AI agents use roughly 600 times more energy than a simple chat prompt
Climate scientist Zeke Hausfather tracked the energy usage of AI agent Claude Code, finding it consumes 600 times more energy than a typical AI chat. This challenges the low energy figures reported by Google and OpenAI.
What happened
Zeke Hausfather tracked his usage of Claude Code over eight weeks, generating 3.2 billion tokens and using 170 kWh of data center electricity. This works out to roughly 600 times more energy per prompt than a typical AI chat. His findings contradict the low energy estimates reported by Google and OpenAI.
Why it matters
As you consider implementing AI agents in your business, be aware of the significant energy costs involved. This could impact your bottom line, especially if you're already operating on thin margins. It's essential to factor in these costs when evaluating the feasibility of AI agents.
The takeaway
Be cautious of low energy estimates from AI providers and factor in the actual energy costs when deciding whether to implement AI agents in your business.
Our plain-English take, written from public reporting for operational business owners. Always read the original for full context.
Nayre builds the AI systems behind stories like this.
Chatbots, workflow automation, finance intelligence, and internal knowledge systems. Built for operational teams, shipped in days.