UK Researchers Propose Foundational Ways for Environmentally Responsible AI

Technological development has transformed our lives in many shapes and forms. Artificial intelligence (AI) is no exception. However, opportunities and doubts simultaneously grow over the development and usage of AI, straying between the enhancement of life and negative social and environmental impacts. Then, is it possible to develop a responsible AI system?

Opportunities and Challenges

From industrial machinery to smartphones and the internet, innovations have always been integral to development by providing opportunities for more productive work and better lives. Some examples include the use of satellite imagery for methane mapping, smartphones for early wildfire detection, and e-books for widening information access.

However, these innovations are not without trade-offs. The proliferation of AI, in particular, has generated debate over its complex relationship with the environment. Besides their potential benefits, AI technologies also consume extensive resources, such as critical minerals and rare earth elements, that can potentially be harmful if extracted too much.

According to the United Nations Environment Programme, around 800 kilograms of raw materials are needed to build a two kilogram computer. Moreover, data centers built to store information from AI technologies consume extensive amounts of water and energy, often from fossil fuels. There is also a concern about electronic waste produced by data centers, which often contain hazardous materials.

Responsible AI

In light of the climate crisis, biodiversity loss, extreme weather events, and other challenges we face today, leveraging the benefits of AI means we must find a way to implement a responsible AI use and system as soon as possible.

In February 2025, the Royal Academy of Engineering, an engineering academy based in the UK, published a report titled “Foundations for environmentally sustainable AI,” which highlighted foundational ways to minimize risks from AI usage. The report was developed in partnership with the Institution of Engineering and Technology and BCS, the Chartered Institute of IT, under the National Engineering Policy Centre (NEPC).

Recognizing the potential benefits and concerns of AI technologies, the report proposes foundational steps to create progress for responsible AI:

Expanding environmental reporting mandates, including introducing mandatory reporting systems for the energy, carbon, and water consumed during the development and use of AI services.
Addressing information asymmetries across the value chain to ensure a clear understanding of the environmental impacts of AI systems and services between stakeholders. This is to encourage a more tailored approach towards responsible AI by individuals, businesses, and organizations.
Setting environmental sustainability requirements for data centers, including water usage, waste recycling, and carbon emissions, to support reducing resource demands from data centers.
Reconsidering data collection, transmission, storage, and management practices through creating incentives and good practices to reduce harmful impacts and encourage responsible AI procurement.
Leading the way with government investment by embedding environmental sustainability in AI policy and programs.

These steps are formulated with practical actions the UK government can implement in the near term. However, they are also intended to be internationally applicable.

Technology and Innovation for Sustainable Development

Technology will continue to evolve, and innovations will continue to emerge. It is the responsibility of governments, businesses, and civil society to ensure that the development and use of these tools do not cause further harm to people and the planet. Beyond environmental sustainability, it is also crucial to expand access for the people and eliminate possible bias in these tools so that they can be accessed by everyone and do not create further disparity.

Kresentia Madina

Reporter at Green Network Asia | Website | + posts

Madina is the Assistant Manager of Stakeholder Engagement at Green Network Asia. She holds a bachelor’s degree in English Studies from Universitas Indonesia. As part of the GNA In-House Team, she supports the organization's multi-stakeholder engagement across international organizations, governments, businesses, civil society, and grassroots communities through digital publications, events, capacity building, and research.