Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush Wadie WJ Habboush

AI Is Straining the Grid — Can SMRs Offer the Answer?

Artificial intelligence (AI) is no longer just a buzzword — it’s now one of the world’s fastest-growing drivers of electricity demand. From powering large-scale language models to operating autonomous systems, AI is already reshaping the global energy map.

But all that progress comes with a cost:

AI needs power — and a lot of it.

A new report from the International Energy Agency (IEA) highlights just how sharp the curve is. And with the current power grid still heavily reliant on fossil fuels, there’s rising urgency to find cleaner, more scalable sources of electricity.

AI’s Electricity Demand Is Skyrocketing

The IEA projects that global data center electricity consumption could reach 945 terawatt-hours (TWh) by 2030 — more than triple the 2020 level. That figure surpasses the current electricity consumption of the entire country of Japan.

AI-related workloads are set to quadruple, driven by compute-heavy applications in cloud computing, natural language processing, and image generation. These workloads require high-density power delivered consistently — something traditional renewables alone struggle to guarantee at scale.

The near-term power supply for data centers remains largely fossil-based:

Natural gas is expected to supply an additional 175 TWh by 2035.
Coal, though in global decline, is still projected to expand in certain regions.

This raises a serious challenge: if AI's power needs are met with fossil fuels, the climate consequences could undermine its benefits.

Enter Small Modular Reactors (SMRs) — an emerging nuclear technology that offers low-emission, scalable, and always-on electricity, tailored for the exact kind of demand AI workloads require.

Some key advantages:

Lower capital costs (typically under $2 billion)
Smaller physical footprint
Greater compatibility with private investment and modular deployment

The IEA report notes that plans for up to 25 GW of SMR capacity have already been announced — much of it aimed at supporting data centers in the United States. The IEA's findings are clear: the world doesn’t just need more electricity — it needs the right kind of electricity.

If data centers continue to rely on fossil fuels, emissions will spike just as global climate goals grow more urgent. SMRs present a rare opportunity:

Reliable, round-the-clock energy
Zero carbon emissions at the point of generation
Scalable to meet future AI and industrial demands

The world stands at a unique crossroads: AI is accelerating at full speed, and energy systems must catch up — without adding to the climate crisis.

SMRs may be the cleanest, most resilient answer we have.

And as companies like Entra1 and NuScale work to bring these reactors online, a new energy narrative is emerging — one that matches the ambition of AI with the stability and sustainability of advanced nuclear power.

Read more from Wadie Habboush or download the IEA report "Energy and AI"

Wadie Habboush on Medium
Wadie Habboush website
Energy and AI by IEA

AI Is Straining the Grid — Can SMRs Offer the Answer?

Subscribe

This website uses cookies.