Green AI
The practice of reducing the environmental impact of AI through energy-efficient algorithms and sustainable computing practices.
Definition
An emerging movement that prioritizes AI research and operations with low carbon footprints—optimizing model architectures for efficiency (distillation, pruning), using low-power hardware, and scheduling large training jobs when renewable energy is available. Governance initiatives include carbon accounting for AI workloads, efficiency KPIs (FLOPs per accuracy gain), and incentives for teams that meet sustainability targets without compromising quality.
Real-World Example
A social-media company schedules its large-scale model pretraining on weekends when its data centers run exclusively on wind power, tracks the kWh consumed per training run, and publishes annual “AI carbon footprint” reports, driving research into more efficient architectures.
.svg)
