Every ChatGPT query, every image generation, every AI-powered search consumes real resources. Energy to power servers. Water to cool data centres. Materials to build hardware.
This isn't an argument against AI. It's an argument for using AI thoughtfully.
The Energy Equation
- Training costs: Training a large language model requires energy equivalent to hundreds of households for a year.
- Inference costs: Billions of daily queries add up to substantial ongoing consumption.
- Comparison: A single ChatGPT query uses roughly 10x the energy of a Google search.
The Water Problem
Data centres generate heat. Cooling requires water—lots of it.
Microsoft reported a 34% increase in water consumption in 2022, largely attributed to AI development. Google's water use increased 20%. Many data centres are located in water-stressed regions.
The Other Side: AI for Good
AI also offers environmental benefits:
- Climate modelling and intervention planning
- Energy optimisation in buildings and transport
- Materials science for sustainability
- Conservation monitoring and protection
- Precision agriculture reducing resource use
The question isn't whether AI can help—it's whether the help outweighs the cost.
Our Approach
We believe in responsible AI use:
- Net positive goal: Create AI implementations that generate more value than they consume
- Efficiency focus: Recommend appropriate models for each task
- Transparency: Honest about AI's limitations and costs
- Charitable giving: 10% of profits to effective charities, including environmental and AI safety causes
The most responsible approach: use AI where it creates genuine value, choose efficient implementations, and advocate for industry-wide improvements.