TBNEnergy
The US power market’s forecasting problem comes at a cost to ratepayers
Publication: Latitude Media
Author: Rebecca Egan McCarthy
Source published: 03/11/2026, 19:58:01 UTC
Briefing created: 03/11/2026, 20:59:24 UTC
Based on a feed excerpt; full article content has not been acquired. Source evidence retrieved 03/11/2026, 20:38:25 UTC.
Briefings
Aurora Energy Transition Forum in Manhattan (Nov 3, 2026) highlighted the US power market's forecasting challenges driven by AI data center growth. Speakers noted that while global electrification is 87% AI-driven, US data centers account for half of expected growth, creating significant uncertainty. This uncertainty could cost ratepayers between $79 billion and $624 billion in capital expenditure, with PJM facing additional transmission costs. Key uncertainty remains regarding the success of the new foreign equipment ban policy and the long-term viability of behind-the-meter generation as a workaround.
Key points
- Data center load forecasts vary widely, with the highest expecting 94 to 131 gigawatts of load growth by 2030, while conservative estimates hover around 22 GW.
- If Aurora's projections are correct, overbuilding could saddle ratepayers with steep, unnecessary costs, while underbuilding could strain the grid and threaten reliability.
Context
The US power market faces unique challenges due to AI-driven data center growth, which complicates industry ability to predict future changes. This uncertainty comes with a cost for ratepayers, as grid operators expect significantly more demand by 2030 than current forecasts suggest.
Limitations
Evidence is limited to a single day-long conference session in November 2026. Specifics on the impact of the new foreign equipment ban policy are uncertain, as rules are expected to be released in late December. Historical precedent suggests only 15% of planned projects will materialize, but the exact impact of the ban on project timelines is not fully quantified.
Generated by Bittrees-hosted qwen35-2b sha256:aaf42c8b7c3c on 03/11/2026, 20:59:24 UTC. Based on publisher-provided feed evidence; may not cover the full original article. Check the source for details.
View archived briefing on IPFS
bafybeifczdosqgsyjeyrvtd6twmqno2mwck45xahkyu5nj4xy5kjaekniu