Site iconSite icon ForkLog

PJM Reports $23.1 Billion Cost for AI Data Centers

PJM Reports $23.1 Billion Cost for AI Data Centers

The demand for electricity from artificial intelligence data centers has resulted in additional expenses amounting to $23.1 billion within the jurisdiction of PJM Interconnection, the largest grid operator in the US. This is according to a report by Monitoring Analytics.

The document indicates that the load from computing centers is already reshaping the capacity market. Over three auction periods (from 2025/2026 to 2027/2028), the projected increase in consumption has generated billions of dollars in additional revenue for generators, ultimately impacting consumers.

Experts highlighted that in just the last two auctions, end-user bills increased by $13.8 billion. This occurred despite existing price caps established by an agreement between the state of Pennsylvania and PJM.

According to the report for the first quarter of 2026, the average wholesale cost of electricity in the region surged by 75.5% year-on-year to $136.53 per MWh. The main impact was on capacity payments (generation reservation), which soared nearly 400%.

Monitoring Analytics analysts described the current trend as “unique and unprecedented.” They noted that the capacity market was initially designed for gradual demand growth, whereas AI projects require the connection of massive amounts of energy in a short timeframe.

As a solution, Monitoring Analytics proposed tightening rules for data center operators:

Specialists warned that the issue is systemic. While Bitcoin miners have traditionally sought cheap excess energy and were prepared for disconnections, AI centers demand high reliability. This raises the question for regulators: who should pay for infrastructure development—tech giants or ordinary households?

In May, TeraWulf acquired a site in Eastern Kentucky to develop HPC infrastructure with a potential capacity of over 1 GW.

Exit mobile version