (TWh) of electricity by 2030, up sharply from about 415 TWh in 2024. This growth is largely driven by AI.
The result is clear. Energy is no longer just a cost. It is a key factor in where AI infrastructure gets built.
Regulation Adds Another Layer of Risk
Energy is only part of the challenge. Regulation is also slowing investment. In the UK, uncertainty around AI rules, especially copyright laws for training data, has created hesitation among companies.
Earlier proposals to allow AI firms to use copyrighted content were withdrawn after backlash. This left companies without clear guidance on compliance.
For large infrastructure projects, this uncertainty increases risk. Data centers require billions in upfront investment. Companies need stable rules before committing capital.
Planning delays and grid connection timelines also add friction. These factors increase both cost and project timelines.
Together, energy costs and regulatory uncertainty create a difficult environment for hyperscale AI infrastructure.
OpenAI’s Global Infrastructure Expands, But More Selectively
Despite the pause, ChatGPT-maker is still expanding globally. The company is investing heavily in AI infrastructure through partnerships with Microsoft, NVIDIA, and Oracle. It is also linked to a much larger $500 billion “Stargate” initiative in the United States, focused on building next-generation AI data centers.
At the same time, the company faces rising costs. Reports suggest OpenAI could lose billions of dollars annually as it scales infrastructure to meet demand.
This reflects a broader industry shift. AI is becoming more like energy or telecom infrastructure. It requires large capital investment, long timelines, and stable operating conditions.
The pause also highlights a deeper issue. AI growth is increasing pressure on energy systems and the environment.
The Hidden Carbon Cost Behind Every AI Query
ChatGPT and similar tools rely on large data centers. These facilities already account for about 1% to 1.5% of global electricity use. Projections for their energy use vary widely due to various factors.
Each individual query may seem small. A typical ChatGPT request can use about 0.3 watt-hours of electricity, which is relatively low. However, usage at scale changes the picture.
ChatGPT now serves hundreds of millions of users. Even small energy use per query adds up quickly. Training models is even more energy-intensive. For example, training GPT-3 required about 1,287 megawatt-hours of electricity and produced roughly 550 metric tons of CO₂.
Newer models are even larger. Some estimates suggest training advanced models like GPT-4 could emit up to 15,000 metric tons of CO₂, depending on the energy source.
At the system level, the impact is growing fast. AI systems could generate between 32.6 and 79.7 million tons of CO₂ emissions in 2025 alone. By 2030, AI-driven data centers could add 24 to 44 million tons of CO₂ annually.
Looking further ahead, global generative AI emissions could reach up to 245 million tons per year by 2035 if growth continues. These numbers show a clear pattern. Efficiency is improving, but total demand is rising faster.
Big Tech Scrambles to Balance AI Growth and Emissions
OpenAI has not published a detailed standalone net-zero target. However, its operations rely heavily on partners such as Microsoft, which has committed to becoming carbon negative by 2030.
The company has acknowledged that energy use is a real concern. Leadership has pointed to the need for more renewable energy, including nuclear and clean power, to support AI growth.
Across the industry, companies are responding in several ways:
- Improving model efficiency to reduce energy per query
- Investing in renewable energy and long-term power contracts
- Exploring new cooling systems to reduce water and energy use
Efficiency gains are already visible. Some AI systems have reduced energy per query by more than 30 times within a year, showing how quickly technology can improve. Still, total emissions continue to rise because demand is scaling faster than efficiency gains.
The Global AI Infrastructure Race
The pause in the UK highlights a larger trend. AI infrastructure is becoming a global competition shaped by energy, policy, and cost.
Regions with lower energy prices and faster permitting processes have an advantage. The United States and parts of the Middle East are attracting large-scale AI investments due to cheaper power and supportive policies.
At the same time, governments are trying to attract these projects. The UK has pledged billions to support AI growth and improve compute capacity. But this case shows that policy ambition alone is not enough. Companies need reliable energy, clear rules, and predictable costs.
AI’s Next Phase Will Be Decided by Energy, Not Code
The decision by OpenAI does not signal a retreat from AI investment. Instead, it reflects a shift in priorities.
Companies are becoming more selective about where they build infrastructure. They are focusing on locations that offer the right mix of energy access, cost stability, and regulatory clarity.
The UK project may still move forward, but only if conditions improve. For now, the message is clear. The future of AI will not be shaped by technology alone. It will also depend on energy systems, policy frameworks, and long-term investment conditions.
The post OpenAI Hits Pause on $40B UK AI Project: Energy Costs Shake Data Center Economics appeared first on Carbon Credits.
What auditors actually test, where projects actually fail, and the contract clauses that protect you before signature.
The meeting happens about fourteen months after the contract was signed. Your assurance provider has reached the nature-based investment line in your Scope 3 file, and the partner across the table has exactly two questions. How do you know the reductions happened? And how do you know they would not have happened anyway?
The first question is MRV: measurement, reporting, and verification. The second is additionality. Between them, they decide whether your nature-based investment counts, in your inventory, in your disclosure, and in front of your board. Everything else in the project documentation is supporting material for these two answers.
This article walks through what each question actually tests, where projects most commonly fail, what digital MRV has changed (and what it has not), and the contract clauses that protect you. The goal is to give you the diligence framework before you sign, because after the credit issues is the wrong time to discover the answers were weak.
What MRV actually verifies
MRV is the machinery that turns a field intervention into a defensible number. Measurement covers the data: biomass surveys, soil sampling, remote sensing, activity records from participating farms. Reporting covers the translation of that data into claimed reductions under a recognised methodology. Verification covers the independent check: an accredited third party tests the reporting against the methodology and the evidence.
The methodologies live in registries. Verra’s Verified Carbon Standard and the Gold Standard are the two largest for nature-based projects, and each publishes the methodology documents, monitoring requirements, and verification protocols that a project must follow. The ICVCM Assessment Framework now sits above the registries, assessing whole methodologies against the Core Carbon Principles and granting the CCP label to those that pass.
For a buyer, the practical questions are concrete. What is the monitoring frequency, and is it specified in the project design document or left vague? Who is the verifier, how were they selected, and how often do they rotate? What raw data do you, the buyer, get access to, and in what format? A project that answers these in writing is a different procurement than one that answers them in a sales call.
What additionality actually proves
Additionality asks whether the intervention caused the reduction, or whether the reduction would have happened anyway. The test is a counterfactual: what would this landscape, this farm, this forest have done without the project’s money?
Three forms matter in practice. Financial additionality asks whether the project needed the carbon revenue to proceed. Regulatory additionality asks whether the activity was already required by law. Common-practice additionality asks whether the activity is already standard in the region, in which case paying for it buys you nothing the world was not getting for free.
The reason additionality dominates audit conversations is recent history. Research published in 2023, including the Science paper examined at length in our piece on conventional offsets and boardroom credibility, found that a large share of REDD+ credits failed the counterfactual test because baselines were inflated. The market response was a wave of methodology revisions at Verra and the arrival of independent ratings agencies whose entire business is re-testing additionality claims. The Carbon Credit Quality Initiative publishes transparent scoring of methodologies on exactly this dimension, and it is free to consult before you sign anything.
Where projects most commonly fail the test
Five failure modes account for most of the wreckage.
- Inflated baselines. The counterfactual assumes more deforestation, more degradation, or lower yields than the evidence supports. The claimed reduction is the gap between reality and the baseline, so an inflated baseline manufactures reductions from nothing.
- Unaccounted leakage. The project protects one forest and the logging moves to the next valley. The methodology is supposed to net this out; weak projects estimate it optimistically.
- Thin permanence protection. Nature-based carbon can reverse: fire, pest, drought, or a change of landowner. Buffer pools and insurance mechanisms exist for this, but their adequacy varies enormously between projects.
- Attribution and double counting. In supply chain settings, the same reduction can be claimed by the supplier, the buyer, and a credit purchaser unless contracts prevent it. Our Insetting vs Offsetting piece covers the inventory rules; the point here is that the auditor will ask who else is counting this tonne.
- Stale monitoring. Data collected at validation and never refreshed. The IPCC AR6 Working Group III land-sector chapter documents how quickly carbon stocks respond to disturbance; a three-year-old measurement is a historical artifact, not a current claim.
What digital MRV changes, and what it does not
Digital MRV is the genuine improvement in the field. Satellite remote sensing, including the free archives at NASA Earthdata, allows biomass and land-cover change to be monitored continuously rather than at multi-year verification intervals. Soil carbon models calibrated with physical sampling reduce the cost of agricultural measurement. The practical effect is more frequent data at lower cost, which compresses the window in which a problem can hide.
What digital MRV does not change is judgment. Baselines are still human decisions about counterfactuals. Additionality is still an argument, not a measurement. Research groups such as the Oxford Smith School have been clear on this point: better sensors improve the M in MRV, but the integrity questions live in the assumptions, and assumptions need governance, not gadgets.
For a buyer, the test is simple. Ask the provider what is measured by instrument, what is estimated by model, and what is assumed by methodology. A provider who can answer that question crisply understands their own evidence chain. A provider who cannot is selling you their confidence rather than their data.
What to require in your contract
The diligence above converts into five contract clauses.
- Monitoring cadence and buyer data access, specified by dataset and frequency.
- Verifier independence, named accreditation, and rotation terms.
- Baseline revision triggers, so the counterfactual updates when the methodology or the evidence changes.
- Reversal liability and buffer adequacy, with the mechanism named and sized.
- Documentation handover in audit-ready form, so the evidence file your assurance provider needs already exists.
None of these clauses is exotic. All of them are absent from weak contracts, and their absence is the most reliable early signal that the MRV and additionality answers will be weak too.
If you are evaluating a nature-based investment and want the MRV and additionality stress-tested before signature rather than after, the carbon and sustainability experts at Carbon Credit Capital can run that review against any project on your shortlist, and design nature-based supply chain investments where the evidence chain is built audit-first. Schedule a consultation.
Sources and further reading
- ICVCM: Core Carbon Principles Assessment Framework
- Verra: Verified Carbon Standard
- Gold Standard for the Global Goals
- Carbon Credit Quality Initiative: Methodology quality scores
- University of Oxford Smith School: Sustainable finance research
- IPCC AR6 Working Group III, Chapter 7: AFOLU
- NASA Earthdata satellite remote sensing archive
EU Directive: Empowering Consumers for the Green Transition (ECGT)
The EU Directive, Empowering Consumers for the Green Transition (ECGT), takes effect on September 27, 2026.(1) The goal of ECGT is to protect consumers by ensuring that environmental claims are fair, understandable, and reliable. This regulation does create a new compliance requirement for businesses, but it also provides sustainability and marketing teams with important guidance that helps create consistency in sustainability communications.
Key takeaways
- ECGT takes effect September 27, 2026, and prohibits claims that a product or service has a neutral, reduced, or positive environmental impact based on offsetting alone.
- Named example phrases the regulation prohibits include climate neutral, CO2 neutral certified, carbon positive, climate net zero, climate compensated, reduced climate impact, and limited CO2 footprint.
- ECGT does not want to deter investment in carbon credits. It wants companies to communicate the real benefits of the projects they support instead.
- SBTi’s guidance recommends framing carbon credits as taking responsibility for ongoing emissions, not as making a product or company neutral.
- Voluntary carbon projects deliver real climate progress: reducing super-pollutants, protecting and restoring ecosystems, and supporting communities.
Regarding carbon credits specifically, voluntary carbon projects deliver important climate progress and environmental benefits that provide many talking points for companies. They reduce climate super-pollutants by removing industrial emissions like methane, N2O, HFCs and others. They protect and restore valuable ecosystems and carbon sinks like forests, mangroves and grasslands. They help communities by reducing local pollution, creating employment opportunities, improving access to healthcare, and more.
The Science Based Targets Initiative (SBTi), a global leader in business climate action, concludes that alongside aggressive decarbonization, we should also use high quality carbon credits to take responsibility for our ongoing emissions. SBTi recognizes that carbon credits are important “to help limit temperature overshoot, mitigate transition risks, and support climate solutions.”(2)
ECGT language on carbon offsetting says that they do not want to deter investment in carbon credits. They just want companies to focus on communicating the benefits of the projects they support and avoid claims beyond the scope of carbon credits, which is good for everyone, companies and consumers alike.
The regulation reinforces that carbon credits do not change the sustainability of your products, so carbon credit buyers should not suggest that their products are more sustainable because of carbon credits. Instead, companies need to promote their climate contributions as a way to compensate or take responsibility for their carbon emissions by supporting projects that do great things like reducing global carbon emissions, reducing pollution, preventing deforestation, restoring forests, and more.
ECGT language related to carbon offsetting
The regulation is particularly focused on prohibiting claims, based on offsetting greenhouse gas emissions, that a product or service has a neutral, reduced, or positive impact on the environment in terms of greenhouse gas emissions. These claims are prohibited in all circumstances because they mislead consumers into believing the claim relates to the product itself, or to how it was made and supplied, or into thinking that using the product carries no environmental impact at all.
Named examples of prohibited claims include:
- climate neutral
- CO2 neutral certified
- carbon positive
- climate net zero
- climate compensated
- reduced climate impact
- limited CO2 footprint
These claims are only allowed when they rest on a product’s actual lifecycle impact, not on offsetting emissions outside that product’s value chain, since the two are not equivalent. This prohibition does not stop companies from advertising their investments in environmental initiatives, including carbon credit projects, as long as they present that information in a way that is not misleading and that meets the other requirements of Union law.(1)
SBTi also provides guidance on climate contribution language in its Corporate Net Zero Standard Version 2.0 Draft for Second Public Consultation, November 2025. While the SBTi language is fairly technical, it has a good framework for crafting a climate contribution message.
SBTi Language for Carbon Credits(3)
- Take responsibility for ongoing emissions by delivering mitigation impact contributions
- Carbon credits certify the mitigation outcomes of projects that reduce, avoid, or remove carbon emissions
- Activities that reduce emissions from emission sources not located within the company’s value chain
- Activities that conserve, protect, and enhance natural carbon sinks
- Activities that capture and store carbon in storage pools
SBTi’s draft standard also walks through sample claim language for this kind of contribution. In general, the samples move from a simple percentage statement, to naming a specific verified tonnage tied to that percentage, to a fuller statement that breaks the tonnage into reductions versus removals. Across all three, the framing stays consistent: a company took responsibility for a defined share of its ongoing emissions over a set period, by funding a specific, verified amount of mitigation, achieved through emission reductions or removals.(3)
FAQ: ECGT and Carbon Credit Claims
When does the ECGT directive take effect?
The rules apply across the EU from September 27, 2026, after member states transposed the directive into national law by March 27, 2026.
Does ECGT ban carbon offsetting?
No. It bans specific marketing claims that a product or service is environmentally neutral, reduced impact, or positive based on offsetting. Advertising investment in carbon credit projects themselves is still allowed if it is not misleading.
What phrases does ECGT specifically prohibit?
Named examples include climate neutral, CO2 neutral certified, carbon positive, climate net zero, climate compensated, reduced climate impact, and limited CO2 footprint, when those claims are based on offsetting rather than a product’s actual lifecycle impact.
How should a company describe its carbon credit purchases instead?
SBTi’s guidance recommends stating the specific verified tonnage of emissions reductions or removals funded and describing that as taking responsibility for a defined share of ongoing emissions, rather than claiming the company or product is neutral.
Does this rule apply to company level sustainability claims too?
ECGT is focused on claims about specific products and services in consumer marketing. Broader company level sustainability communication is a separate matter still governed by other existing rules.
While ECGT does add a new compliance burden for businesses, it helps create consistency in sustainability messaging that is important to building confidence in voluntary carbon projects and scaling the industry to help us achieve progress on global carbon emissions.
Disclaimer: Terrapass does not provide legal or regulatory advice. Any interpretation of regulation must be approved by your legal representative.
References:
(1) https://eur-lex.europa.eu/eli/dir/2024/825/oj
(2) https://files.sciencebasedtargets.org/production/files/Corporate-Net-Zero-Standard-version-2.pdf
(3) https://files.sciencebasedtargets.org/production/files/CNZS-V2-Second-Consultation-Draft.pdf
The post The EU’s New Green Claims Rules and Carbon Credits appeared first on Terrapass.
Most businesses that decide to act on their net-zero targets reach the same point of friction. Buying carbon credits has meant tracking down brokers, sitting through sales calls, and requesting a quote just to learn a price, sometimes with limited proof of what you are buying.
-
Climate Change1 year ago
Guest post: Why China is still building new coal – and when it might stop
-
Greenhouse Gases2 years ago嘉宾来稿：满足中国增长的用电需求 光伏加储能“比新建煤电更实惠”
-
Greenhouse Gases1 year ago
Guest post: Why China is still building new coal – and when it might stop
-
Climate Change2 years ago嘉宾来稿：满足中国增长的用电需求 光伏加储能“比新建煤电更实惠”
-
Renewable Energy11 months agoSending Progressive Philanthropist George Soros to Prison?
-
Climate Change2 years ago
Bill Discounting Climate Change in Florida’s Energy Policy Awaits DeSantis’ Approval
-
Greenhouse Gases1 year ago
嘉宾来稿：探究火山喷发如何影响气候预测
-
Carbon Footprint2 years agoUS SEC’s Climate Disclosure Rules Spur Renewed Interest in Carbon Credits