Risks to the environment
General- purpose AI is a moderate but rapidly growing contributor to global environmental impacts through energy use and greenhouse gas (GHG) emissions. Current estimates indicate that data centres and data transmission account for an estimated 1% of global energy- related GHG emissions, with AI consuming 10–28% of data centre energy capacity. AI energy demand is expected to grow substantially by 2026, with some estimates projecting a doubling or more, driven primarily by general-purpose AI systems such as language models.
ENTITY
2 - AI
INTENT
2 - Unintentional
TIMING
3 - Other
Risk ID
mit1030
Domain lineage
6. Socioeconomic and Environmental
6.6 > Environmental harm
Mitigation strategy
1. Prioritize the deployment of AI data centers in locations with low water stress and high access to clean, renewable energy sources (e.g., wind or hydropower) to minimize the associated carbon footprint and reduce strain on local water resources, with siting being the most critical factor. 2. Mandate the adoption of energy-efficient model architectures and training techniques, such as model pruning, knowledge distillation to smaller models, and low-precision inference (e.g., FP16 or INT8 quantization), to fundamentally reduce the computational and energy demand per AI inference. 3. Implement intelligent, AI-driven operational efficiency measures within data centers, including optimizing cooling systems and dynamically adjusting resource allocation (e.g., carbon-aware scheduling and workload management) to ensure energy is not wasted on idle capacity or suboptimal environmental controls.