-1-1.png?height=120&name=Curation%20Logo%20Transparent%20(1)-1-1.png)
Google has partnered with PJM Interconnection, the largest US grid operator, to deploy AI for streamlining power grid connections. The move marks the first use of AI to manage interconnection queues at scale. Electricity demand is soaring as increasing numbers of data centres are built, spurred by AI development. Current wait times for new grid connections have reached record highs, exacerbating energy supply issues. Tapestry – backed by Alphabet – aims to automate labour-intensive planning tasks, beginning implementation in 2025. PJM serves 67 million people, including northern Virginia – home to the world’s largest data centre hub. (Reuters)
Why does this matter? Electricity consumption in the US is set to reach record levels in 2025 and 2026, according to the US Energy Information Administration (EIA). While this is primarily driven by data centres, homes and businesses also increasingly rely on electricity for heating and transportation. The International Energy Agency has projected that by 2030 the US will use more electricity for data processing than for manufacturing all major energy-intensive goods combined, including cement, steel, aluminium and chemicals.
As AI technologies continue to rapidly grow and evolve, data centres may start experiencing power outages in the near future, according to research by Gartner. Therefore, resolving the complexity of grid connection is crucial to maintaining their operations. Currently, there is a backlog of around 2,300 power projects in grid interconnection queues across the US, predominantly solar and batteries. As of 31 January 2025, PJM’s interconnection queue consisted of 79 GW of solar, 37 GW of storage, 23 GW of wind and 4 GW of natural gas.
The multi-year collaboration with Google will see Tapestry develop AI tools powered by Google Cloud and Google DeepMind. The project aims to reduce bottlenecks to therefore accelerate integration of new, affordable energy projects on PJM’s grid, which spans 13 US states and Washington D.C. The initiative will also unify multiple databases into a single secure platform that enables faster, more efficient grid planning. It also aligns with PJM’s continuing automation efforts and the US interconnection reform implemented in mid-2023.
Clean energy industry advocates applauded the move, with Advanced Energy United’s Director, Jon Gordon, hailing it as an innovative and “critical step” towards modernising the grid to get projects online faster while lowering costs for customers. Nevertheless, Gordon added that policy reforms will be needed to further advance the market and “make the grid more nimble, more open to new resources, and more prepared for the clean energy transition already underway.”
Google’s announcement came shortly after President Donald Trump issued an executive order to bolster the reliability and security of the US power grid to meet rising demand from data centres. The order mandates the Secretary of Energy to fast-track reviews and develop a standard method for assessing reserve margins across regions regulated by the Federal Energy Regulatory Commission. It also includes measures to retain critical generation assets and prevent the retirement or fuel conversion of resources exceeding 50 MW.
Simultaneously, Trump signed a series of executive orders aimed at shoring up the US’ dwindling coal industry in the name of energy security. The measures suspend federal pollution emissions requirements for coal-fired plants for two years and allow ageing facilities to continue operating. They also direct federal agencies to identify coal-rich on federal lands and fast-track mining leases. However, energy experts suggest any resurgence in coal under Trump is likely to be short-lived, as natural gas remains more cost-effective and there is sustained demand for renewables, regardless of the political administration in office.