Conservation Notes: Environmental impact of data centers & AI

The rapidly-growing demand for electric power

By Tom Molloy

With the continued avalanche of bad news about the environment and climate coming from policies of the Trump regime, I write today about something not directly related to the regime: data centers, artificial intelligence and energy usage.

Tens of thousands of job layoffs have just been announced in the media largely due to artificial intelligence, 14,000. For Amazon alone. That is scary enough, but another scary part of AI—and there are many—is the energy required to power it.

The U.S. Energy Information Agency reported that total annual U.S. electricity consumption hit a record high in 2024, and that ceiling could rise if data centers continue expanding at their current pace.

U.S. data centers—which process workloads other than just AI—consumed more than 4% of the country’s total electricity consumption last year, roughly equivalent to the annual electricity demand of Pakistan. By 2030, this figure is projected to grow by 133%.

Pew Research reports that a typical AI-focused hyperscaler* annually consumes as much electricity as 100,000 households. The larger ones currently under construction are expected to use 20 times as much, estimates the International Energy Agency.

Data center cooling systems require a large amount of water, though some types use more than others. In 2023, the country’s data centers directly consumed about 17 billion gallons of water. Hyperscale data centers alone are expected to consume between 16 billion and 33 billion gallons of water annually by 2028.

As of 2024, natural gas supplied over 40% of electricity for U.S. data centers, according to the IEA. Renewables supplied about 24% of electricity while nuclear power supplied around 20% and coal around 15%.

Utilities typically make expensive upgrades to power grids to be able to handle increased energy demands from new data centers. Smaller businesses and U.S. households often pay these costs, unless ratepayer protections are put in place.

One study from Carnegie Mellon University estimates that data centers and cryptocurrency mining could lead to an 8% increase in the average U.S. electricity bill by 2030, potentially exceeding 25% in the highest-demand markets.

Pew research surveys of American attitudes towards AI and its impact on the environment reveal that, absent government regulatory oversight, which is not likely to happen in the next 3 years at least, the trends noted above are not likely to be mitigated. A quarter of U.S. adults think AI will have a negative impact on the environment, a quarter say it will have an equally positive and negative impact, 20% foresee a positive impact, while 30% aren’t sure.

---

*Large-scale data center that provides extensive computing power and storage capacity, to support cloud computing and vast numbers of users.