Leading artificial intelligence companies are staying silent on their climate impact even as their data centers consume enormous amounts of energy, a dramatic reversal from the corporate sustainability trend of just a few years ago. OpenAI, Anthropic, and SpaceX's AI division have not published greenhouse gas emissions reports, set net-zero targets, or released sustainability plans, despite reaching valuations near one trillion dollars.

Why Are AI Giants Avoiding Climate Commitments?

The silence reflects a broader retreat from environmental, social, and governance (ESG) investing across Wall Street. Funds focused on ESG or environmental themes attracted about $485 billion in 2021, but those same funds lost $82 billion in 2025. This shift has removed much of the investor pressure that once pushed companies to disclose their environmental footprint and set ambitious climate goals.

The contrast with earlier corporate behavior is striking. When Tesla went public in 2010, the company prominently cited environmental concerns with gasoline-powered cars and discussed how regulatory support for clean energy helped it scale. When Rivian Automotive filed for its initial public offering in 2021, founder Robert Schultze said climate impact was behind "every decision" the company made. But when SpaceX went public this year, environmental concerns barely appeared in its filing, despite the company deploying gas turbines to power massive data centers.

The political environment has shifted dramatically. Under the second Trump presidency, the Securities and Exchange Commission (SEC) has moved to scrap climate disclosure rules, and Republican lawmakers have attacked ESG initiatives. Major Wall Street banks, including BlackRock, departed the Net-Zero Banking Alliance, leading it to cease operations.

"Big institutional investors have kind of pulled away from the pressure that they used to put on companies around sustainability," said Amanda Urquiza, a partner at Wilson Sonsini Goodrich and Rosati, who advises pre-IPO companies.

Amanda Urquiza, Partner at Wilson Sonsini Goodrich and Rosati

What Is the Scale of AI's Climate Impact?

The stakes are enormous. New gas plants being built to power data centers in the United States alone could soon generate as much climate pollution as the entire country of Australia, according to research by the nonprofit Environmental Integrity Project. Yet the public has almost no visibility into how much greenhouse gas pollution is tied to the large language models (LLMs) that power ChatGPT, Claude, and other AI systems, or whether companies plan to reduce their emissions.

Anthropic has joined a carbon-removal coalition called Frontier, and the company is working with Watershed, a carbon accounting platform, to measure its company-wide footprint ahead of regulatory deadlines. OpenAI is coordinating with its data center partners and has established a sustainability steering committee, though the company has not publicly disclosed its emissions.

How Will Regulation Force Disclosure?

Relief may come from California. The state's law SB253 begins to take effect in November, requiring companies with more than $1 billion in revenue doing business in California to report their Scope 1 and Scope 2 greenhouse gas emissions. These measures account for the planet-warming pollution caused by a company's operations and energy usage, and California's regulations apply to activity outside the state.

The AI companies are already staffing up to meet these deadlines. Anthropic hired a veteran of Salesforce to help build out its sustainability and non-financial reporting capabilities and posted a job listing for a director of infrastructure and energy accounting. OpenAI has posted on LinkedIn for a head of ESG reporting and operations as well as for clean energy and environmental leads.

"If I don't have to talk about a potentially large risk that could sit on my balance sheet, and I'm not required to do it and shareholders are not asking about it, then what incentive do I really have to talk about it?" said Todd Cort, a senior lecturer in sustainability at Yale School of Management.

Todd Cort, Senior Lecturer in Sustainability at Yale School of Management

What Are the Challenges in Measuring AI Emissions?

Even as companies prepare for California's reporting deadline, the technical work of measuring AI's climate footprint remains incomplete. There is no standardized way to measure emissions from training a large language model or to account for the embodied carbon in AI hardware, making the accounting complex and uncertain.

A London Business School researcher emphasized the importance of transparency. "They should absolutely be disclosing," said Ioannis Ioannou, an associate professor whose research focuses on integrating sustainability into corporate strategy. "We're talking about the potential environmental impact of a scale that we haven't seen before".

How Can Cities Use AI to Address Climate Change?

While major AI companies remain quiet on their own emissions, researchers are exploring how AI itself can help cities and municipalities reduce their climate impact. A comprehensive study called "cityclimAIte" examined 29 AI solutions designed to help cities achieve climate neutrality and adapt to climate change.

The study evaluated AI applications across five key areas where cities can reduce emissions and build climate resilience:

• Mobility and Logistics: AI systems that optimize transportation routes and reduce fuel consumption in urban delivery and public transit networks.
• Electricity and Heating Networks: AI tools that balance energy supply and demand across city grids, reducing waste and improving efficiency.
• Energy-Efficient Buildings: Machine learning systems that optimize heating, cooling, and lighting in residential and commercial structures.
• Urban Climate and Climate Change Adaptation: AI applications that predict heat waves, heavy rainfall, and other climate impacts to help cities prepare and respond.
• Energy-Efficient Data Centers and Waste Heat Recovery: AI systems that reduce the energy consumption of data centers and capture excess heat for reuse.

How to Evaluate AI Climate Solutions for Your City

The cityclimAIte study created a free, web-based AI technology radar to help city leaders assess which climate solutions are ready to deploy. The tool classifies 29 different AI applications based on their technological maturity and market readiness, ranging from early-stage research to fully available products.

• Technology Maturity Assessment: The radar shows which solutions are still in research and development, which are being tested in pilot projects, and which are already available on the market for immediate use.
• Complexity Evaluation: Each application is rated for its technical and organizational complexity, helping decision-makers understand the resources and expertise required to implement it.
• Impact and Feasibility Analysis: The tool evaluates each solution's potential climate impact and practical feasibility, allowing cities to prioritize projects that deliver the greatest benefit with manageable implementation challenges.
• Filtering and Comparison: Users can filter solutions by criteria such as feasibility, impact, and application area, making it easy to compare options and identify the best fit for a city's specific needs and constraints.

The study emphasized that AI's effectiveness in achieving climate goals depends on more than just the technology itself. High-quality data, interoperable systems, necessary expertise, and a clear legal framework, such as that provided by the General Data Protection Regulation (GDPR) and the EU AI Act, are all essential for success.

"The Technology Radar allows you to compare AI solutions for cities. It shows at a glance which solutions are already market-ready, which are currently being tested, and where further research is needed. This allows decision-makers to set priorities transparently and plan pilot projects effectively," said Veronika Hornung-Prähauser, an innovation researcher specializing in digital technologies at Salzburg Research.

Veronika Hornung-Prähauser, Innovation Researcher at Salzburg Research

The cityclimAIte study was conducted from 2024 to 2026 by Salzburg Research, the AIT Austrian Institute of Technology, and Tech Meets Legal, with support from the Austrian Climate and Energy Fund and the Austrian Research Promotion Agency. The research provides a practical roadmap for cities seeking to harness AI's potential while navigating the complex landscape of climate technology options.

As AI companies face mounting pressure to disclose their own environmental impact, the contrast is becoming clearer: the technology that powers modern AI systems consumes vast amounts of energy, yet the companies building these systems remain largely silent about their carbon footprint. Meanwhile, cities and municipalities are beginning to explore how AI itself can become a tool for climate action, provided they have the right guidance and frameworks to evaluate and implement these solutions responsibly.