AMSTERDAM, NETHERLANDS — NVIDIA, currently crowned the world’s most valuable publicly traded company by market capitalization, stands accused of deploying sophisticated greenwashing strategies to obscure a staggering surge in greenhouse gas emissions. A landmark investigative briefing released by a consortium of international civil society groups reveals that NVIDIA’s indirect supply chain emissions have skyrocketed by an astonishing 725% since 2020.
Furthermore, analysts estimate that undisclosed emissions stemming from the operational lifespan of the millions of advanced graphics processing units (GPUs) shipped globally could double that footprint. As the primary engine behind the global generative artificial intelligence (AI) boom, the tech giant faces mounting scrutiny over the true ecological cost of its meteoric rise.
Executive Overview
The intersection of artificial intelligence and environmental sustainability has become one of the defining geopolitical and economic flashpoints of the decade. At the center of this storm is NVIDIA, whose hardware dominates the global market for deep learning and generative AI applications. While the corporate leadership in Santa Clara projects an image of environmental stewardship—frequently touting the potential of AI to optimize energy grids and combat climate change—critics argue this narrative serves as a smokescreen.
The new briefing, spearheaded by Greenpeace East Asia in collaboration with organizations such as Greenpeace International, Beyond Fossil Fuels, Friends of the Earth, the Green Screen Coalition, Stand.Earth, Kairos Fellowship, and Climate Action Against Disinformation, marks the first comprehensive dissection of NVIDIA’s environmental messaging. Authored by analyst Ketan Joshi, the report systematically dismantles the company’s public relations claims, revealing a widening chasm between corporate rhetoric and physical reality.
The core of the controversy lies in NVIDIA’s business model: chips are heavily reliant on energy-intensive manufacturing ecosystems in East Asia and deployed worldwide into power-hungry data centers. These facilities are increasingly drawing on fossil-fuel energy sources—exemplified by NVIDIA’s recent hardware commitments to massive, gas-powered data centers in the United States—driving local environmental degradation and accelerating global climate instability.
Detailed Chronology and Investigative Findings
To understand the trajectory of NVIDIA’s environmental footprint, analysts examined corporate disclosures, financial reports, and supply chain data spanning from 2020 through projected figures for 2026 and 2027.
2020–2023: The Incubation of the AI Gold Rush
At the dawn of the decade, NVIDIA’s primary market expansion began shifting aggressively toward accelerated computing and large-scale data center infrastructure. During this period, the company’s Scope 3 emissions—indirect greenhouse gas emissions occurring up and down its value chain, particularly in raw material extraction and manufacturing—began a steep upward trajectory. Despite early warnings from environmental watchdogs regarding the carbon intensity of semiconductor fabrication plants in Taiwan and broader East Asia, corporate disclosures largely treated these impacts as peripheral externalities.
2024–2025: The Generative AI Explosion and Disclosures
As generative AI captured the global imagination following the widespread adoption of large language models, demand for NVIDIA’s H100 and subsequent generation chips created unprecedented financial records. However, this surge in hardware production catalyzed an exponential rise in environmental strain. The new briefing reveals that between 2020 and the mid-2020s, NVIDIA’s Scope 3 emissions surged by 725%.
Simultaneously, calculations tracking the energy consumption of deployed hardware indicate that emissions resulting from the active use of sold chips could reach up to 21 million tonnes of carbon dioxide equivalent ($textCO_2texte$) by 2025. This figure underscores a critical blind spot in standard corporate carbon accounting: the emissions generated by third parties utilizing a company’s products often dwarf operational footprints.
2026–2027: Financial Supremacy Meets Physical Reality
By fiscal year 2027, financial metrics highlighted the absolute dominance of a single sector within NVIDIA’s balance sheet. According to NVIDIA’s Q2 FY27 financial disclosures, the data center segment—driven almost entirely by generative AI hardware demand—accounted for a staggering 92.5% of the company’s total revenue.
Yet, this financial triumph coincided with intensifying physical backlashes. In mid-2026, public controversies erupted over massive data center projects, such as the high-profile integration of NVIDIA hardware into a mega-scale facility in Ohio, reported to be powered directly by one of the world’s largest new natural gas plants. Rather than accelerating a transition to green grids, the localized deployment of NVIDIA’s technology has frequently anchored digital infrastructure to fossil fuel baseloads for decades to come.
Supporting Context & Metrics
The investigation contextualizes NVIDIA’s environmental impact through several distinct operational pillars, metrics, and academic frameworks:
1. The Anatomy of Scope 3 Emissions
Corporate greenhouse gas accounting is divided into three scopes. Scope 1 covers direct emissions from owned or controlled sources. Scope 2 covers indirect emissions from the generation of purchased electricity, steam, heating, and cooling consumed by the reporting company. Scope 3, however, encompasses all other indirect emissions that occur in a company’s value chain.
For NVIDIA, Categories 1 through 8 of Scope 3 are the primary drivers of its carbon expansion. These include upstream emissions from silicon wafer fabrication, specialized chemical sourcing, and component assembly, as well as downstream emissions from the multi-year operation of millions of GPUs running continuous machine learning workloads.
2. Debunking NVIDIA’s Three Core Climate Claims
The Greenpeace briefing identifies and debunks three primary rhetorical strategies utilized by NVIDIA to deflect environmental criticism:
- The "Net Benefit" Fallacy: NVIDIA frequently claims that the downstream analytical power of AI will ultimately solve climate change by optimizing global systems. Analysts point out that this is a speculative distraction used to justify guaranteed, immediate, and compounding emissions in the present.
- The Renewable Growth Distortion: The company suggests that data center demand response programs actively foster renewable energy grid growth. In practice, as demonstrated by fossil-fuel-backed facilities in Ohio and elsewhere, tech conglomerates often sign power purchase agreements that strain local grids and perpetuate fossil fuel reliance.
- Efficiency as Sustainability: NVIDIA heralds the generational energy-efficiency gains of its silicon architecture as proof of progress. While individual chips require less energy per floating-point operation (FLOP) than predecessors, the Jevons paradox applies: increased efficiency lowers the effective cost of computing, leading to a massive net expansion in total computational volume and absolute energy consumption.
3. Transnational Inequity and "Tech Colonialism"
The environmental burden of the AI revolution is not distributed evenly. According to research cited from the United Nations University and Greenpeace East Asia’s Nvidia’s Green Illusion, the lifecycle of an AI chip relies heavily on industrial centers in East Asia, where power grids historically feature high fossil-fuel mixes.
Furthermore, the physical infrastructure required to support data centers—ranging from massive water consumption for cooling towers to rare-earth mineral mining and e-waste generation—disproportionately impacts local communities and lower-income nations in the Global Majority. Critics characterize this dynamic as "tech colonialism," wherein private corporations monopolize astronomical financial profits while offloading ecological degradation onto vulnerable populations.
Official Statements
The release of the briefing has drawn sharp rebukes from leading environmental advocates and civil society leaders who demand immediate accountability from the tech sector.
Avex Li, Project Lead for Greenpeace East Asia’s AI Supply Chain Campaign, stated:
"NVIDIA wants the world to believe that AI will solve the climate crisis, while ignoring the rapidly growing emissions behind its own products and billion-dollar supply chain networks. As one of the world’s biggest beneficiaries and a key player in the AI frenzy, with immense market power, cash flow, and influence, NVIDIA cannot distance itself from the real-world environmental impacts of its business."
Xueying Wu, Campaigner on Big Tech for Greenpeace International, added:
"NVIDIA is driving a system of tech colonialism: monopolising the profits of AI while offloading its ecological impacts onto the Global Majority. Meanwhile, NVIDIA together with other tech giants is inflating the AI bubble that gambles with global financial stability, concentrating more and more power in tech corporations’ hands."
Civil society coalitions are demanding a fundamental pivot in how technology companies approach public utility and environmental limits, asserting that digital infrastructure must serve the public good rather than speculative financial markets at the expense of planetary boundaries.
Future Outlook
As NVIDIA continues to post record-breaking valuations and expand its footprint across global markets, the collision course between exponential computing demand and global climate targets is reaching a critical juncture.
The findings from Ketan Joshi and the international consortium of NGOs signal the end of an era where tech companies could market hardware innovation as inherently green. Moving forward, regulatory bodies, institutional investors, and climate advocates will likely demand unprecedented transparency regarding Scope 3 supply chain footprints and real-world energy sourcing.
To maintain any credible claim to corporate responsibility, NVIDIA and its industry peers face an unambiguous imperative: they must transition from rhetorical mitigation to direct, capital-intensive investments in local renewable energy infrastructure, transparent lifecycle accounting, and strict accountability for the downstream emissions of their technologies. Without such a transformation, the silicon foundation of the AI revolution risks destabilizing the very global climate upon which human civilization depends.
