The head of Hitachi Energy, the world’s largest transformer manufacturer, has warned that governments must urgently regulate the electricity consumption of artificial intelligence data centres to prevent grid instability and blackouts. In an interview with the Financial Times, CEO Andreas Schierenbeck said AI’s power usage is unlike any other industrial demand and poses unique challenges to energy systems worldwide.
“AI data centres are very, very different from these office data centres because they really spike up,” Schierenbeck explained. “If you start your AI algorithm to learn and give them data to digest, they’re peaking in seconds and going up to 10 times what they have normally used.”
He emphasized that no other industry would be permitted to draw power in such a volatile manner. “If you want to start a smelter, you have to call the utility ahead,” he said, calling for similar regulatory oversight of data centres to smooth out sudden and extreme demand surges.
Schierenbeck’s warning adds a new dimension to the growing debate around AI’s environmental and infrastructure impact. While much attention has focused on the overall electricity consumption of AI-related data centres, the Hitachi Energy chief highlighted the dangerous unpredictability of these power demands — especially when combined with the intermittent nature of renewable energy sources.
“Volatility on top of volatility” is how he described the combination of AI power spikes and uneven clean energy generation, warning that this could undermine grid reliability.
The International Energy Agency projects that data centre power consumption could nearly double to 945 terawatt-hours by 2030 — more than the total electricity used by Japan today. Some European countries, including Ireland and the Netherlands, have already imposed restrictions on new data centre developments over concerns about strain on national grids.
According to Financial Times reporting, analysts at Rystad Energy argue that AI’s power needs could actually help stabilize electricity networks if properly managed. This would require tech companies to cap energy use and schedule AI training during periods of abundant renewable supply — something not currently mandated in most jurisdictions.
Schierenbeck, who previously led German energy utility Uniper, is one of the first major industry executives to publicly urge regulatory intervention on this front. His company, Hitachi Energy — born out of ABB Power Grids’ $11 billion acquisition in 2020 — is now at the epicenter of a global power transformer shortage, driven by surging demand for grid upgrades.
The CEO revealed that Hitachi Energy’s order backlog has tripled over the past three years to $43 billion, and that clearing it could take up to three years due to bottlenecks in expanding manufacturing capacity. A critical constraint is the availability of specialist contractors to construct reinforced factory floors required to produce heavy transformers weighing hundreds of tonnes.
To meet skyrocketing demand, the Japanese-owned group plans to invest $6 billion and hire 15,000 new workers by 2027. Schierenbeck expressed confidence in filling the roles, noting that laid-off engineers from Europe’s automotive and chemical industries are increasingly available for retraining.
As governments worldwide grapple with how to manage AI’s rapid rise, Schierenbeck’s call to action signals a growing consensus among infrastructure experts: the sector’s energy demands can no longer be left unregulated if power systems are to remain stable.
