The energy needed to power the AI revolution

(BPT) - On the night of March 2, 2026, millions of workers around the world opened their laptops and found Claude — one of the world's most widely used AI assistants — was down. The outage lasted hours. It was a small but pointed reminder of how dependent modern work has become on AI, and how fragile the infrastructure behind it can be. The grid that powers that infrastructure is under more strain than most people realize — and the vulnerabilities don't stop at the data center door. They come home with us.

Today, the world is spending hundreds of billions of dollars building the energy infrastructure needed to power AI, including data centers, grid upgrades and new power plants. That scale is just beginning. Fast forward to 2030, when the International Energy Agency (IEA) projects global data center electricity consumption will nearly double.