Around the world, the rapid development of artificial intelligence is changing governance, economies, and environmental management. Recognising both the promise and the risks of this technology, the UN Environment Assembly (UNEA) has adopted the UN’s first AI and environment resolution, marking a historic step in global environmental governance. The resolution, put forth by Kenya, aims to maximize the benefits of artificial intelligence (AI) while mitigating its environmental impacts. It also requires the UN Environment Programme (UNEP) to prepare a report on the advantages, risks, and environmental effects of AI systems.

However, despite its innovative nature, the resolution has come under fire for its omissions. Most significantly, it makes no mention of tracking AI systems over their entire lifecycle, from energy use and end-of-life disposal to mineral extraction and data center building. Without this comprehensive approach, experts caution, the world risks underestimating AI’s environmental impact and repeating unfair trends from past technological shifts. The debate surrounding the UN’s first AI and environment resolution reveals not only technical gaps but also deeper geopolitical and economic tensions shaping global climate action.

Why Does Ignoring the AI Lifecycle Matter for the Environment?

AI’s effects on the environment don’t stop when a server is turned down or when an algorithm is implemented. To fully comprehend AI’s impact, a lifecycle approach is essential.

Previous drafts of the resolution specifically requested that UNEP investigate AI systems “across their lifecycle” during negotiations in Nairobi. Kenya, Norway, Colombia, and the European Union were among the nations that promoted this language. However, it was removed from the final text because of criticism from Saudi Arabia, Russia, and the United Arab Emirates.

The importance of lifecycle assessment:

• Critical minerals: AI hardware relies on rare earths and minerals, the exploitation of which frequently results in social unrest and ecological harm.
• Water use: Data centers, which are often located in water-stressed areas, consume large amounts of water for cooling.
• Energy demand: AI systems are driving up electricity usage, which, if renewable energy sources are insufficient, could force nations to switch to fossil fuels.
• End-of-life waste: The growing e-waste problem is exacerbated by outdated servers and chips.

Sustainability cannot be attained without monitoring AI “from extraction to disposal,” according to Faith Munyalo, Kenya’s AI focal point. Without lifecycle accountability, the UN’s first AI and environment resolution risks addressing symptoms rather than causes.

Also Read: The Impact Of Artificial Intelligence On Electrical Energy Grids

What Political and Financial Tensions Shaped the Final Resolution?

Beyond technical arguments, geopolitics significantly influenced the outcome. Due to financial differences, the AI resolution was on the verge of collapsing midway through negotiations. Iran and Saudi Arabia contended that rich nations should be the primary source of funding for AI capability, particularly in the Global South. The UK and the EU, on the other hand, insisted that funding should come from a variety of sources.

The final agreement spared wealthier countries from having to support the development of AI capabilities in developing nations directly. Instead, the resolution promotes:

• Improved global collaborations.
• A rise in philanthropic and private sector investment.
• Mobilization of resources voluntarily.

Consequences of this compromise:

• Financing environmentally friendly AI projects may be difficult for developing nations.
• Countries with low financial resources once again bear the burden of capacity-building.
• Global disparities in AI infrastructure could get worse.

For instance, a nearly $7 million demand for an AI-based climate and weather forecasting system is part of Sierra Leone’s updated climate plan. AI is already helping with carbon stock estimations, reforestation, and forest monitoring in Kenya. Such programs remain susceptible in the absence of consistent funding. Critics argue that the UN’s first AI and environment resolution misses an opportunity to support these climate-positive applications directly.

Also Read: Artificial Intelligence’s Environmental Footprint: Balancing Innovation With Sustainability

Is AI Helping or Harming the Clean Energy Transition?

In the energy transition, AI holds a paradoxical position. On the one hand, it provides practical tools for biodiversity protection, climate risk prediction, and energy system optimization. Conversely, it is driving global power demand sharply higher.

AI must be created to prevent repeating the mistakes of previous extractive transitions, according to Somya Joshi of the Stockholm Environment Institute. This necessitates comprehending how the entire AI value chain affects the environment.

The AI energy conundrum:

• In 2024, data centers accounted for roughly 1.5% of global electricity consumption.
• By 2030, this demand is predicted to have more than doubled.
• The International Energy Agency warns that fossil gas and coal will continue to expand substantially this decade, even as renewables and batteries could provide half of the additional electricity.

UN Secretary-General António Guterres has asked Big Tech to run all of its data centers entirely on renewable energy by 2030. However, growing demand for AI could jeopardize climate goals in the absence of legally binding agreements. Critics argue that by omitting energy and water use from its final language, the UN’s first AI and environment resolution weakens its ability to guide a sustainable AI transition.

Also Read: The Impact Of Renewable Energy On Nature: Balancing Progress And Preservation

What Does the Nairobi Outcome Say About Global Environmental Governance?

A heated political environment that hindered progress on several environmental fronts led to the adoption of the AI resolution. Many observers hailed the resolution as a symbolic victory for UNEA, while others saw it as a sign of deeper issues.

The Trump administration’s climate skepticism was reflected in the United States’ rejection of the results, which it dismissed as “climate change theatre.” Oil-rich countries, on the other hand, attempted to soften words related to climate science, such as glacier melting. Small island nations like Barbados and Fiji, as well as the EU and Australia, strongly condemned this.

More general UNEA results included:

• A feeble resolution on transition minerals and mining, delaying the implementation of more environmentally friendly supply chains.
• There is relief that discussions for a global plastics convention have not been derailed and will soon resume.

In this context, the UN’s first AI and environment resolution represents progress, but also highlights how geopolitical divides continue to constrain ambition in multilateral environmental forums.

Also Read: How To Choose The Right Life Cycle Assessment Methodologies And Tools For Your Product

Looking Ahead: Can Future Negotiations Close the Gaps?

Faith Munyalo and other negotiators view the resolution as a beginning rather than a finish. They contend that the “blind spots” left behind, including lifetime accountability and equitable finance, must be addressed at the subsequent UNEA meetings.

A more effective strategy would:

• Require AI system lifecycle assessments.
• Establish standards for data centers powered by renewable energy.
• Encourage the development of capabilities in the Global South with steady funding.
• Include AI governance with more comprehensive frameworks for biodiversity and climate change.

The UN’s first AI and environment resolution may struggle to achieve its goal of coordinating technological advancement with environmental sustainability without these actions.

Also Read: Implementing Lean Manufacturing To Minimize Waste And Maximize Efficiency

Frequently Asked Questions (FAQs)

Q1. Why is lifecycle assessment crucial for the environment and artificial intelligence?

AI affects the environment at every stage, from energy use and mineral extraction to e-waste production. Its actual footprint is underestimated if these stages are ignored.

Q2. Does the resolution prohibit or limit the advancement of AI?

No. Instead of restricting innovation or deployment, the resolution seeks to promote environmentally conscious AI.

Q3. Will the UN revisit AI lifecycle issues in the future?

Yes. Future discussions should include stronger wording on lifetime accountability and financial matters, according to several nations and experts.

Also Read: How Is Green Manufacturing Revolutionizing Business Sustainability?