When AI firms become power firms, should governments foot the bill?

It is rare for a policy announcement in Washington to feel like the real-time validation of a thought experiment I undertook. Yet that is precisely what happened last week when the Trump administration unveiled the “Ratepayer Protection Pledge”, urging America’s largest artificial intelligence firms to build, or procure, their own electricity supply for the data centres that power AI.

The non-binding pledge asks hyperscalers like Alphabet, Microsoft, Amazon, Meta, Oracle, and xAI to pay the full cost of energy generation and grid upgrades required to run their facilities, rather than passing those costs on to ordinary consumers.

For anyone who has been modelling the economics of AI infrastructure, this proposal is striking. Interestingly, I ran a sort of Monte Carlo-style exercise examining Alphabet’s “century bonds” in one of my earlier columns. One of the plausible long-term scenarios from that exercise was hyperscalers slowly evolving into quasi-utility companies whose competitive advantage lies not only in algorithms and data but also in electricity generation, transmission, and physical compute capacity. The Trump administration’s pledge seems to nudge the industry in that direction.

The reason is simple: AI is an energy revolution disguised as a technological revolution.

Training and running large AI models requires enormous computing clusters with thousands of GPUs operating continuously inside data centres that consume electricity at industrial scale. In the United States alone, data centres currently account for roughly 4–5% of national electricity demand, and projections suggest this could rise to 9–17% by 2030 as AI infrastructure expands.

That growth has triggered political backlash as communities hosting large data centres have complained about rising power bills and strained grids. This pledge is therefore as much a political gesture as an economic one. And, by the way, it is not binding and its operational details are sketchy at best. So let us not even get into the enforcement mechanism. Still, the pledge signals something profound about the next phase of the AI race.

The constraint will no longer be about chips. It will be about energy.

If the pledge becomes reality, the hyperscalers may resemble a hybrid between a technology company and an independent power producer. These big tech firms may end up building and controlling the energy supply needed for computing.

If AI infrastructure evolves into a capital-intensive energy business with massive fixed assets, regulated pricing structures and predictable demand, then issuing ‘century bonds’ —as in the case of Google’s long-term debt raise —begins to look less eccentric and more rational.

The power generation industry has long been capable of sustaining long-term liabilities, and it is now on the cusp of being disrupted by hybrid technology companies building massive data centres across the globe.

While the heads of these AI companies have signed a non-binding pledge to build and run their own electricity supply in the United States, they are being aggressively courted by the Indian government with a tax holiday.

In the Union Budget earlier this year, the government announced a tax holiday until 2047 for foreign cloud providers that deliver global services through data centres located in India. The government’s pitch fails to fully account for the downstream effects of such incentives.

In effect, India risks subsidising capital-intensive infrastructure whose largest resource costs — power, water and land — will ultimately be borne by local ecosystems and public utilities.

Unlike the software companies India courted in the past, hyperscale data centres are heavy industrial facilities that consume gigawatts of electricity, enormous volumes of water for cooling, and large tracts of land. The economic benefits they generate locally — in terms of employment — are relatively modest once construction is complete.

Offering a multi-decade tax holiday to attract such facilities could place additional strain on already stressed urban ecosystems. There is also the question of grid financing as hyperscale data centres require dedicated transmission lines, substations and network upgrades — costs that often fall on public utilities unless explicitly recovered from operators. These concerns are beginning to surface at the state level.

Tamil Nadu, for instance, is letting its data-centre policy lapse. In 2021, the state government incentivised data-centre investment with a range of perks. Now officials are openly acknowledging the trade-offs involved, particularly the large electricity demand and water consumption associated with AI data centres in Chennai.

This divergence between states such as Tamil Nadu and the Union government highlights a deeper policy gap. The physical consequences of hosting these facilities — including power demand, water stress, land use and grid upgrades — are borne primarily by state governments. And those pressures will only grow as AI workloads expand.

While Washington’s approach — at least rhetorically — is that hyperscalers should eventually finance their own power supply, India’s current approach appears to be the opposite: offering generous incentives without fully considering the potential toll on the country’s power infrastructure and water resources. If AI is indeed becoming the next layer of global industrial infrastructure, the question India must confront is not merely how to attract data centres, but who will ultimately pay for the energy systems that sustain them.

Published – March 14, 2026 08:12 am IST