Search Climate Insights Directory

The US faces a critical bottleneck in connecting new electricity generation and storage to the grid, with over 2.2 terawatts of projects currently in queues. While the US Department of Energy has pushed for streamlined load interconnection for data centers, the author argues that this is insufficient without parallel reforms to generator interconnection. Current processes are characterized by long delays—averaging nearly five years from request to operation in 2024—and high network upgrade costs that often exceed 10% of project capex, leading to high attrition rates. The brief proposes several solutions: adopting advanced AI and automation software for studies, integrating Grid-enhancing technologies (GETs) to reduce costs, improving transparency in network upgrade construction, and implementing 'fast track' pathways such as energy-only interconnection, surplus interconnection service, and generator replacement.

Gham Power installed an 83.16 kWp rooftop solar PV system for Datahub Nepal in Tilottama, Butwal, to replace a reliance on diesel generators and an unreliable utility grid. The project, completed on 2 July 2024, provides stable daytime power, reducing daytime electricity costs by 68.11%. Over a 25-year period, the system is projected to displace 38 kiloliters of diesel and reduce carbon emissions by 220 tonnes.

The document outlines the climate and social risks associated with the expansion of data centers, highlighting their massive energy and water consumption. It warns journalists against uncritically accepting corporate claims about job creation and investment, urging them to investigate local resource competition, 'digital sacrifice zones,' and extractivist dynamics in the Global South.

The report examines the risks of AI integration in the ASEAN power sector, highlighting data limitations, regulatory uncertainty, and cybersecurity vulnerabilities. It argues that while AI is a 'low-hanging fruit' for accelerating the energy transition, its success depends on strengthening data foundations, updating grid codes, and building AI-ready workforces through phased roll-outs and 'AI sandboxes'.

The report argues that AI can significantly accelerate the energy transition in ASEAN by managing the complexity of rising variable renewable energy (VRE) shares. It identifies five key AI applications—forecasting, predictive maintenance, dispatch optimisation, real-time control, and dynamic line rating—that can reduce operational costs and emissions. While ASEAN shows strong digital readiness, current AI adoption is fragmented and pilot-based. The report estimates that widespread AI adoption could save up to $67 billion and reduce nearly 400 million tonnes of CO2 by 2035, provided that regulatory, data, and cybersecurity challenges are addressed.

This report finds that widespread AI adoption in ASEAN power systems could deliver up to $67 billion in cumulative cost savings and reduce nearly 400 million tonnes of CO2 by 2035.

A survey of 930 registered U.S. voters reveals broad support for clean energy research and policies, such as soil restoration (88%) and carbon dioxide regulation (77%). While 58% prefer candidates who support climate action, significant political polarization remains regarding the priority of global warming for the president and Congress. Notably, 67% of voters believe global warming affects the U.S. cost of living, primarily impacting utility bills and groceries. The report also highlights a majority opposition (58%) to local data center construction due to concerns over electricity costs, and a general trust in climate scientists (71%) and primary care doctors (65%) as information sources.

The briefing proposes Bring-Your-Own (BYO) tariffs to allow large energy users, such as data centers, to fund their own power resources to speed up grid connection and reduce financial risks for other ratepayers. It distinguishes these from 'baseline' tariffs and highlights specific examples of BYO and Clean Transition programs in the US, including initiatives by PJM, NV Energy, and Georgia Power.

The report highlights that 'next generation geothermal' technologies, such as Enhanced Geothermal Systems (EGS), have expanded the geographic viability of geothermal power beyond volcanic regions. In the EU, approximately 43 GW of capacity could be developed at costs below 100 €/MWh, potentially generating 301 TWh of electricity annually. While Europe has a strong historical foundation in geothermal energy, the report warns that the EU risks losing its leadership to the US and Canada due to slow permitting and a lack of coordinated policy frameworks. It also notes geothermal's potential to meet the rising energy demands of AI data centres and its ability to provide firm, low-carbon power and mineral recovery, such as lithium.

The report argues that the AI industry uses a 'bait-and-switch' greenwashing tactic, where the proven climate benefits of 'traditional' AI (such as predictive models) are used to justify the massive energy and water consumption of 'consumer generative AI' (such as ChatGPT). An analysis of 154 climate benefit claims found that 97% related to traditional AI, while only 3% related to consumer generative AI, with no evidence found that the latter leads to substantial emissions reductions. Furthermore, the report finds that evidence for these benefits is generally weak, with 36% of claims citing no evidence and only 26% citing published academic research. This narrative masks the immediate climate impacts of data centre expansion, which the report states is boosting demand for fossil fuels and risking the ability of regions like Ireland and Australia to meet carbon targets.

Reclaim Finance's 2026 report reveals that 62% of analyzed unit-linked accounts (UC) in French life insurance are exposed to fossil fuel developers, despite growing saver demand for sustainable investments. The study highlights a systemic 'blind spot' where insurers apply climate policies to their general funds (fonds euro) but ignore them for UC. Specifically, 61% of UC are exposed to oil and gas developers, and 15% to thermal coal. The report also identifies significant greenwashing risks, noting that 47% of funds with 'sustainable' or 'ESG' labels still hold fossil fuel developers. It recommends that insurers stop referencing new fossil-exposed funds, demand robust climate strategies from asset managers, and calls for public authorities to implement an 'eco-score' for savings products.

The United States now leads the world in gas-fired power capacity in development, with nearly 252 GW planned, driven significantly by the energy demands of AI and data centers. Globally, the pipeline for gas power grew by 31% in 2025 to 1,047 GW, though expansion is currently bottlenecked by turbine production backlogs stretching to 2030. Texas is the primary hub for this growth in the U.S., with 80.6 GW in development, nearly half of which is intended to power data centers directly.

A joint submission by multiple NGOs and experts to the UN Special Rapporteur detailing systemic human rights violations in the U.S. linked to per- and polyfluoroalkyl substances (PFAS). The authors argue that the U.S. government has failed to regulate PFAS as a class, allowed military and industrial contamination of drinking water, and permitted corporate structures to shield polluters from accountability. The report highlights disproportionate impacts on Arctic Indigenous Peoples and Black and Latino communities, while noting the global transboundary movement of PFAS waste and pollution originating from the U.S.

The document details the use of a modified Regional Energy Deployment System (ReEDS) model to project US electricity demand from data centers and the resulting effects on the energy transition. It outlines five demand scenarios, including 'high data center demand growth', and evaluates them against various policy frameworks such as the Inflation Reduction Act (IRA) and the One Big Beautiful Bill Act (OBBBA). The analysis incorporates specific state-level emissions caps for Illinois, Michigan, and Wisconsin, and adjusts technology costs for nuclear and carbon capture and storage (CCS) to reflect conservative estimates. It also details the calculation of health damages from criteria pollutants and the social cost of greenhouse gas emissions.

The Union of Concerned Scientists analysis finds that data center load growth could increase Illinois electricity system costs by $24 billion to $37 billion by 2050 and drive the state to become an energy importer as early as 2030. Under current policies, this growth would increase reliance on in-state fossil fuel plants and out-of-state imports, leading to higher CO2, methane, NOx, and SO2 emissions. The report estimates that this growth could cause between 68 and 403 additional mortalities and up to $124 billion in climate damages by 2050. To mitigate these effects, the document recommends requiring data centers to secure new carbon-free electricity resources, increasing transparency in power reporting, and restoring federal clean energy tax credits.

The Union of Concerned Scientists analysis finds that Michigan's electricity demand could nearly double by 2050, driven significantly by AI-related data centers. While current policies may lead to increased fossil gas capacity and emissions due to an energy export loophole, a proposed CO2 reduction policy could enable Michigan to meet 95% of its 2050 demand with renewables and low-carbon sources, avoiding $3.5 billion in local health costs and $408 billion in global climate damages.

The report argues that the United States can meet surging electricity demand from AI data centers using clean energy while phasing out fossil fuels, provided that stronger state and federal policies are implemented. It warns that under current policies, data center growth will increase reliance on gas and coal, leading to higher emissions and significant costs for ratepayers. Conversely, adopting clean energy policies—such as restoring federal tax credits and implementing carbon-free electricity standards—could save trillions of dollars in climate and health damages, far outweighing the costs of the energy transition.

The report finds that data centers are the primary driver of near-term electricity demand growth in Wisconsin, which risks increasing reliance on fossil fuels under current policies. By implementing a Clean Electricity Standard and CO2 reduction policies, Wisconsin could generate 83% of its electricity from clean energy by 2050 and reach net-zero CO2 emissions. Such policies would avoid significant health-related costs and mortalities compared to current trajectories. The report recommends reintroducing Integrated Resource Planning and establishing a clean electricity standard to protect ratepayers and the climate.

In 2025, solar and wind power expanded fast enough to outpace global electricity demand growth for the first time over a sustained period. Solar growth accelerated in Africa and Central Europe, while BRICS nations now generate over half of global solar power. The rise of 'electrotech'—led by China—and falling battery costs are enabling fossil-free baseload power and supporting the energy demands of AI and data centres.

Earthjustice responds to a DOE request for information on accelerating power delivery for AI and industrial growth. The organization argues against reviving retired coal plants due to high costs and unreliability, instead advocating for the rapid deployment of wind, solar, and battery storage. It recommends that the DOE address speculative and inflated data center load forecasts, automate interconnection studies to reduce timelines, and deploy Advanced Transmission Technologies (ATTs) to increase grid capacity quickly. Additionally, it emphasizes the need for early community engagement in permitting and the long-term expansion of interregional transmission lines.