As more than 200 laws develop globally to regulate artificial intelligence (AI), questions arise about the environmental implications of this rapidly expanding sector. With over 100 countries implementing legislation, the focus has largely centered on privacy, bias, disinformation, and cybersecurity, while often neglecting the environmental consequences of AI technology.
The AI industry is known for its significant energy consumption, leading to increased greenhouse gas emissions and electronic waste. This impact is partly attributed to the manufacture and use of complex computer chips, particularly graphics processing units (GPUs), which are essential for training AI models. The environmental ramifications extend across the entire AI lifecycle, from hardware production and model training to deployment and eventual disposal.
Research indicates that the manufacture of AI components, which rely heavily on the extraction of rare earth elements, can contaminate soil and water, pollute the air, and lead to habitat loss. Notably, a study from 2025 estimated that training the GPT-3 model—released by OpenAI in 2020—consumed approximately 700,000 liters of freshwater for electricity generation and cooling data centers. Despite advancements in energy efficiency for AI models, the overall energy consumption and associated emissions continue to rise, especially as the size and application of AI technologies expand.
The European Union’s AI Act, which took effect on August 1, 2024, is the world’s first comprehensive regulation governing AI. While it acknowledges some environmental consequences, it primarily emphasizes human-centric technology. It mandates that AI systems be developed in a sustainable manner and requires disclosure of energy consumption data upon request by the AI Office within the European Commission. However, this approach has drawn criticism for its lack of stringent requirements and its anthropocentric focus.
In contrast, the UK currently lacks specific AI legislation, with existing laws governing the sector. The UK government’s 2023 white paper on AI regulation does not prioritize sustainability, noting that while AI can aid in addressing climate change, it does not explicitly address environmental risks. The document states, “The proposed regulatory framework does not seek to address all of the wider societal and global challenges that may relate to the development or use of AI,” including sustainability concerns.
Calls for increased transparency in the AI sector have intensified, urging developers to disclose data on energy and water consumption, carbon emissions, and the use of rare earth elements and plastics in production. This transparency would create a baseline for setting targets and limits for energy efficiency and emissions, ultimately improving sustainability within the industry.
Several proposals have emerged to mitigate environmental effects, such as training AI models on less carbon-intensive energy grids and utilizing water-efficient data centers. Additionally, warnings regarding the environmental impact could accompany AI-generated outputs, informing users of the carbon emissions or water usage linked to specific queries. An AI labeling system, akin to the EU’s energy efficiency labels, could help consumers assess the sustainability of different AI systems.
Proposals for an “energy star” rating system for AI, along with social and environmental certification, aim to empower consumers in making informed choices about their AI usage. Incentives such as tax breaks for tech firms adopting sustainable practices could also bolster these efforts. By embedding sustainability into AI legislation, stakeholders can safeguard the environment while allowing for the continued growth of AI technologies.
The implications of AI’s environmental impact are significant as the industry continues to evolve. By addressing these concerns head-on, regulatory authorities can create a framework that balances technological innovation with ecological responsibility, ensuring that the benefits of AI do not come at the expense of the planet.
See also
AI Technology Enhances Road Safety in U.S. Cities
China Enforces New Rules Mandating Labeling of AI-Generated Content Starting Next Year
AI-Generated Video of Indian Army Official Criticizing Modi’s Policies Debunked as Fake
JobSphere Launches AI Career Assistant, Reducing Costs by 89% with Multilingual Support
Australia Mandates AI Training for 185,000 Public Servants to Enhance Service Delivery
