Building Artificial Intelligence That Serves People and the Planet

Artificial intelligence is increasingly framed as both a climate threat and a climate solution. Critics highlight the rapid growth in electricity demand from data centres, rising water use for cooling, and growing pressure on supply chains for critical minerals. At the same time, proponents argue that AI can accelerate decarbonisation by improving energy efficiency, supporting renewable integration, and strengthening environmental monitoring.

This tension has become a central issue for policymakers, researchers, and companies, as highlighted in recent analysis by Reuters. The debate suggests that AI’s environmental impact will depend less on the technology itself and more on how it is governed and deployed.

Energy Use and Data Centre Growth

One of the most immediate concerns is energy consumption. Training and operating advanced AI models requires significant computing power, much of it concentrated in large data centres. These facilities already represent a fast-growing share of global electricity demand, with further increases expected as AI applications spread across sectors such as finance, industry, transport, and public administration.

If powered by fossil-heavy grids, this expansion risks locking in higher emissions. However, experts note that AI can also improve the efficiency of the very infrastructure it relies on. Advanced algorithms are being used to optimise cooling systems, manage workloads, and reduce downtime in data centres, lowering overall energy intensity.

When combined with renewable power sourcing, energy storage, and grid flexibility, AI-driven optimisation can help limit emissions growth even as digital demand rises.

Supporting the Energy Transition

Beyond digital infrastructure, AI is increasingly being used to support the clean energy transition. In power systems, machine learning improves forecasting for wind and solar generation, allowing grid operators to balance supply and demand more effectively. This reduces reliance on fossil-fuel-based backup generation and supports higher shares of renewables.

In transport and industry, AI tools are enabling route optimisation, predictive maintenance, and process control that cut fuel consumption and material waste. These applications offer practical emissions reductions, particularly in sectors where full electrification or fuel switching remains challenging.

Environmental Monitoring and Land Use

AI is also playing a growing role in environmental monitoring. Satellite imagery combined with machine learning is being used to detect deforestation, track biodiversity loss, and monitor land-use change in near real time. In agriculture, AI-supported systems can optimise irrigation, fertiliser application, and crop management, reducing water use and emissions while improving yields.

However, experts caution that these benefits are not guaranteed. Without clear objectives and safeguards, AI efficiency gains can lower costs and unintentionally drive higher overall consumption, offsetting environmental benefits.

Governance, Metrics, and Transparency

A recurring theme in the debate is governance. Researchers stress the need for clearer standards to measure and disclose the environmental footprint of AI systems. This includes not only electricity use during operation, but also emissions embedded in hardware manufacturing, supply chains, and end-of-life disposal.

Consistent metrics would allow regulators, investors, and companies to compare AI deployments more effectively and align them with net-zero targets. Transparency is also seen as essential to avoid greenwashing and ensure that AI-driven sustainability claims are credible.

Equity and Global Access

Equity considerations are another critical factor. While AI has the potential to support climate adaptation and resilience, many low- and middle-income countries lack the infrastructure, data access, and skills required to benefit fully. Without targeted investment and international cooperation, AI risks reinforcing existing inequalities in climate vulnerability and economic development.

Experts argue that capacity-building, open data initiatives, and inclusive policy design will be necessary to ensure AI supports global sustainability goals rather than concentrating benefits in a few regions.

Aligning AI With Climate Goals

For businesses, AI deployment is increasingly linked to sustainability strategies and reporting requirements. Companies are under growing pressure to disclose emissions associated with digital operations and demonstrate that AI-driven growth aligns with long-term climate commitments.

The emerging consensus is that moving beyond both hype and alarmism requires practical governance, credible measurement, and alignment with climate and development objectives. AI can support people and the planet, but only if environmental considerations are embedded into its design, deployment, and regulation from the outset.

Source: www.reuters.com