UN Chief Calls on AI Firms to Disclose Emissions and Go Renewable by 2030

UN Secretary-General António Guterres called on every major AI company to disclose the full environmental footprint of their systems and power them entirely with renewable energy by 2030, speaking at the opening of the World Artificial Intelligence Conference in Shanghai. The address, delivered alongside Chinese President Xi Jinping, framed AI's environmental impact as one of three priorities, alongside capacity-building and safety, that Guterres argued must be addressed for the technology to advance sustainable development rather than deepen global inequality.

 

Why AI's Environmental Footprint Has Become a Governance Issue

 

Guterres acknowledged that AI carries genuine potential to accelerate the energy transition, citing its ability to manage electricity grids, integrate renewable generation and cut waste. But he argued that potential is undermined if the technology's own resource consumption, its growing demand for energy, water and land to run data centres and train models, remains hidden from public scrutiny. His call for disclosure treats AI's environmental footprint as a transparency problem first: without visibility into how much energy and water AI systems actually consume, neither governments nor the public can assess whether the technology's net effect on climate goals is positive or negative.

The 2030 renewables deadline attached to that disclosure call sets a concrete, time-bound target rather than an open-ended aspiration, mirroring the structure of corporate net-zero commitments that have become common across other industries. Guterres pointed to China as already requiring data centres to increase their use of renewable power year over year, presenting it as a model other governments could adopt by incorporating clean power requirements for AI directly into national energy planning.

 

Read more: All 27 EU States Miss Deadline to Transpose Building Energy Rules

 

The Inequality Framing Behind the Speech

 

The environmental disclosure call sat within a broader argument about AI deepening global inequality. Guterres noted that one-third of humanity remains offline, computing power and technical expertise remain concentrated in a small number of countries and companies, and hundreds of billions of dollars in private AI investment flow overwhelmingly to wealthy nations while developing countries receive comparatively little. That imbalance, he argued, risks widening gaps in income, opportunity and security between the Global North and South rather than the technology serving as a shared tool for development.

To address that gap, Guterres pointed to two initiatives already in motion: a UN-supported Global Network for Exchange and Cooperation on AI Capacity Building, which he said has already drawn nominations from more than 20 countries including China, and a forthcoming proposal for a Global Fund for AI aimed at the same goal of supporting developing countries. Both are framed as mechanisms to extend AI capability, including affordable computing power and training, beyond the small group of countries and firms currently at the technology's frontier.

 

Explore OneStop ESG Marketplace: AI (Artificial Intelligence)

 

A Speech, Not Yet a Commitment

 

It is worth being clear about what this event represents. Guterres's remarks are a call to action rather than a binding commitment; no AI company has yet agreed to the disclosure and renewables targets he proposed, and the Global Fund for AI remains a forthcoming recommendation rather than an established mechanism. The speech's significance lies in placing environmental transparency and renewable power squarely on the agenda of a major international AI governance conference, alongside safety and capacity-building, rather than in any concrete regulatory or corporate action taken as a result.

Whether major AI companies respond to the disclosure and renewables call with actual commitments, and whether governments follow China's example of mandating rising renewable power shares for data centres, will determine whether this speech marks a genuine turning point in how AI's environmental footprint is governed or remains an aspirational statement without material follow-through.